Abstract
Living organisms are far superior to state-of-the-art robots as they have evolved a wide number of capabilities that far encompass our most advanced technologies. The merging of biological and artificial world, both physically and cognitively, represents a new trend in robotics that provides promising prospects to revolutionize the paradigms of conventional bio-inspired design as well as biological research. In this review, a comprehensive definition of animal–robot interactive technologies is given. They can be at animal level, by augmenting physical or mental capabilities through an integrated technology, or at group level, in which real animals interact with robotic conspecifics. Furthermore, an overview of the current state of the art and the recent trends in this novel context is provided. Bio-hybrid organisms represent a promising research area allowing us to understand how a biological apparatus (e.g. muscular and/or neural) works, thanks to the interaction with the integrated technologies. Furthermore, by using artificial agents, it is possible to shed light on social behaviours characterizing mixed societies. The robots can be used to manipulate groups of living organisms to understand self-organization and the evolution of cooperative behaviour and communication.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
1 Introduction
Animal–robot interactive technologies represent a relatively novel research field of bio-robotics and are opening up to new opportunities for multidisciplinary studies, including biological investigations, as well as bio-inspired engineering design. This new field introduces the possibility to have, beyond the traditional bioinspiration, the merging of natural and artificial worlds in synergistic systems (Webb 2000; Krause et al. 2011; Garnier 2011; Halloy et al. 2013).
In this review, the hybrid context connecting the natural and artificial components that takes place at individual level has been defined as “bio-hybrid organism”. In bio-hybrid organisms, also called cyborgs, bio-robots or animal–robots (Sato et al. 2008; Peng et al. 2011; Wang et al. 2015), an artificial component or a biological organ is incorporated into an animal or into a robot, respectively, with a direct interaction between the nervous or muscular system and the electronics. Nowadays, bio-hybrid organisms are a hot spot in the research community since they merge animal capacities to adapt to dynamic environments with the possibility to control them as robots (Krause et al. 2011). Indeed, animals have physical and cognitive features that are often source of inspiration for hardware design and control methods (Cham et al. 2002; Pfeifer and Bongard 2006; Long et al. 2006; Kim et al. 2008; Bonsignori et al. 2013; Lin et al. 2011; Zahadat et al. 2015; Bodi et al. 2015; Schmickl et al. 2011). In addition, the integration of biological entities with artificial devices can also involve biological information processing units and computing elements. This research domain named “cyborg intelligence” produces bio-hybrid cognitive functions such as perception and learning (Wu et al. 2016; Brown and Brown 2017). Despite great efforts in developing novel biologically inspired artefacts, animals are still outperforming the current bio-inspired robots by orders of magnitude. Therefore, the possibility to directly control their actions could give interesting advantages (Yang et al. 2015). Their potential applications are numerous, involving high multidisciplinary efforts. For instance, they could be used as reliable micro-vehicles for security and rescue tasks. Besides, they represent valid tools for advancement in neuroscience research and bio-electronic interfaces (Wang et al. 2013; Katz 2014).
When living animals and artificial agents interact and modulate each other, following the principles of social behaviour, an animal–robot mixed society is established. These mixed societies are dynamic systems (Halloy et al. 2013), where artificial agents are no longer simple dummies triggering specific reactions in animals. In mixed societies, robots can evoke behavioural responses adjusting the behaviour according to the animal’s one (Halloy et al. 2013). Behaviour traits play an important role in biological adaptations and conservation. Robots can be useful to study behavioural adaptation, since they are easier to handle if compared to real animals, and it is possible to control their position in the environment, allowing a highly standardized and reproducible experimental design. This novel paradigm for behavioural ecology investigation, merging robotics with ethology, is also known as ethorobotics (Partan et al. 2011; Kopman et al. 2013; Romano et al. 2017b). Experiments for social behaviour have been revolutionized by introducing physical models providing important methodological advantages. In an animal–robot mixed society, a biomimetic animal replica is located in a definite place and time simultaneously with real animals, and it may be accepted as a conspecific by animals. In order to establish a mixed society, the robot, accepted as a conspecific, can interact and modulate the animal behaviour and/or can adjust its behaviour according to the animal response. However, since this domain involves both roboticists and zoologists, it should be considered that from a biological point of view, the term “population” instead of “society” is more appropriate, since “society” in zoology has a precise and rigorous definition, often confused with gregariousness and other terms, referring to temporary animal aggregations of any nature (Sherman et al. 1995).
The state of the art of the current research on bio-hybrid organisms, as well as animal–robot mixed societies, was surveyed by following a comprehensive and innovative approach. Indeed, one of the novelties of this review, if compared with the previous ones (Garnier 2011; Krause et al. 2011; Mondada et al. 2013; Wang et al. 2013; Katz 2014; Mitri et al. 2013; Frohnwieser et al. 2016; Stojnić 2017), is that here the interactions between living organisms and artificial systems were investigated from individuals to populations, by summarizing in a more detailed and comprehensive way the different animal species used as biological models as well as the biological interfaces involved in each research. Pioneer works on animal–robot interactions (Vaughan et al. 2000; Halloy et al. 2007) provided different interactive concepts mainly based on the physical localization and mobility of the artificial devices with respect to living organisms (Garnier 2011; Krause et al. 2011; Mondada et al. 2013). In particular, Mondada et al. (2013) named “mobile nodes” the autonomous mobile robots interacting and moving with animals, “static nodes” the immobile artificial agents including distributed sensors and actuators that are fixed in the environment and “mounted nodes” consisting in sensors and actuators fixed on living organisms.
In this review, the nature of the interaction has been described not only at level of spatial localization of the artificial device with respect to the animal (e.g. embedded in the organism or mixed in the animal population), but it branches at level of the physiological/behavioural inputs and outcomes produced during the interaction. In particular, how the artificial components of bio-hybrid organisms affect the physiology of the animal to manage their motor outputs, sensory systems, as well as metabolic processes is reported. Furthermore, which behavioural stimuli, communication channels and/or blend of cues are produced by artificial agents to establish different ecological interactions with animals in mixed societies is also documented.
In addition, compared to the previous review papers, where these domains were still emerging, this paper shows many results that have been obtained very recently and how they are applied in various research fields, providing an accurate analysis of the different species selected as model organisms.
Although this review focuses on the animal kingdom, it is interesting to mention that the use of artificial devices is not only restricted to animals, but is also currently deployed in other types of organisms such as plants (Hamann et al. 2015). Furthermore, interactions between human beings and artificial entities represent an increasingly common context in human society (Ruhland et al. 2015), with promising application as therapy or assistive functions (Pennisi et al. 2016; Breazeal et al. 2016).
Figure 1 shows the scheme of the classification of the animal–robot interactions, based on the aim outlined above.
2 Methods
Scopus and Web of Science databases (accessed June 2018) have been used to survey the literature. As keywords in the bio-hybrid organism domain, we used “bio-hybrid organism”, “bio-hybrid animal”, “bio-hybrid system”, “cyborg”, “animal cyborg” “bio-robot”, “animal–robot”. Concerning the animal–robot mixed society domain, the used keywords were: “animal–robot mixed society”, “mixed society”, “animal–robot interaction”, “animal–robot bio-hybrid society”, “robotic fish”, “robotic animal”, “ethorobotics”. We double-checked the literature by using the same keywords on these domains also in another database (Google Scholar). In addition, journals in which the articles reviewed in this work were published were defined as biology oriented, engineering oriented or multidisciplinary oriented by reading the aim and scope of the different journals on their Web platforms, as well as by searching the journals’ subject areas on Scimago (scimagojr.com).
3 Research progress in bio-hybrid organisms
Bio-hybrid organisms, endowed with animal visual, audio and tactile sensory skills, represent a new category of robots showing outstanding abilities to navigate in the environment and carry out complex tasks under extreme conditions. Recently, a growing number of studies focused on bio-hybrid organisms, aiming to exploit the great potential of this research area (Fig. 2). The main tested species and research topics are detailed in Fig. 3a, b. In the following paragraphs, we examined several attempts to design controllable systems, according to different bio-artificial interactions and integrations.
3.1 Movement control
3.1.1 Controlling animal terrestrial locomotion
Generating stable and robust legged locomotion devices is still a challenging topic in the current robotics. Arthropods, despite the relatively limited number of motoneurons, outperform locomotion performances of a number of vertebrates. An early study on neural control of locomotion in lobsters was conducted, where the output of leg muscles during walking was analysed by electrophysiological analyses (Ayers and Davis 1977; Ayers and Clarac 1978). Later, a cockroach cyborg (Periplaneta americana Linnaeus; Blattodea: Blattidae), controlled by electrical stimuli at the antennal level routing locomotion decisions, was developed. Measurements collected on a styro-foam trackball–computer interface system allowed the record of the response to antennal electrical stimulation (Holzer and Shimoyama 1997). These data was used to establish a simple mathematical model and to design a insect cyborg by placing a series of electronic backpacks on the back of the animal. A microcontroller sent electrical stimulations to the insect that was steered on a coloured route by two photosensors.
Very recently, motion identification and localization were obtained in a cockroach species of public health importance, Gromphadorhina portentosa Schaum (Blattoidea: Blaberidae), equipped with a custom board with a five degrees of freedom inertial measurement unit, transmitting accelerometer and gyroscope data, to accurately detect the cockroach’s movements (Cole et al. 2017).
Later, Zhang et al. (2016) controlled the leg motion of the beetle Mecynorhyna torquata Drury (Coleoptera: Scarabaeidae), by exploiting a fuzzy feedback control system that sends directly to the antagonistic pair of muscles two separate electrical stimulations. The fuzzy control system modulated its internal parameters depending on different situations. The authors were successful in moving M. torquata legs in a predetermined trajectory. This system was very simple, requiring a low power for hardware, and it could be used on different beetles with different responses to the same electrical stimulation. Moreover, Cao et al. (2016) electrically stimulated eight leg muscles via eight pairs of implanted electrodes in M. torquata producing protraction/retraction and elevation/depression motions in both forelegs of the insect. Also, a vision-based automated system for insect navigation in G. portentosa has been designed (Latif et al. 2016).
Notably, in another study, human intention was translated to a cyborg G. portentosa, remotely controlled by applying micro electrical stimulation to the nerves of its antennas, where steady-state visual evoked potential (SSVEP)-based electroencephalography (EEG) was used as a robust brain–computer interface (BCI) (Li and Zhang 2016, 2017).
Besides, in terrestrial scenarios, jumping locomotion represents a valuable strategy adopted by a number of animal species, since it can be highly functional to locate food and mates, as well as to avoid predators. It has been reported that jumping insects generate accelerations far exceeding gravitational acceleration, thus producing exceptional jumping performances. To control this ability, the jumping response was elicited in the desert locust Schistocerca americana Drury (Orthoptera: Acrididae), by stimulating, with constant current square wave pulses, the metathoracic T3 ganglion, which coordinates the neuromuscular activity leading to a jump (Giampalmo et al. 2011).
Furthermore, several attempts aimed at the control of terrestrial locomotion have been carried out in vertebrates. Wenbo et al. (2009) induced artificial locomotion on the lizard gecko Gekko gecko Linnaeus (Squamata: Gekkonidae), by stimulating the midbrain with electric signals. Electric stimulations supplied by the implanted electrodes successfully elicited several locomotor behaviours in both anesthetized and awake geckos.
Kim et al. (2016) developed a display mounted on the head of the turtle Trachemys scripta elegans Wied. (Testudines: Emydidae), with wireless communication, and a stimulation device that produced the turtle’s escape behaviour remotely controlling its movements relying to a human brain–computer interface.
Rats (Rattus norvegicus Berkenhout; Rodentia: Muridae) were widely exploited to develop zoo-artifical organisms because of their well-developed motion and perceptual abilities, as well as their ease of breeding in laboratory. Movement control is generally based on the principles of “virtual award” and/or “virtual punishment” (Wang et al. 2013). Since an electrical stimulus can produce a cue or a reward, depending on the site of brain stimulation, virtual learning, involving direct stimulation of somatosensory cortices (SI) as cues and of medial forebrain bundle (MFB) as rewards, can be developed to guide the rat (Talwar et al. 2002; Xu et al. 2004). Talwar et al. (2002) used the virtual award principle, producing the first example where a rat was controlled by humans to steer, jump or climb in several environment scenarios. Huai et al. (2009) steered rats with the principle based on virtual punishment, which eliminates training in advance. The thalamic ventral posterolateral nucleus (VPL) and amygdale nucleus (AMY) of the rat were stimulated with electrical signals to affect its locomotion causing changes of direction and run-off. This method showed better outcomes in controlling rat moves with respect to the method based on virtual rewards, and a lower stimulation intensity was necessary. In order to induce a turning response in rats, Xu et al. (2016) investigated two different areas of the brain. The effect of the electric stimuli was observed in the ventral posteromedial (VPM) thalamic nucleus and the barrel field (BF) cortex. The authors demonstrated that a better control of a rat navigation can be achieved by stimulating VPM. Maintaining immobility during navigation is also crucial. Lin et al. (2010) investigated the immobile behaviour evoked by dorsolateral periaqueductal gray (dlPAG). By implanting electrodes in the brain of rats, motion and motionless states were triggered after stimulating the MFB or dlPAG during navigation. Interestingly, Yu et al. (2016a, b) developed a vision-based automatic training system, controlling a rat by brain electrical stimulation, to reduce the time-consuming training procedure.
Later on, Wang et al. (2015, 2017) designed a rat robot able to perform automatic navigation and to find target objects thanks to incorporated object detection algorithms. Furthermore, since a single stimulus was not enough to trigger rat motion, a closed-loop stimulation model giving a series of stimuli to the rat was developed allowing it to perform a motion successfully. Videos captured by a miniature camera mounted on the rat were wireless transferred to a computer developing detection algorithms that enable the rat robot to navigate automatically and find targeted objects. Remarkably, Yu et al. (2016a, b) demonstrated the principle of a “cyborg intelligence” (e.g. an animal with augmented intelligence), by producing a computer-aided rat able to quickly deal with the maze escape task. Two microelectrodes were implanted into the medial forebrain bundle of the rat’s brain, and the whisker barrel fields of left and right somatosensory cortices were the areas where the other two pairs of micro electrodes were implanted. The computer-aided maze solving system sent directional stimuli to the backpack of the rat via Bluetooth. Brain–machine interface systems have been recently developed since advances on methods for the recording of neural signals have enabled to collect brain neural activities (Wu et al. 2013). The first attempt of using brain–machine interfaces for interacting with a rat was proposed by Wu et al. (2014). Researchers incorporated a speech translator module, which produces electric stimuli to the brain to control the rat locomotion by translating speech commands by humans. Virtual punishment was exploited by Yang et al. (2015) to develop a navigation control system in pigeons (Columba livia domestica Gmelin; Columbiformes: Columbidae). This was obtained by stimulating the left and right nucleus dorsalis intermedius ventralis anterior (DIVA), and the periaqueductal gray (PAG) region of the brain. This system was able to send multi-mode electrical stimuli to the brain of the pigeon, which turned left or right to escape the virtual fear by stimulating the DIVA and moved forward when the PAG was stimulated. A micro-embedded computer sent remotely digital commands to a receiver microprocessor module mounted on the pigeon’s head producing biphasic TTL pulses to the brain. Huai et al. (2016) proposed microelectrodes for multiple brain region synchronization, which were implanted in different regions of the brain of a pigeon at the same time. In this way, the animal control was higher, if compared to the original electrodes.
In a pilot study, liquid crystal polymers were used as depth electrodes implanted in the pigeon brain for wirelessly navigation (Seo et al. 2017).
The walking forward has been recently obtained in pigeons by microstimulating the intercollicular nucleus of this avian species that is thought to correspond to the midbrain periaqueductal gray region in mammals (Wang et al. 2018). The time and the accuracy of the response were significantly improved compared to the previous experiments in which archistriatum was stimulated. Developing a swarm of bio-hybrid organisms aimed to form a mobile sensor network would be very useful in environmental monitoring applications or to locate survivors after natural disasters (Bozkurt et al. 2016). This was the purpose of some studies that—after implanting electrodes to the antennae of Madagascar hissing cockroach G. portentosa—equipped them with wireless electronic backpacks and solar panels (Latif et al. 2014; Dirafzoon et al. 2017a, b). The bio-hybrid cockroaches were kept in a target area by using an automated platform, which defined virtual barriers. Furthermore, solar panels ensured automatic charging of electronic backpacks improving the duration of the cockroach searching activities. To make feasible a swarm of bio-hybrid insects, Erickson et al. (2015) investigated and quantified several electrical stimuli parameters (i.e. pulse types, amplitude, frequency and duration) eliciting controlled navigation in G. portentosa (Fig. 4a). In addition, locomotor response to electrical stimuli was investigated through a trackball. This study demonstrated a notable response by delivering bipolar voltage stimuli to antennae and cerci of cockroaches.
3.1.2 Controlling animal aquatic locomotion
Aquatic and underwater environments represent scenarios where bio-hybrid organisms can have successful applications, although researches on locomotion control in fishes are scarce. Artificially induced swimming behaviour was obtained in the goldfish Carassius auratus auratus Linnaeus (Cypriniformes: Cyprinidae) (Fig. 4b) by stimulating a small mesencephalic area via an electrode inserted in the fish brain (Kobayashi et al. 2009).The fish was free to move and was equipped with a wireless controller delivering electrical stimuli to the medial longitudinal fasciculus (Nflm), which has an important role in activating fish swimming movements (Kobayashi et al. 2009; Uematsu and Todo 1997). Forward movement was elicited by stimulating sites on and off the midline, steering fish towards the stimulated side. Thus, Nflm direct stimulation, affecting trunk and tail movements, allowed to control swing fish robot in the horizontal plane.
Peng et al. (2011) designed and controlled fish robots by delivering electric stimuli to the brain of carps, Cyprinus carpio Linnaeus (Cypriniformes: Cyprinidae). The authors sent stimuli through electrodes implanted into the corpus cerebelli of fish, observing that steering directions occurred in the opposite side of the stimulation sites, whereas forward and backward movements were all induced by stimulating the midline of the corpus cerebelli. Overall, these studies showed successful artificially induced locomotion in animals for the development of swimming zoo-artifical organisms.
3.1.3 Controlling animal flight
Potential roles of micro-air-vehicles and nano-air-vehicles (MAVs/NAVs) are well known by industry, commercial and military applications (Ellington 1999). Even if great advancements have been carried out to improve MAVs/NAVs performances, insects still show incomparable flight abilities. Flight-controlled insects could be directly used as MAVs/NAVs by artificially modulating the wing movements using flight muscles (Sato and Maharbiz 2010). However, the electronic system implantation is a critical phase since it can produce tissue damages and negatively affect the insect flight performances (Bozkurt et al. 2008).
Bozkurt et al. (2007) developed a method to insert a microsystem in insects at the pupal stage, named early metamorphosis insertion technology (EMIT), able to electrically stimulate wing muscles. This method exploits the fact that in holometabolous insects, the latest juvenile stage (pupa), undergoing metamorphosis, is almost motionless. The authors inserted polyimide thin-film-based flexible microprobes in the thorax of Manduca sexta Linnaeus (Lepidoptera: Sphingidae) pupae and, afterwards, they actuated the wing motion in emerging adults, controlling steering during tethered flight. Implanting artificial components during metamorphosis allowed a better interface and a strong attachment with the insect tissues enabling to avoid adverse effects on the insect flight performance. Furthermore, the microsystem was implanted in the moth before its emergence; thus, the insect did not consider it as a foreign object. The manipulation of flight direction changes was explored in tethered M. sexta, designing a flexible electrode to implant in the insect abdomen at pupal stage (Tsang et al. 2008). Electrical stimulation of the abdominal nerve cord in adult moths produced variations in abdomen angle that, during flight, allowed to modify the moth flight direction.
Flight control can be obtained not only in tethered insects, but also sending radio controlled signals triggering wing stroke movements (Bozkurt et al. 2009). By implanting electrodes in the brain and the thorax of M. sexta via EMIT, emerged adults were connected to the control electronics as well as to a plastic stick holding a helium balloon. The lifting force of the balloon relieved the weight of foreign components. A remote controller sent radio signals to a receiver mounted on the mouth and used to stimulate the neuromuscular system of the insect routing the flight.
Sato et al. (2009) guessed that insects with asynchronous flight muscles do not need to continuously send signals to the muscles for each wing oscillation; thus, less controller power would be involved over insects with synchronous flight muscles. They inserted neural probes in M. torquata beetles bodies at pupal stage and then mounted a radio frequency system in emerged adults (Fig. 4c). The radio controller mounted on the pronotum delivered stimuli to the brain and to the flight muscles, enabling to start and stop flight behaviour and to steer the beetle during flight.
Bao et al. (2011) selected Apis mellifera ligustica Spinola (Hymenoptera: Apidae) for developing a flying insect robot. After implanting two electrodes into the left and the right optic lobes, various electrical stimulus pulses were delivered to a tethered honeybee with the purpose to investigate how they condition flight behaviour. Below a specific amplitude threshold, flight did not occur. Exceeding the threshold of 3.5 V, the flight was successfully triggered and continued for a period of time without additional stimulus, since honeybees had asynchronous muscles.
Investigating which muscles are involved in grading turning in insect free flight is essential to obtain an accurate steering control. For this purpose, a group of researchers used a miniature radio system mounted on M. torquata for remote electrical stimulation, proving that the third axillary muscle plays a key role in regulating steering in beetles (Sato et al. 2015).
Poon et al. (2016) successfully evoked flight initiation in M. torquata by mounting a miniature stimulator on the insect pronotum and applying 2–4 V, 100 Hz, 20% duty cycle, biphasic square pulse trains between two electrodes implanted into the optic lobes, massive neural clusters of compound eyes. Flight cessation was obtained by a single DC pulse applied to the same sites. To turn the insect, the same pulse trains were applied to the left or right basilar flight muscle. Also, Choo et al. (2016) developed a reliable electrical stimulation protocol, obtained by analysing a pair of dorsal longitudinal muscles that elicited flight in M. torquata. To study insect flight, a wireless backpack was mounted on the pronotum of M. torquata, relying to a silver electrode implanted on the 3Ax muscle on each side of the insect (Doan and Sato 2016). During free flight, the beetle was successfully turned depending on the side where the muscle was stimulated.
Electric stimuli directed to the basalar and third axillary muscles of M. torquata, enabled the control of the beetle’s horizontal flight (Li et al. 2018). Authors developed a controller based on a proportional derivative feedback to decrease the effect due to the insect adaptation to long-lasting electrical stimulations.
Simulation results of a novel chip architecture were obtained by Wang et al. (2016), showing that the system would work efficiently to control the flight of free honeybees.
However, avoiding the invasive implantation of electronics in animals represents the best path to follow in this research context in order to produce highly efficient systems as well as more acceptable approaches from an ethical point of view.
In this framework, Zheng et al. (2017) attempted to control the flight of bumblebees (Bombus spp. Hymenoptera: Apididae) in a non-invasive way. The authors proposed a control method based on real-time virtual reality and reinforcement learning that consisted in a configurable LED display system as visual stimulus that was projected on the insect compound eyes.
3.2 Controlling robots by biological inputs
The control of animal behaviour by sending artificial stimuli to their nervous and/or muscular systems is of great interest in ethorobotics. Besides, another fascinating field of research is the development of robots responding to cues produced by biological interfaces. In this context, the biological part drives the artefact, creating a reverse condition over zoo-artificial organisms discussed in the paragraphs above.
Animals have extraordinary sensorial skills to detect and recognize olfactory, auditory and light stimuli, which can be used to control machines. Kuwana et al. (1995) used the antennae of silkworm moths, Bombyx mori Linnaeus (Lepidoptera: Bombycidae), to develop a bio-hybrid sensor able to measure pheromone concentration in the air, thus guiding a mobile robot towards the source of this stimulus. Two male moth antennae were fixed to a mobile robot, and their electrical activity was monitored by an electroantennograph and then used to successfully steer the robot along the pheromone trace, as naturally occurs in male moths. A similar mobile robot, mounting silkworm moth antennae, but characterized by a smaller size and a higher level of performances, has been developed (Fig. 4d), in order to test it in a wind tunnel and analyse the robot behaviour with the same experimental apparatus normally employed for moths (Kuwana et al. 1999). With the aim of locating an odour source in several downwind areas by using an insect antenna bio-hybrid sensor, Myrick and Baker (2010) combined an electroantennogram (EAG) odour detection system with a GPS and a 2D anemometer. The EAG was simultaneously recorded in two moths species, Heliotis virescens Fabricius and Helicoverpa zea Boddie (Lepidoptera: Noctuidae). In addition, by using Lagrangian atmospheric dispersion models, to describe odour parcel turbulence and Bayesian inference to quantify the uncertainty in the considered source parameters, they obtained the odour source location in a 0.2 m radius from the device.
A description of a protocol for tracking odour traces and sources using insect antennae was recently provided by Martinez et al. (2014). They also used electroantennograms to detect the response of olfactory neurons. In order to guide an autonomous robot mounting real insect antennae, they developed an interface between the electroantennograph electrodes and the robot. This robotic platform represents a valid appliance to investigate animal odour search strategies. Besides, in a very recent study, a tethered male of B. mori successfully routes a robot to localize a targeted odour source (Ando and Kanzaki 2017).
Biological inputs also can be obtained by photoreceptors, driving robots by light stimuli. Jadhav et al. (2012) designed flexible microfabricated neural interfaces to be inserted at level of eyes of Zophobas morio Fabricius (Coleoptera: Tenebrionidae) during the pupal stage. Adult beetles with these interfaces were subjected to optical stimulation experiments allowing to record their neural activity. This interface system represents a worthy attempt to simplify the production of bio-hybrid organisms exploiting animal sensory systems.
As a further step, a bio-hybrid approach was proposed by Stefanini et al. (2012), who developed a new bio-inspired robot extensively tested in aquatic environment. This robot was inspired by the lamprey Lampetra fluviatilis Linnaeus (Petromyzontiformes: Petromyzontidae), a vertebrate model where the neural control system for goal-directed locomotion is well documented at a cellular level (Grillner et al. 2007). This robotic platform enabled the idea of interfacing the artefact with real lampreys, to directly control the robot by the neural activity of the animal (Manfredi et al. 2013).
Lastly, rat heart muscle cells were added to a microfabricated gold skeleton and a rubber body to produce a bio-hybrid robot swimming like a ray fish (Park et al. 2016). The undulatory movements of the robot were obtained by the genetically engineered cells that gave responses to light cues allowing the robot to follow a light source.
3.3 Energy harvesting from bio-hybrid organisms
A number of efforts aimed to develop bio-hybrid organisms as new methods to harvest and manage energy from animal physical exertion as well as biological metabolic compounds. Aktakka et al. (2011) presented a vibration energy scavenger that can be employed during the flight of Cotinis nitida Linnaeus (Coleoptera: Scarabaeidae). This generator utilizing non-resonant piezoelectric bimorphs was able to convert 1–10% of mechanical energy deriving from beetle wing vibrations into electricity. One generator for each wing was placed, and electrical output was obtained from wing motions without disturbing effects on the flight of the beetle. Further studies focused on harvesting energy from internal metabolic compounds of animals. Biofuel cells were implanted in a snail, Neohelix albolabris Say (Stylommatophora: Polygyridae), producing electrical power over a long period of time using physiologically produced glucose as a fuel (Fig. 4e), (Halámková et al. 2012; Katz and MacVittie 2013). In addition, enzyme-based biofuel cells implanted into living American lobsters, Homarus americanus Milne-Edwards (Decapoda: Nephropidae), were able to activate an electrical device by generating electrical energy (MacVittie et al. 2013). Schwefel et al. (2015) developed a bio-hybrid cockroach by implanting a biofuel cell in living Blaberus discoidalis Audinet-Serville (Blattodea: Blaberidae). The biofuel cell had a bienzymatic anode that dissociated trehalose to glucose and then oxidized it to gluconolactone by the enzyme glucose oxidase. The cockroach, presenting a custom-designed oscillator mounted on his back powered by the biofuel cell, was capable of generating and transmitting wireless signals to an external receiver. The same experiment was successfully conducted by implanting biofuel cells in the moth M. sexta. Shoji et al. (2016) produced a self-powered robot represented by a G. portentosa endowed with a biofuel cell and a micro-wireless sensor module for environmental monitoring. The biofuel cell generated electric power from glucose in the insect haemolymph.
These self-powered bio-hybrid organisms could have biomedical applications, and they could be utilized in environmental monitoring, homeland security and military applications (MacVittie et al. 2013).
Recently, El Ichi-Ribault et al. (2018) developed a device including an enzymatic biofuel cell implanted in the abdomen of a rabbit and connected to a wireless tele-transmission system, providing an interesting achievement concerning the output obtained from an enzymatic biofuel cell integrated in a mammal.
4 Research progress in animal–robot mixed societies
In recent years, as illustrated in Fig. 5, robotic devices have been increasingly proposed to shed light on hot topics in animal behaviour and behavioural ecology, with special reference to the study of social interactions among animals and robots. Recent technology has made it possible, by developing interactive robots, which can perform complex behavioural tasks, adapting their behaviour to signals from living animals and the environment (Garnier 2011). Furthermore, these robots allow to deliver selected cues triggering animal responses or manipulating the animal collective behaviour.
Robotics devices in mixed society can be useful for several reasons, i.e. to validate in silico systems, creating a closed-loop society with real animals, or to observe and modify the collective behaviour. Thus, robotics represents a sophisticated approach contributing to the study of social behaviours in animals, with potential applications in the control of animal populations in agriculture or in the improvement in animal farming conditions (Garnier 2011; Halloy et al. 2013). In the following paragraphs, we examined the main issues in these research fields.
Figure 6a, b shows the current trend of behavioural studies using robot replicates. Figure 6a shows the number of published research items per year considering the different topics. Most of the articles focus on the animal–robot interactions as well as on the level of attraction of the replica. The latest was investigated to clarify which morphological features boost the chances of acceptance by living animals, thus the possibility to influence their behaviour. Figure 6b shows the number of published research items per year according to the species under investigation. The largest number of studies is on fish species. This can be due to the easier conditions of fish maintenance in the laboratory, compared to other vertebrates, such as mammals. Besides their easy rearing, fishes allow us to investigate a number of behavioural interactions, such as the influence of swimming acceleration and speed, as well as peculiar morphological features. However, studies involving fish collective behaviour are often focused on single mechanisms acting on group dynamics that limits experimental demonstrations of these displays (Ioannou 2017). Indeed, multiple mechanisms are involved in producing improved group behaviours in fish species. In insects, collective behaviour is often demonstrated by the relatively clarity on how information and decision-making occur (Halloy et al. 2007), while in fish the stimuli transferring information in the shoal are more complex and unclear (Ioannou 2017). In addition, contrary to colonies of many social insect species, shoals of fish include unrelated subjects and thus cooperative behaviours are in several contexts eclipsed by competitive interactions (Halloy et al. 2007; Couzin 2009). Also, the heterogeneity of cognitive abilities among individuals of a shoal of fish can pose some difficulty in their study compared to other animals such as several arthropods species (Ioannou 2017).
4.1 Animal perception of robots as conspecifics
Robot features are extremely important for their acceptance by single individual and animal groups as well. In order to elicit interactions with single individuals or group, the robot as to be treated as a conspecific. This is particularly important, especially in gregarious animals. In this context, the use of robots may help to determine which features and signals are critical to trigger conspecific attraction.
Motor patterns seem to play an important role in conspecific attraction. For example, Marras and Porfiri (2012) determined the effect of tail-beat frequency of a robotic replica on individual golden shiners, Notemigonus crysoleucas Mitchill (Cypriniformes: Cyprinidae). They found that locomotion is a determinant cue to evoke fish preference, since N. crysoleucas individuals are more attracted towards the robot with tail-beat movement rather than when it is statically immersed in the water. An explanation can be because it provided considerable hydrodynamic advantages.
Another study showed that the water flow is determinant for schooling, because the coordinate swimming reduced energy expenditure (Polverino et al. 2013). A similar experiment, but in a closed-loop interaction, was performed with zebrafish, Danio rerio Hamilton (Cypriniformes: Cyprinidae) (Kopman et al. 2013; Bonnet et al. 2016a, b), where the tail beat was real-time controlled based on fish motion. Bonnet et al. (2016b) showed that a robotic zebrafish replica is able to attract a shoal of zebrafish inside of a circular corridor according to the speed of the device, even if this effect was not enough to guarantee a full integration. Later on, the authors developed a robotic system that achieved similar motion patterns displayed by zebrafish for direct interactions with D. rerio, establishing a closed-loop interaction (Bonnet et al. 2017a, b).
Another important issue in conspecific attraction is the visual information, which seems to be crucial in many species, especially colour and shape features. Interestingly, to understand how the body size affects the social behaviour in zebrafish, Bartolini et al. (2016) investigated behavioural responses of zebrafish to shoals of 3D printed conspecifics of different sizes. In addition, several experiments combined visual and motion features to demonstrate the effective attraction of the robotic replica by living animals. It was found that the variations in morphophysiological and locomotory features were determinant of attraction towards the robotic fish (Abaid et al. 2012; Polverino et al. 2012). The behavioural response of zebrafish individuals to small shoals changed according to the variation in the aspect ratio, colour pattern, tail-beat frequency and speed of the robotic fish. The attraction is maximized when the robotic fish replicated the animal counterpart in aspect ratio and colour (Bonnet et al. 2014). In contrast to the aforementioned observations on zebrafishes, results on mosquitofish, Gambusia affinis Baird &Girard (Cyprinodontiformes: Poeciliidae), showed that they were repelled by mosquitofish-inspired robotic replica, independently of its aspect ratio or swimming depth (Polverino and Porfiri 2013). A recent study on the acceptance of biomimetic replica of the Trinidadian guppy, Poecilia reticulata Peters (Cyprinodontiformes: Poeciliidae), provides a description of the effect triggered by different appearances, motion patterns and interaction modes on the acceptance of the artificial fish replica. Integration of realistic eye dummies along with natural motion patterns led to a significant improvement in the robotic replica acceptance level (Landgraf et al. 2016). Bierbach et al. (2018a) investigated which cues produced by a robotic fish evoke acceptance in fish species. Authors tested the effect of the biomimetic artefact on two populations of Poecilia mexicana Steindachner (Cyprinodontiformes: Poeciliidae): the first population was adapted in dark habitats (e.g. caves), while the second population was adapted to natural light conditions. In light conditions, the acceptance of the robotic fish occurred in both populations, while in dark conditions the robotic fish did not have effects, since probably other cues are needed in the absence of light (e.g. chemical cues, auditory stimuli). In addition, Bierbach et al. (2018b) studied individual responsiveness to social stimuli in P. reticulata individuals, by controlling a biomimetic robot, to avoid influences rising from mutual interactions among fish. Authors observed that responses to social stimuli are independent traits, not correlated with other individual behavioural displays.
Since zebrafish behaviours depend on social interactions as well as on their position in the environment, a multilevel model describing the zebrafish collective behaviours was developed to control a robot that was integrated socially in zebrafish group (Cazenille et al. 2017). In this case, optimization methods to calibrate automatically the controllers of a robotic agent according to the animal behaviour were developed (Cazenille et al. 2017).
However, the reconstruction of 3D trajectories should be considered (Macrì et al. 2017), since traditional behavioural displays observed in 2D can undermine data integrity. Ruberto et al. (2016, 2017) studied the zebrafish response to a 3D-printed conspecific replica moving along realistic trajectories (Ruberto et al. 2017).
As observed, the design of biomimetic robots able to interact with fish is complex since it should ensure a luring capability, as well as the acceptation of the robots by the animals as a conspecific. Stochastic model-based behaviours of the robot were proposed by Cazenille et al. (2018a), to integrate it in a group of D. rerio.
Animal–robot interactions are benefitting from progress in technology that provides more complex systems to be exploited in animal behavioural studies. Worth to be mentioned is the recent use of virtual reality to control the animal behaviour (Stowers et al. 2017). This strategy boosts detailed surveys into neural and behavioural functions through the accurate control of sensorimotor feedbacks in animals moving in 3D scenarios. Besides the study of morphological and mobility features, Worm et al. (2014) investigated the acceptance of a robotic counterpart of the weakly electric fish Mormyrus rume Valenciennes (Osteoglossiformes: Mormiridae). This fish is able to communicate by using electric signals, and for this reason, the dummy fish was equipped with electrodes for the generation and reception of signals. Results showed that the animals preferred to interact with the dummy fish in the presence of electric signals (Donati et al. 2016). Worm et al. (2018) observed an enhanced interaction between M. rume individuals and a robotic fish when the artificial agent generated dynamic echo playback of biomimetic electric organ discharges compared to inter-discharge intervals displays statically and randomly produced.
4.2 Robots for modification of behaviour through different biomimetic stimuli
Animals rely to different senses to perceive external stimuli. A complex sensory system allows to perceive the surrounding environment and its dynamic changes. Therefore, understanding how it works is a fundamental question of the animal cognition. Robots can represent an effective solution to this question, creating different combinations of perceptual cues, by modifying their features, which can elicit different responses in animal behaviour. For example, a mechanical honeybee model (Michelsen et al. 1992) was used to understand the role of dance in the transfer of foraging information to follower bees. The robot was able to separate waggle dance and sound-producing wing vibration of the dance, and the experiments demonstrated that the waggle dance is important to transmit information about distance and direction of a given food source. Close-related studies have been conducted subsequently to investigate the acceptance of the robot in the hive (Landgraf et al. 2010, 2012).
Recently, robotic agents were used in neuroethology studies focused on invertebrate brain lateralization. In particular, Romano et al. (2017b) investigated the lateralization of escape and surveillance responses in Locusta migratoria Linnaeus (Orthoptera: Acrididae), during predator–prey interactions with a robot inspired to a natural enemy of locusts, the Guinea fowl, Numida meleagris (Linnaeus) (Galliformes: Numididae). Benelli et al. (2018) developed a mechatronic device mimicking a potential host, providing a combination of visual and olfactory cues important in the host-seeking behaviour in ticks. Results showed a population-level left-biased use of forelegs in Ixodes ricinus L. (Acarina: Ixodidae), during climbing on the robotic host. The action performed by these artefacts enabled a high degree of standardization and accuracy of the stimuli presented, which represent crucial issues in the study of lateralization, especially in experiments involving invertebrates.
The possibility to deliver different kinds of cues separately through robots interacting with animals allows us to dissect the relative importance of the different stimuli triggering a response in living organisms. This is hard in classical natural observations because all the cues are presented together and the response to a single feature in a complex (multi-stimuli) behaviour is not measurable. Robots make possible the deconstruction of different behavioural components to measure the individual responses. For example, a study focused on display modification to visual signals in the lizard Sceloporus graciosus Baird & Girard (Squamata: Phrynosomatidae) used a robotic counterpart to investigate two types of headbob displays (special, typical and unusual) both in short- and in long-term assays (Martins et al. 2005). No evidence for immediate changes in signal structure or long-term differences was detected. However, the lizard was more agitated and produced highly aggressive displays when exposed to unusual headbob displays. Another study focused on the impact of different displays in Anolis sagrei Duméril & Bibron (Squamata: Dactyloidae) (Partan et al. 2011), showing that the social response to the movement of the robot was higher in signature pattern than alternate pattern, despite that they had a high degree of variability in signature bobbing display. In another study, the frog Epipedobates femoralis (Anura: Dendrobatidae) defends their territory and emits visual cues (i.e. vocal sac pulsations) and auditory cues (i.e. advertisement cues). In this study, the two stimuli have been presented with spatial disparity or/and time delay and the result showed that bistimuli with temporal overlap evoked aggressive behaviour while with lacking overlap they were ineffective. For spatial disparity, the response was the same up to 12 cm. Similar studies have been conducted on ground squirrel, Spermophilus beecheyi Richardson (Rodentia: Sciuridae) (Rundus et al. 2007), where the importance of infrared signal to deter rattlesnake predator was investigated. Natural observation showed that squirrels, when confronting infrared sensitive rattlesnakes, add an infrared component to their snake-directed tail-flapping signals, whereas, when confronting infrared-insensitive snakes, the tail flagging is without augmenting infrared emission. The robot squirrel simulated the same behaviour, and the results showed a greater shift from predatory to defensive behaviour of the rattlesnake when the infrared was present. In similar studies on multi-stimuli with the squirrel Sciurus carolinensis Gmelin (Rodentia: Sciuridae) (Partan et al. 2009, 2010), a robotic counterpart presented different alarm cues (i.e. visual and acoustic ones), in different combinations, investigating the response of wild squirrels (Fig. 7a). Multimodal signals showed higher responses to cue combinations over separate signals.
Siamese fighting fish males perform stereotyped and vigorous aggressive displays towards conspecific males, related to their highly territoriality. On this basis, Betta splendens Regan (Perciformes: Osphronemidae) was tested as model system to investigate the effect of a robot fish eliciting aggression (Romano et al. 2017a). The authors evaluated how multiple signal systems, including a light stimulus, affect aggressive responses in this species. The efficacy of an artificial opponent eliciting aggressive behaviour in fish can be boosted by exposure to multiple signals. The optimized cue combination delivered by the robotic fish demonstrated its help to predict escalating levels of aggression.
A biomimetic soft robot able to 3D move in underwater environments was developed by Katzschmann et al. (2018), to be perceived as a fish, thus approaching and studying the aquatic life.
4.3 Robots in social interaction
Animals live in a dynamic social environment with frequent changes, and they continuously update the information they hold from conspecifics (e.g. foraging for food and mates). The transfer of information between individuals is a complex behaviour, hard to fully understand and control. The discrimination between different features of a selected cue and information processed by the individual still needs research efforts. In this scenario, robots can be extremely useful to solve this problem, allowing to control independently specific aspects of behaviour.
The relevance of social interactions was investigated in a study on dogs interacting with commercially available quadrupedal robots (Kubinyi et al. 2004), where the latter elicited either aggressive or playful responses from live dogs. Results showed that age and context influence the social behaviour of dogs. A laboratory study on learning showed that rats followed a remote-controlled electromechanical rat (WM-2) to sources of food (Takanishi et al. 1998). Live rats recognized the movement of robot and it helped them to learn response to stimulations. Later, the authors increased the complexity of the interaction by using a legged rat robot, which successfully taught the rat a lever pushing task to get food (Laschi et al. 2006; Patanè et al. 2007). Recently, Shi et al. (2010, 2013, 2015) investigated the interaction between a robotic rat and living rats (Fig. 7b), observing that rats with more active behavioural parameters are more susceptible to being adjusted by the robot.
In animal interactions, it is possible manipulate the behaviour of individuals by using animal–mimicking robots (Fernández-Juricic et al. 2006). In this context, the foraging and scanning behaviour of the house finch Carpodacus mexicanus Muller (Passeriformes: Fringillidae) in response to different types of behaviours from artificial flocks were studied. Finches spent more time foraging when the robots simulated body movement that could be associated with successful foraging behaviour. A comparison on social information transfer in three different bird species (i.e. European starling, house finch, brown-headed cowbird) (Fernández-Juricic and Kowalski 2011) showed that in all species, a nonlinear decrease in social information flow with increasing distance between the robots and live bird was found. This was more pronounced in species with lower visual acuity, because animals with higher visual acuity can detect changes in the behaviour of conspecifics from farther distances, which may have consequence in spatial distance between individuals within a flock. High visual acuity is due to regions of the retina with a localized high density. These regions project into a visual space, which may require, in order to detect changes, to move the body and modify the position within the group (Butler and Fernández-Juricic 2014).
Robots can be used to study the learning occurring at a particular early life stage, namely “imprinting”. There is a brief receptive period, typically soon after birth or hatching, in which the animal is able of rapid learning and establishing a long-lasting response to a specific stimuli (Lorenz 1935). The input can be a visual, auditory or tactile experience that creates the attachment with a given object. Usually, in nature, this object is a parent, but in experiments other animals and inanimate objects, as robots can be used. The idea is to use the robot to fully control an animal’s experience by exposing it to a single or multiple robots from the birth. It was shown that chicks, Gallus gallus domesticus Linnaeus (Galliformes: Phasianidae), can be imprinted on robot introduced as a surrogate hen which can spatially interact (Gribovskiy et al. 2010, 2015, 2018) (Fig. 7c). In a social experiment with the Japanese quail chick Coturnix coturnix japonica Temminck & Schlegel (Galliformes: Phasianidae), a mobile robot carrying a heat source was used to control the motion of quail chicks. Chicks showed better spatial abilities when raised with heated mobile robot than when exposed to an immobile heater. These experiments demonstrated that there was a measurable attachment to the robot (favoured when the chick encounters the robot early after the birth), and this attachment was also combined with a synchronization of chick and robot activity (Jolly et al. 2016). Finally, in a study assessing social cues in the Australian brushturkey Alectura lathami Gray (Galliformes: Megapodiidae), a series of robot chicks was built with the same characteristics of the real one, except for one made with pecking movements. Chicks preferred a pecking model over static or scanning models (Göth and Evans 2004), suggesting that social responses of chicks depend upon conspecific motion patterns.
Animal behaviour may vary post-exposure to different traits. This is particularly true for courtship behaviour, where males often differ in their courtship sequence and these differences modify the rate of success at convincing female to mate. Interactive robot can provide different combinations of mate values to evaluate their effect to investigate the relevant features in mating choice. A study on courtship in fish Lucania goodei Jordan (Cyprinodontiformes: Fundulidae) (Phamduy et al. 2014) measured the preference of a fertile female for male robot movements by varying aspect parameters (e.g. colour: red, yellow or blue skin) during motion of classical courtship behaviour. In satin bowerbirds Ptilonorhynchus violaceus Vieillot (Passeriformes: Ptilonorhynchidae) (Patricelli et al. 2002, 2006), a robotic female robot was used to study male displays adjusted to female’s response and to test how and when the males reduce the intensity of their courtship displays after females starling (Fig. 7d). Another example is the study of head-bobbing movements in the lizard S. graciosus that can elicit both aggressive and mating displays (Martins et al. 2005). Results showed that two aspects of head-bobbing displays are independently meaningful components interpreted different by different receivers. Males are attended to posture and females to number of head bobbing, using it to distinguish male courtship. Multimodal signals during courtship were also investigated in Tùngara frogs (Physalaemus pustulosus Cope; Anura: Leptodactylidae) (Taylor et al. 2008) to test if females preferentially respond to multimodal stimuli (i.e. auditory plus visual cues) over a unimodal stimulus (i.e. auditory cues only). In this study, the females interacted with a robotic calling male frog, validating the theory that females positively select multimodal cue combinations over the same stimuli tested singly (Fig. 7e). Another recent example is the study of fiddler crabs (Uca mjoebergi Rathbun; Decapoda: Ocypodidae), arm movement in mating preferences, where four robotic male crabs were used to produce asynchronous and synchronous movements during the courtship dance (Reaney et al. 2008). A last example is the study of cricket behaviour during courtship, based on pose estimation (Kawabata et al. 2013). The real cricket interacted with a mobile robot, and the studies about behaviour were based on the animal pose. In a subsequent study (Kawabata et al. 2014), the robot robes cricket’s pheromone and its position was controlled by designing motion patterns based on visual motion tracking of the both agents. Results showed a change in the behavior of crickets due to experience.
4.4 Robots in collective behaviour
A further application of robots is to understand and influence the animal collective behaviour (Mondada et al. 2013). The collective behaviour is a complex system that presents several levels of organization (i.e. hierarchical organization). One of the major challenges in mixed society is to design robots that can modulate the natural society towards a desired behaviour influencing the decision-making process. The resulted system is a closed-loop feedback system between artificial and natural animals where robots can react to sensory input triggered by the animals. To investigate the mechanism of shelter seeking in cockroaches (P. americana) (Halloy et al. 2007), a robot with the same behaviour of real counterpart was developed. The robots were autonomous, did not look like roaches, but were pheromone scented. It was able to recognize the shelters and to interact with real cockroaches as well as to elicit novel collective decisions. They were programmed to lead cockroaches from the favourite shelter into unsafe one in open area. Also, the Robot Sheep dog project developed a mobile robot to control a flock of ducks to go to a specific and safety place (Vaughan et al. 2000). The robot moved round behind ducks with respect to the goal and the flock moved away from the robot to the goal. When the flock reached the goal started the repulsive phase and the robot was less attracted.
Studies on collective behaviour in fish shoals are rapidly increasing. One of the reasons is the possibility to have easy access to fish and the limited space they needed. Zebrafishes is one of the main species currently studied, due to their high reproduction rate, short intergeneration time and evident shoaling tendency. For example, in Butail et al. (2013) a robot able to move at different speed varying tail beating was used to explore the fish responses (Fig. 7g). It has been showed that for group cohesion, speed is a determinant feature. Indeed, the relative distance increases with speed of the robot. Further experiments investigated the influence in shoal size and configuration (Butail et al. 2014) of fish increasing the number of robots. Stress measurements showed increased value in the presence of more robots, or with fast robot swimming alone instead of slow swimming of two robots. Other experiments focused on the collective behaviour of different fish species, including the analysis of self-organization and aggregation in Giant danios, Devario aequipinnatus McClelland (Cypriniformes: Cyprinidae) (Aureli et al. 2012). A similar study with robotic replica, but different species of fishes (i.e. sticklebacks and guppies) involved one robot that moved according to the position of fish detected through an external camera (Landgraf et al. 2013). Risk-taking behaviour of individual golden shiners (N. crysoleucas) was investigated in the presence of a self-propelled robotic fish to test the hypothesis whether the behaviour of fish can be modulated by varying the behaviour of a robotic fish and to investigate whether such response depended on fish individual boldness (Abaid et al. 2013).
Robots can be used to explore how animals select the leader and in which contests they follow it. Leadership occours when one or more individuals initiate a new direction of locomotion, followed by other group members. For example, in Faria et al. (2010) a fish-like replica was used to visually attract and drive single fishes out of a refuge and to initiate new swimming directions in both individuals and groups. Ward et al. (2008) investigated the decision-making process in a Y-maze by using a robotic replica moved on a guide line, to demonstrate the role of “quorum responses” in the movement of the fish Gasterosteus aculeatus Linnaeus (Gasterosteiformes: Gasterosteidae). An information-theoretic approach to infer leadership starting from positional data of fish was also proposed (Butail et al. 2016, 2017). Rashid et al. (2012) presented another work about leadership, which involved the use of mobile light sources for guiding swarms of brine shrimp larvae (Artemia salina Linnaeus; Anostraca: Artemiidae).
Recently, computer vision and real-time control have enabled the development of closed-loop control systems that boost the degree of biomimicry of the artefacts, by improving the attraction and the interaction of D. rerio with the robotic stimuli (Kim et al. 2018). These artefacts that are perceived as conspecifics can be used to study social processes and to affect collective decision in the fish (Bonnet et al. 2018). Interestingly, Cazenile et al. (2018a, b) presented a strategy for real-time calibration of behavioural models based on an evolutionary algorithm, to improve the integration of the robots into the shoal.
4.5 Robots as tools for scientific validation
Finally, robots can be useful for testing and validating models of behaviour in embodied simulators. We define this possibility “in artefacto”, compared to in vivo, in vitro and in silico well-established approaches. Interactive robots can be used to assess mechanistic models and the assumption they are based on (Krause et al. 2011; Manfredi et al. 2013), allowing scientists to test hypotheses about mechanical mechanisms and interactions with external cues. They can be effective tools for validation theories and natural models, thanks to the achievement of high accessibility to the environment. It is possible to embed the biological knowledge and models directly on the robots; thus, the observations can be made on such a complete physical model interacting with the real environment.
5 Lesson learned from animal–robot interaction studies and future perspectives
Overall, our review on animal–robot interaction and integration shows the mutual contribution between robotics and biological systems in science and technology. Bio-hybrid organisms can be considered advanced engineered organisms resulting from a hot trend in biotechnology, aimed at coupling the biological and robotic worlds in one entity. This approach outlined fascinating applications for both research fields, and its potential is still minimally exploited (see Table 1). Bio-robotics researchers can boost the still clumsy adaptability of robots to the real world by exploiting animal complex architectures and their capability of negotiating unknown and unstructured environments. Furthermore, their energy autonomy is worth to be considered in long-lasting explorations. These bio-hybrid organisms could significantly outperform traditional robots used for monitoring, searching and/or rescuing activities in dangerous and intricate scenarios. Neuroscientists and biologists could rely to bio-hybrid organisms to investigate and verify neurophysiological functions and mechanisms involved in muscle activation. The latter could be exploited to carry out behavioural interactions with conspecific animals.
In addition, a visionary pathway to target is that of tightly merging a biomimetic robot with a biological nervous system (e.g. Stefanini et al. 2012; Fry et al. 2009). In this way, electrophysiological signals should control the artefacts, and conversely, sensory signals recorded by the robot could be used as feedbacks by the animals, in order to affect its behaviour.
Finally, some limitations have to be challenged to develop this scientific context. A key problem concerns the interface between living tissues and artificial devices. Here, although several studies described above were addressed to it, efforts are still needed to investigate the most suitable materials possessing good compatibility with biological tissues, as well as to reduce the difficulties for the operator to manipulate the animal model. Ethical disputes could also constrain researchers involved in this field making even more complex bio-hybrid approaches. One of the main ethical issue concerns if the contribution in science can justify the killing of animals used to develop bio-hybrid organisms. The growing sensibility in animal rights has produced many efforts to find the most suitable experimental treatment ensuring animal wellness (Lockwood 1988; ASAB/ABS 2004).
Several parameters have been exploited to define the features proving animal sentience, to understand the limit between research and moral issues (Tye 2016), and often the domain of bio-hybrid organisms is considered to go too far compared to their contribution in science (Dodd 2014). Concerning this issue, a remarkable attempt has been made by some authors, which proposed control methods based on real-time virtual reality and reinforcement learning in order to control animal locomotion in a non-invasive way (Wang et al. 2016; Zheng et al. 2017).
Regarding animal–robot mixed societies, recent technological advances made possible to control and manipulate animal behaviour. These robots are able to perceive, communicate and interact/adapt with the animals. Therefore, robots can be used to bring new capabilities producing shared collective intelligence (Halloy et al. 2013). Animal–robot interactions, in addition to new knowledge, can hold to a remarkable socioeconomic impact on our daily lives. Their application includes the control of animal populations in agriculture, the improvement in animal farming conditions, improving breeding conditions and the protection of endangered species. An interesting area for application is the use of robots as leaders, guiding animals away from a dangerous situation. Another possibility is to use robots during the imprinting phase, allowing animals to follow the robots along a suitable migration route or away from danger. A further potential application is in herding animals prior to catching or transportation. An effective herding robot must be able to interrupt ongoing animal behaviour without causing panic or flight reactions.
However, although many robots have been developed for research purposes, robots available outside of the laboratories are very rare in these contexts. A good example of artificial agents interacting with animals in the “real world” can be represented by milking robots that are becoming more and more used in cattle husbandry and dairy production (Jacobs and Siegford 2012). In this case, although robots are not biomimetic, animals directly interact with the machine without involving humans (Pastell et al. 2006). Also smart collars, wearable devices mounted directly on animals (e.g. “mounted nodes” described by Mondada et al. 2013), are very common in farms (King 2017). Furthermore, an impressive biomimetic robot (e.g. Robird) has been recently developed to be marketable as a sustainable product for bird control (e.g. to relocate birds around airports) (Folkertsma et al. 2017). This robot mimics in its appearance and flight a rapacious searching for a prey.
Service robots are having a great impact on the market in the domestic environment (Decker et al. 2011; Forlizzi and DiSalvo 2006), and innovative artificial agents used to interact/control animals would have a huge market potential as well.
The main challenge that could still limit the developing of these kinds of robots for real applications is represented by the design of biomimetic agents able to perform long-term interactions and to behave in highly unstructured scenarios (Beer et al. 1997). Further efforts on this are imperative, considering the potential that these devices would have.
Concerning ethical issues, they can rise also in this domain, since people might erroneously be concerned by the fact that scientists are building tools to potentially control all types of animals in the planet, as well as human being, using artificial devices. However, concerning potential applications of this domain should be considered that traditional strategies to manage livestock as well as to control pest and wildlife populations are often carried out by adopting unethical and non-eco-friendly methods (e.g. employment of pesticides and selection hunting). Thus, one of the aims of this domain, concerning real-world applications, is to produce advancements in animal wellness and environmental sustainability. In addition, the approach of this domain in studying animal–robot interactions is elegant and minimally stressful for animals as stated in most the works mentioned in this review. Furthermore, the use of predators in laboratory-staged predator–prey experiments and real opponents in aggressive behavioural studies are no longer ethically acceptable (Huntingford 1984; ASAB/ABS 2004). The use of robotics in studies focused on predator–prey interactions and aggressive behaviour turned out to be an excellent solution to these issues (Ladu et al. 2015; Romano et al. 2017b; El Khoury et al. 2018).
Table 2 shows the list or recent studies on animal–robot behavioural interactions. It is evident how, in the last years, the studies focused on experimental behaviour on fish, due to easy access to them and the simple breeding farm they required, although different limitations, already discussed, complicate the use of these animals as model organisms. The overlapping and managing of several communication channels, according to different behavioural contexts, as well as diverse personalities characterizing each individual, should be also considered, to ensure a high degree of reliability and biomimicry of the artefacts during the interaction.
A recent successful employment of biomimetic robots is represented by their use in neuroethology studies, with special reference to research on brain lateralization in invertebrates (Romano et al. 2017b; Benelli et al. 2018). Robots accurately provide stimuli that are fully controllable by humans and that overcome limits related to the direction/orientation of cues during these experiments.
In addition, a fascinating and noble role of biomimetic robots concerns their potential use as “artificial hosts” for feeding, in laboratory conditions, haematophagous arthropods, to study their vector competence for parasites and pathogens representing a key threat for humans and animals (Romano et al. 2018).
The employment of these agents would be perfectly in line with the 3R principle “reduce, replace and refine” addressed to limit the use of experimental animals in research, thus avoiding the transmission of diseases, ethical problems regarding animal welfare and the costs of animals rearing as well.
Lastly, an issue to be discussed concerns the fact that robots can be seen as fancy devices to “sell” such studies to the public (Garnier 2011). As many of the proposed studies have demonstrated, robotics has a crucial role in both engineering and biology researches. In addition, they could also have an interesting role as tools for the scientific dissemination and for educational purposes. However, in the case of bio-hybrid organisms, a larger number of papers have been published on engineering journals/conferences compared to biology journals/conferences (Fig. 8a). This is maybe due to weight that ethical issues have on research communities, especially in biology.
Conversely, concerning animal–robot mixed societies, the number of research articles published on engineering journals is rather balanced with research papers published on biology journals (Fig. 8b), highlighting the mutual benefit that this domain ensures to both engineering and biology research communities. However, robotic devices are mainly used by few groups that are highly multidisciplinary. We should not forget the importance that traditional dummies and mirrors had in the study of animal behaviour (Tinbergen 1948). Although these tools have many limitations, they are still commonly used in laboratories, since they are easier to manufacture and to use, if compared to mechatronic devices. Further efforts are needed for improving worldwide the number of collaborations involving researchers in engineering and biology. In addition, the training of novel scientists with a multidisciplinary background would greatly advance this research field, with advantages to both engineering and biology contexts.
6 Conclusions
In our review, recent research about the most significant works on animal–technology interaction, from individual to animal population, is critically discussed. To our mind, this innovative studies can lead to the development of new applications, such as the possibility to merge the two separate fields and create mixed society with cyborgs, revolutionizing the bio-inspired design, as well as leading new biological researches in the fields of animal cognition, evolutionary ecology, conservation biology, as well as parasitology research. In addition, in this review, an accurate analysis on the animal species and abundance of articles for each research topic was carried out, summarizing the current knowledge on bio-hybrid systems and mixed society. Results show that both areas are rapidly growing. Thus, it is crucial to identify their potential future impact on bio-behavioural research. Our review aims to be a concise and timely tool to help researches in the identification and discussion of exciting research challenges in animal behaviour and bio-robotics.
References
Abaid N, Bartolini T, Macrì S, Porfiri M (2012) Zebrafish responds differentially to a robotic fish of varying aspect ratio, tail beat frequency, noise, and color. Behav Brain Res 233(2):545–553
Abaid N, Marras S, Fitzgibbons C, Porfiri M (2013) Modulation of risk-taking behaviour in golden shiners (Notemigonus crysoleucas) using robotic fish. Behav Process 100:9–12
Aktakka EE, Kim H, Najafi K (2011) Energy scavenging from insect flight. J Micromech Microeng 21(9):095016
Ando N, Kanzaki R (2017) Using insects to drive mobile robots—hybrid robots bridge the gap between biological and artificial systems. Arthropod Struct Dev. https://doi.org/10.1016/j.asd.2017.02.003
ASAB/ABS (2004) Guidelines for the treatment of animals in behavioural research and teaching. Anim Behav 99:1–9
Aureli M, Fiorilli F, Porfiri M (2012) Portraits of self-organization in fish schools interacting with robots. Phys D 9:908–920
Ayers J, Clarac F (1978) Neuromuscular strategies underlying different behavioural acts in a multifunctional crustacean leg joint. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 128(1):81–94
Ayers JL Jr, Davis WJ (1977) Neuronal control of locomotion in the lobster, Homarus americanus. J Comp Physiol 115(1):1–27
Bao L, Zheng N, Zhao H, Hao Y, Zheng H, Hu F, Zheng X (2011) Flight control of tethered honeybees using neural electrical stimulation. In: 2011 5th international IEEE/EMBS conference on neural engineering (NER). IEEE, pp 558–561
Bartolini T, Mwaffo V, Showler A, Macrì S, Butail S, Porfiri M (2016) Zebrafish response to 3D printed shoals of conspecifics: the effect of body size. Bioinspir Biomim 11(2):026003
Beer RD, Quinn RD, Chiel HJ, Ritzmann RE (1997) Biologically inspired approaches to robotics: What can we learn from insects? Commun ACM 40(3):30–38
Benelli G, Romano D, Rocchigiani G, Caselli A, Mancianti F, Canale A, Stefanini C (2018) Behavioral asymmetries in ticks–lateralized questing of Ixodes ricinus to a mechatronic apparatus delivering host-borne cues. Acta Trop 178:176–181
Bierbach D, Lukas JA, Bergmann A, Elsner K, Höhne L, Weber C, Weimar N, Arias-Rodriguez L, Mönk HJ, Nguyen H, Romanczuk P, Landgraf T, Krause J (2018a) Insights into the social behavior of surface and cave-dwelling fish (Poecilia mexicana) in light and darkness through the use of a biomimetic robot. Front Robot AI 5:3. https://doi.org/10.3389/frobt.2018.00003
Bierbach D, Landgraf T, Romanczuk P, Lukas J, Nguyen H, Wolf M, Krause J (2018b) Using a robotic fish to investigate individual differences in social responsiveness in the guppy. bioRxiv https://doi.org/10.1101/304501
Bodi M, Möslinger C, Thenius R, Schmickl T (2015) BEECLUST used for exploration tasks in autonomous underwater vehicles. IFAC PapersOnLine 48(1):819–824
Bonnet F, Binder S, de Oliveria M E, Halloy J, Mondada F (2014) A miniature mobile robot developed to be socially integrated with species of small fish. In: 2014 IEEE international conference on robotics and biomimetics (ROBIO). IEEE, pp 747–752
Bonnet F, Crot N, Burnier D, Mondada F (2016a) Design methods for miniature underwater soft robots. In: 2016 6th IEEE international conference on biomedical robotics and biomechatronics (BioRob). IEEE, pp 1365–1370
Bonnet F, Kato Y, Halloy J, Mondada F (2016b) Infiltrating the zebrafish swarm: design, implementation and experimental tests of a miniature robotic fish lure for fish–robot interaction studies. Artif Life Robot 21(3):239–246
Bonnet F, Cazenille L, Séguret A, Gribovskiy A, Collignon B, Halloy J, Mondada F (2017a) Design of a modular robotic system that mimics small fish locomotion and body movements for ethological studies. Int J Adv Robot Syst 14(3):1729881417706628
Bonnet F, Cazenille L, Gribovskiy A, Halloy J, Mondada F (2017b) Multi-robot control and tracking framework for bio-hybrid systems with closed-loop interaction. In: 2017 IEEE international conference on robotics and automation (ICRA). IEEE, pp 4449–4456
Bonnet F, Gribovskiy A, Halloy J, Mondada F (2018) Closed-loop interactions between a shoal of zebrafish and a group of robotic fish in a circular corridor. Swarm Intelligence 12:1–18
Bonsignori G, Stefanini C, Scarfogliero U, Mintchev S, Benelli G, Dario P (2013) The green leafhopper, Cicadella viridis (Hemiptera, Auchenorrhyncha, Cicadellidae), jumps with near-constant acceleration. J Exp Biol 216(7):1270–1279
Bozkurt A, Paul A, Pulla S, Ramkumar A, Blossey B, Ewer J et al (2007) Microprobe microsystem platform inserted during early metamorphosis to actuate insect flight muscle. In: IEEE 20th international conference on micro electro mechanical systems, 2007. MEMS. IEEE, pp 405–408
Bozkurt A, Gilmour R, Stern D, Lal A (2008) MEMS based bioelectronic neuromuscular interfaces for insect cyborg flight control. In: IEEE 21st international conference on micro electro mechanical systems, 2008. MEMS 2008. IEEE, pp 160–163
Bozkurt A, Lal A, Gilmour R (2009) Radio control of insects for biobotic domestication. In: 4th international IEEE/EMBS conference on neural engineering, 2009. NER’09. IEEE, pp 215–218
Bozkurt A, Lobaton E, Sichitiu M (2016) A biobotic distributed sensor network for under-rubble search and rescue. Computer 49(5):38–46
Breazeal C, Dautenhahn K, Kanda T (2016) Social robotics. Springer handbook of robotics. Springer, Cham, pp 1935–1972
Brian Smith C, Martins EP (2006) Display plasticity in response to a robotic lizard: signal matching or song sharing in lizards? Ethology 112(10):955–962
Brown MF, Brown AA (2017) The promise of cyborg intelligence. Learn Behav 45(1):5–6
Butail S, Bartolini T, Porfiri M (2013) Collective response of zebrafish shoals to a free-swimming robotic fish. PLoS ONE 8(10):e76123
Butail S, Polverino G, Phamduy P, Del Sette F, Porfiri M (2014) Influence of robotic shoal size, configuration, and activity on zebrafish behaviour in a free-swimming environment. Behav Brain Res 275:269–280
Butail S, Mwaffo V, Porfiri M (2016) Model-free information-theoretic approach to infer leadership in pairs of zebrafish. Phys Rev E 93(4):042411
Butail S, Mwaffo V, Porfiri M (2017) Inferring leadership in zebrafish pairs: an information-theoretic approach. In: American control conference (ACC), 2017. IEEE, pp 2040–2040
Butler SR, Fernández-Juricic E (2014) European starlings recognize the location of robotic conspecific attention. Biol Lett 10(10):20140665
Cao F, Zhang C, Choo HY, Sato H (2016) Insect–computer hybrid legged robot with user-adjustable speed, step length and walking gait. J R Soc Interface 13(116):20160060
Cazenille L, Chemtob Y, Bonnet F, Gribovskiy A, Mondada F, Bredeche N, Halloy J (2017) Automated calibration of a biomimetic space-dependent model for zebrafish and robot collective behaviour in a structured environment. In: Conference on biomimetic and biohybrid systems. Springer, Cham, pp 107–118
Cazenille L, Collignon B, Chemtob Y, Bonnet F, Gribovskiy A, Mondada F, Bredeche N, Halloy J (2018a) How mimetic should a robotic fish be to socially integrate into zebrafish groups? Bioinspir Biomim 13(2):025001. https://doi.org/10.1088/1748-3190/aa8f6a
Cazenille L, Chemtob Y, Bonnet F, Gribovskiy A, Mondada F, Bredeche N, Halloy J (2018b) How to blend a robot within a group of zebrafish: achieving social acceptance through real-time calibration of a multi-level behavioural model. arXiv preprint arXiv:1805.11371
Cham JG, Bailey SA, Clark JE, Full RJ, Cutkosky MR (2002) Fast and robust: hexapedal robots via shape deposition manufacturing. Int J Robot Res 21(10–11):869–882
Choo HY, Li Y, Cao F, Sato H (2016) Electrical stimulation of coleopteran muscle for initiating flight. PLoS ONE 11(4):e0151808
Cianca V, Bartolini T, Porfiri M, Macrì S (2013) A robotics-based behavioural paradigm to measure anxiety-related responses in zebrafish. PLoS ONE 8(7):e69661
Cole J, Mohammadzadeh F, Bollinger C, Latif T, Bozkurt A, Lobaton E (2017) A study on motion mode identification for cyborg roaches. In: 2017 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, pp 2652–2656
Cord-Cruz G, Ruberto T, Neri D, Porfiri M (2017) Zebrafish response to live predator and biologically-inspired robot in a circular arena. In: SPIE smart structures and materials + nondestructive evaluation and health monitoring. International Society for Optics and Photonics, p 101620P
Couzin ID (2009) Collective cognition in animal groups. Trends in Cognitive Sciences 13(1):36–43
De Margerie E, Lumineau S, Houdelier C, Yris MR (2011) Influence of a mobile robot on the spatial behaviour of quail chicks. Bioinspir Biomim 6(3):034001
De Margerie E, Peris A, Pittet F, Houdelier C, Lumineau S, Richard-Yris MA (2013) Effect of mothering on the spatial exploratory behaviour of quail chicks. Dev Psychobiol 55(3):256–264
Decker M, Dillmann R, Dreier T, Fischer M, Gutmann M, Ott I, genannt Döhmann IS (2011) Service robotics: Do you know your new companion? Framing an interdisciplinary technology assessment. Poiesis Prax 8(1):25–44
Dirafzoon A, Bozkurt A, Lobaton E (2017a) A framework for mapping with biobotic insect networks: from local to global maps. Robot Auton Syst 88:79–96
Dirafzoon A, Bozkurt A, Lobaton E (2017b) Geometric learning and topological inference with biobotic networks. IEEE Trans Signal Inf Process Netw 3(1):200–215
Doan TTV, Sato H (2016) Insect-machine hybrid system: remote radio control of a freely flying beetle (Mercynorrhina torquata). J Vis Exp 115:e54260–e54260
Dodd A (2014) The trouble with insect cyborgs. Soc Anim 22(2):153–173
Donati E, Worm M, Mintchev S, Van Der Wiel M, Benelli G, Von Der Emde G, Stefanini C (2016) Investigation of collective behaviour and electrocommunication in the weakly electric fish, Mormyrus rume, through a biomimetic robotic dummy fish. Bioinspir Biomim 11(6):066009
El Ichi-Ribault S, Alcaraz JP, Boucher F, Boutaud B, Dalmolin R, Boutonnat J, Cinquin P, Zebda A, Martin DK (2018) Remote wireless control of an enzymatic biofuel cell implanted in a rabbit for 2 months. Electrochim Acta 269:360–366
El Khoury R, Ventura R B, Cord-Cruz G, Ruberto T, Porfiri M (2018) Interactive experiments in a robotics-based platform to simulate zebrafish response to a predator. In: Bioinspiration, biomimetics, and bioreplication VIII. International Society for Optics and Photonics, vol 10593. SPIE, p 105930I
Ellington CP (1999) The novel aerodynamics of insect flight: applications to micro-air vehicles. J Exp Biol 202(23):3439–3448
Erickson JC, Herrera M, Bustamante M, Shingiro A, Bowen T (2015) Effective stimulus parameters for directed locomotion in Madagascar hissing cockroach biobot. PLoS ONE 10(8):e0134348
Faria JJ, Dyer JR, Clément RO, Couzin ID, Holt N, Ward AJ et al (2010) A novel method for investigating the collective behaviour of fish: introducing ‘Robofish’. Behav Ecol Sociobiol 64(8):1211–1218
Fernández-Juricic E, Kowalski V (2011) Where does a flock end from an information perspective? A comparative experiment with live and robotic birds. Behav Ecol 22(6):1304–1311
Fernández-Juricic E, Gilak N, Mcdonald JC, Pithia P, Valcarcel A (2006) A dynamic method to study the transmission of social foraging information in flocks using robots. Anim Behav 71(4):901–911
Folkertsma GA, Straatman W, Nijenhuis N, Venner CH, Stramigioli S (2017) Robird: a robotic bird of prey. IEEE Robot Autom Mag 24(3):22–29
Forlizzi J and DiSalvo C (2006) Service robots in the domestic environment: a study of the roomba vacuum in the home. In: Proceedings of the 1st ACM SIGCHI/SIGART conference on human–robot interaction. ACM, pp 258–265
Frohnwieser A, Murray JC, Pike TW, Wilkinson A (2016) Using robots to understand animal cognition. J Exp Anal Behav 105(1):14–22
Fry SN, Rohrseitz N, Straw AD, Dickinson MH (2009) Visual control of flight speed in Drosophila melanogaster. J Exp Biol 212(8):1120–1130
Fujiwara D, Iizuka K, Matsumura Y, Moriyama T, Watanabe R, Enomoto K et al (2012) The effect of mobile robot on group behaviour of animals. J Robot Mechatron 24(6):1071–1079
Garnier S (2011) From ants to robots and back: How robotics can contribute to the study of collective animal behaviour. Bio-inspired self-organizing robotic systems. Springer, Berlin, pp 105–120
Giampalmo SL, Absher BF, Bourne WT, Steves LE, Vodenski VV, O’Donnell P M, Erickson JC (2011) Generation of complex motor patterns in American grasshopper via current-controlled thoracic electrical interfacing. In: 2011 annual international conference of the IEEE engineering in medicine and biology society, EMBC. IEEE, pp 1275–1278
Göth A, Evans CS (2004) Social responses without early experience: Australian brush-turkey chicks use specific visual cues to aggregate with conspecifics. J Exp Biol 207(13):2199–2208
Gribovskiy A, Halloy J, Deneubourg JL, Bleuler H, Mondada F (2010) Towards mixed societies of chickens and robots. In: 2010 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, pp 4722–4728
Gribovskiy A, Mondada F, Deneubourg JL, Cazenille L, Bredeche N, Halloy J (2015) Automated analysis of behavioural variability and filial imprinting of chicks (G. gallus), using autonomous robots. arXiv:1509.01957
Gribovskiy A, Halloy J, Deneubourg JL, Mondada F (2018) Designing a socially integrated mobile robot for ethological research. Robot Auton Syst 103:42–55
Grillner S, Kozlov A, Dario P, Stefanini C, Menciassi A, Lansner A, Kotaleski JH (2007) Modeling a vertebrate motor system: pattern generation, steering and control of body orientation. Prog Brain Res 165:221–234
Griparić K, Haus T, Miklić D, Polić M, Bogdan S (2017) A robotic system for researching social integration in honeybees. PLoS ONE 12(8):e0181977
Halámková L, Halámek J, Bocharova V, Szczupak A, Alfonta L, Katz E (2012) Implanted biofuel cell operating in a living snail. J Am Chem Soc 134(11):5040–5043
Halloy J, Sempo G, Caprari G, Rivault C, Asadpour M, Tâche F et al (2007) Social integration of robots into groups of cockroaches to control self-organized choices. Science 318(5853):1155–1158
Halloy J, Mondada F, Kernbach S, Schmickl T (2013) Towards bio-hybrid systems made of social animals and robots. Biomimetic and biohybrid systems. Springer, Berlin, pp 384–386
Hamann H, Wahby M, Schmickl T, Zahadat P, Hofstadler D, Stoy K, Risi S, Faina A, Veenstra F, Kernbach S, Kuksin I, Kernbach O, Ayres P, Wojtaszek P (2015) Flora robotica-mixed societies of symbiotic robot-plant bio-hybrids. In: 2015 IEEE symposium series on computational intelligence. IEEE, pp 1102–1109
Henderson JV, Nicol CJ, Lines JA, White RP, Wathes CM (2001) Behaviour of domestic ducks exposed to mobile predator stimuli. 1. Flock responses. Br Poult Sci 42(4):433–438
Holzer R, Shimoyama I (1997) Locomotion control of a bio-robotic system via electric stimulation. In: Proceedings of the 1997 IEEE/RSJ international conference on intelligent robots and systems, 1997. IROS’97. IEEE, vol 3, pp 1514–1519
Huai R, Yang J, Wang H, Su X (2009) A new robo-animals navigation method guided by the remote control. In: 2nd international conference on biomedical engineering and informatics, 2009. BMEI’09. IEEE, pp 1–4
Huai RT, Yang JQ, Wang H (2016) The robo-pigeon based on the multiple brain regions synchronization implanted microelectrodes. Bioengineered 7(4):213–218
Huntingford FA (1984) Some ethical issues raised by studies of predation and aggression. Anim Behav 32(1):210–215
Ioannou CC (2017) Swarm intelligence in fish? The difficulty in demonstrating distributed and self-organised collective intelligence in (some) animal groups. Behav Process 141:141–151
Ishii H, Shi Q, Fumino S, Konno S, Kinoshita S, Okabayashi S et al (2013) A novel method to develop an animal model of depression using a small mobile robot. Adv Robot 27(1):61–69
Jacobs JA, Siegford JM (2012) Invited review: the impact of automatic milking systems on dairy cow management, behavior, health, and welfare. J Dairy Sci 95:2227–2247
Jadhav AD, Aimo I, Cohen D, Ledochowitsch P, Maharbiz MM (2012) Cyborg eyes: microfabricated neural interfaces implanted during the development of insect sensory organs produce stable neurorecordings in the adult. In: 2012 IEEE 25th international conference on micro electro mechanical systems (MEMS). IEEE, pp 937–940
Jolly L, Pittet F, Caudal JP, Mouret JB, Houdelier C, Lumineau S, De Margerie E (2016) Animal-to-robot social attachment: initial requisites in a gallinaceous bird. Bioinspir Biomim 11(1):016007
JrJH Long, Schumacher J, Livingston N, Kemp M (2006) Four flippers or two? Tetrapodal swimming with an aquatic robot. Bioinspir Biomim 1(1):20
Katz E (2014) Implantable bioelectronics–editorial introduction. In: Implantable bioelectronics. Wiley, Weinheim. https://doi.org/10.1002/9783527673148.ch1
Katz E, MacVittie K (2013) Implanted biofuel cells operating in vivo–methods, applications and perspectives–feature article. Energy Environ Sci 6(10):2791–2803
Katzschmann RK, DelPreto J, MacCurdy R, Rus D (2018) Exploration of underwater life with an acoustically controlled soft robotic fish. Science Robotics. https://doi.org/10.1126/scirobotics.aar3449
Kawabata K, Aonuma H, Hosoda K, Xue J (2013) Controlled interaction with the cricket based on on-line pose estimation of mobile robot. In: 2013 IEEE international conference on robotics and biomimetics (ROBIO). IEEE, pp 1347–1352
Kawabata K, Aonuma H, Hosoda K, Sugimoto Y, Xue J (2014) Experimental study on robotic interactions to the cricket. In: 2014 IEEE international conference on robotics and biomimetics (ROBIO). IEEE, pp. 949–954
Kim S, Spenko M, Trujillo S, Heyneman B, Santos D, Cutkosky MR (2008) Smooth vertical surface climbing with directional adhesion. IEEE Trans Robot 24(1):65–74
Kim CH, Choi B, Kim DG, Lee S, Jo S, Lee PS (2016) Remote navigation of turtle by controlling instinct behaviour via human brain–computer interface. J Bionic Eng 13(3):491–503
Kim C, Ruberto T, Phamduy P, Porfiri M (2018) Closed-loop control of zebrafish behaviour in three dimensions using a robotic stimulus. Sci Rep 8(1):657
King A (2017) The future of agriculture. Nature 544(7651):S21–S23
Kobayashi N, Yoshida M, Matsumoto N, Uematsu K (2009) Artificial control of swimming in goldfish by brain stimulation: confirmation of the midbrain nuclei as the swimming center. Neurosci Lett 452(1):42–46
Kopman V, Laut J, Polverino G, Porfiri M (2013) Closed-loop control of zebrafish response using a bioinspired robotic-fish in a preference test. J R Soc Interface 10(78):20120540
Krause J, Winfield AF, Deneubourg JL (2011) Interactive robots in experimental biology. Trends Ecol Evol 26(7):369–375
Kubinyi E, Miklósi Á, Kaplan F, Gácsi M, Topál J, Csányi V (2004) Social behaviour of dogs encountering AIBO, an animal-like robot in a neutral and in a feeding situation. Behav Process 65(3):231–239
Kuwana Y, Shimoyama I, Miura H (1995) Steering control of a mobile robot using insect antennae. In: 1995 IEEE/RSJ international conference on intelligent robots and systems 95.’Human Robot Interaction and Cooperative Robots’, Proceedings. IEEE, vol 2, pp 530–535
Kuwana Y, Nagasawa S, Shimoyama I, Kanzaki R (1999) Synthesis of the pheromone-oriented behaviour of silkworm moths by a mobile robot with moth antennae as pheromone sensors. Biosens Bioelectron 14(2):195–202
Ladu F, Bartolini T, Panitz SG, Chiarotti F, Butail S, Macrì S, Porfiri M (2015) Live predators, robots, and computer-animated images elicit differential avoidance responses in zebrafish. Zebrafish 12(3):205–214. https://doi.org/10.1089/zeb.2014.1041
Landgraf T, Oertel M, Rhiel D, Rojas R (2010) A biomimetic honeybee robot for the analysis of the honeybee dance communication system. In: 2010 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, pp 3097–3102
Landgraf T, Rojas R, Nguyen H, Kriegel F, Stettin K (2011) Analysis of the waggle dance motion of honeybees for the design of a biomimetic honeybee robot. PLoS ONE 6(8):e21354
Landgraf T, Oertel M, Kirbach A, Menzel R, Rojas R (2012) Imitation of the honeybee dance communication system by means of a biomimetic robot. In: Conference on biomimetic and biohybrid systems. Springer, Berlin, pp 132–143
Landgraf T, Nguyen H, Forgo S, Schneider J, Schröer J, Krüger C et al (2013) Interactive robotic fish for the analysis of swarm behaviour. In: International conference in swarm intelligence. Springer, Berlin, pp 1–10
Landgraf T, Nguyen H, Schröer J, Szengel A, Clément RJ, Bierbach D, Krause J (2014) Blending in with the shoal: robotic fish swarms for investigating strategies of group formation in guppies. In: Conference on biomimetic and biohybrid systems. Springer International Publishing, pp 178–189
Landgraf T, Bierbach D, Nguyen H, Muggelberg N, Romanczuk P, Krause J (2016) RoboFish: increased acceptance of interactive robotic fish with realistic eyes and natural motion patterns by live Trinidadian guppies. Bioinspir Biomim 11(1):015001
Laschi C, Mazzolai B, Patanè F, Mattoli V, Dario P, Ishii H et al (2006) Design and development of a legged rat robot for studying animal–robot interaction. In: The first IEEE/RAS-EMBS international conference on biomedical robotics and biomechatronics, 2006. BioRob 2006. IEEE, pp 631–636
Latif T, Whitmire E, Novak T, Bozkurt A (2014) Towards fenceless boundaries for solar powered insect biobots. In: 2014 36th annual international conference of the IEEE engineering in medicine and biology society (EMBC). IEEE, pp 1670–1673
Latif T, Whitmire E, Novak T, Bozkurt A (2016) Sound localization sensors for search and rescue biobots. IEEE Sens J 16(10):3444–3453
Li G, Zhang D (2016) Brain–computer interface controlled cyborg: establishing a functional information transfer pathway from human brain to cockroach brain. PLoS ONE 11(3):e0150667
Li G, Zhang D (2017) Brain–computer interface controlling cyborg: a functional brain-to-brain interface between human and cockroach. Brain–computer interface research. Springer International Publishing, Cham, pp 71–79
Li Y, Wu J, Sato H (2018) Feedback control-based navigation of a flying insect-machine hybrid robot. Soft Robotics. https://doi.org/10.1089/soro.2017.0118
Lin JY, Yu CN, Jia J, Zhang SM, Wang YW, Chen WD, Zheng XX (2010) Using dlPAG-evoked immobile behaviour in animal–robotics navigation. In: 2010 5th international conference on computer science and education (ICCSE). IEEE, pp 1295–1298
Lin HT, Leisk GG, Trimmer B (2011) GoQBot: a caterpillar-inspired soft-bodied rolling robot. Bioinspir Biomim 6(2):026007
Lockwood JA (1988) Not to harm a fly: our ethical obligations to insects. Between Species 4(3):12
Lorenz K (1935) Der kumpan in der umwelt des vogels. J Ornithol 83(3):289–413. [Translated in K. Lorenz. 1970. Studies in Animal and Human Behaviour. Harvard University Press, Cambridge.]
Macrì S, Neri D, Ruberto T, Mwaffo V, Butail S, Porfiri M (2017) Three-dimensional scoring of zebrafish behaviour unveils biological phenomena hidden by two-dimensional analyses. Sci Rep 7(1):1962
MacVittie K, Halámek J, Halámková L, Southcott M, Jemison WD, Lobel R, Katz E (2013) From “cyborg” lobsters to a pacemaker powered by implantable biofuel cells. Energy Environ Sci 6(1):81–86
Manfredi L, Assaf T, Mintchev S, Marrazza S, Capantini L, Orofino S et al (2013) A bioinspired autonomous swimming robot as a tool for studying goal-directed locomotion. Biol Cybern 107(5):513–527
Marras S, Porfiri M (2012) Fish and robots swimming together: attraction towards the robot demands biomimetic locomotion. J R Soc Interface 9(73):1856–1868
Martinez D, Arhidi L, Demondion E, Masson JB, Lucas P (2014) Using insect electroantennogram sensors on autonomous robots for olfactory searches. J Vis Exp 90:e51704
Martins EP, Ord TJ, Davenport SW (2005) Combining motions into complex displays: playbacks with a robotic lizard. Behav Ecol Sociobiol 58(4):351–360
Mehta D, Altan E, Chandak R, Raman B, Chakrabartty S (2017) Behaving cyborg locusts for standoff chemical sensing. In: 2017 IEEE international symposium on circuits and systems (ISCAS). IEEE, pp 1–4
Michelsen A, Andersen BB, Storm J, Kirchner WH, Lindauer M (1992) How honeybees perceive communication dances, studied by means of a mechanical model. Behav Ecol Sociobiol 30(3–4):143–150
Mitri S, Wischmann S, Floreano D, Keller L (2013) Using robots to understand social behaviour. Biolo Rev 88(1):31–39
Mondada F, Martinoli A, Correll N, Gribovskiy A, Halloy JI, Siegwart R, Deneubourg JL (2013) A general methodology for the control of mixed natural-artificial societies. Pan Stanford Publishing, Stanford, pp 547–586
Myrick AJ, Baker TC (2010) Locating a compact odor source using a four-channel insect electroantennogram sensor. Bioinspir Biomim 6(1):016002
Narins PM, Grabul DS, Soma KK, Gaucher P, Hödl W (2005) Cross-modal integration in a dart-poison frog. Proc Natl Acad Sci USA 102(7):2425–2429
Park SJ, Gazzola M, Park KS, Park S, Di Santo V, Blevins EL et al (2016) Phototactic guidance of a tissue-engineered soft-robotic ray. Science 353(6295):158–162
Partan SR, Larco CP, Owens MJ (2009) Wild tree squirrels respond with multisensory enhancement to conspecific robot alarm behaviour. Anim Behav 77(5):1127–1135
Partan SR, Fulmer AG, Gounard MA, Redmond JE (2010) Multimodal alarm behaviour in urban and rural gray squirrels studied by means of observation and a mechanical robot. Curr Zool 56(3):313–326
Partan SR, Otovic P, Price VL, Brown SE (2011) Assessing display variability in wild brown anoles Anolis sagrei using a mechanical lizard model. Curr Zool 57(2):140–152
Pastell M, Takko H, Gröhn H, Hautala M, Poikalainen V, Praks J, Veermäe I, Kujala M, Ahokas J (2006) Assessing cows’ welfare: weighing the cow in a milking robot. Biosyst Eng 93(1):81–87
Patanè F, Mattoli V, Laschi C, Mazzolai B, Dario P, Ishii H, Takanishi A (2007) Biomechatronic design and development of a legged rat robot. In: IEEE international conference on robotics and biomimetics, 2007. ROBIO 2007. IEEE, pp 847–852
Patricelli GL, Krakauer AH (2009) Tactical allocation of effort among multiple signals in sage grouse: an experiment with a robotic female. Behav Ecol 21(1):97–106
Patricelli GL, Uy JAC, Walsh G, Borgia G (2002) Sexual selection: male displays adjusted to female’s response. Nature 415(6869):279–280
Patricelli GL, Coleman SW, Borgia G (2006) Male satin bowerbirds, Ptilonorhynchus violaceus, adjust their display intensity in response to female startling: an experiment with robotic females. Anim Behav 71(1):49–59
Peng Y, Wu Y, Yang Y, Huang R, Wu C, Qi X et al (2011) Study on the control of biological behaviour on carp induced by electrophysiological stimulation in the corpus cerebelli. In: 2011 international conference on electronic and mechanical engineering and information technology (EMEIT). IEEE, vol 1, pp 502–505
Pennisi P, Tonacci A, Tartarisco G, Billeci L, Ruta L, Gangemi S, Pioggia G (2016) Autism and social robotics: a systematic review. Autism Res 9(2):165–183
Pfeifer R, Bongard J (2006) How the body shapes the way we think: a new view of intelligence. MIT press, Cambridge
Phamduy P, Polverino G, Fuller RC, Porfiri M (2014) Fish and robot dancing together: bluefin killifish females respond differently to the courtship of a robot with varying color morphs. Bioinspir Biomim 9(3):036021
Polverino G, Porfiri M (2013) Mosquitofish (Gambusia affinis) responds differentially to a robotic fish of varying swimming depth and aspect ratio. Behav Brain Res 250:133–138
Polverino G, Abaid N, Kopman V, Macrì S, Porfiri M (2012) Zebrafish response to robotic fish: preference experiments on isolated individuals and small shoals. Bioinspir Biomim 7(3):036019
Polverino G, Phamduy P, Porfiri M (2013) Fish and robots swimming together in a water tunnel: robot color and tail-beat frequency influence fish behaviour. PLoS ONE 8(10):e77589
Poon KC, Tan DCL, Li Y, Cao F, Doan TTV, Sato H (2016) Cyborg insect: insect computer hybrid robot. In: Meeting abstracts. Electrochemical Society, vol 44. ECS, p 3221
Rashid MT, Frasca M, Ali AA, Ali RS, Fortuna L, Xibilia MG (2012) Artemia swarm dynamics and path tracking. Nonlinear Dyn 68(4):555–563
Reaney LT, Sims RA, Sims SW, Jennions MD, Backwell PR (2008) Experiments with robots explain synchronized courtship in fiddler crabs. Curr Biol 18(2):R62–R63
Romano D, Benelli G, Donati E, Remorini D, Canale A, Stefanini C (2017a) Multiple cues produced by a robotic fish modulate aggressive behaviour in Siamese fighting fishes. Sci Rep. https://doi.org/10.1038/s41598-017-04840-0
Romano D, Benelli G, Stefanini C (2017b) Escape and surveillance asymmetries in locusts exposed to a Guinea fowl-mimicking robot predator. Sci Rep. https://doi.org/10.1038/s41598-017-12941-z
Romano D, Stefanini C, Canale A, Benelli G (2018) Artificial blood feeders for mosquito and ticks—Where from, where to? Acta Trop 183:43–56
Ruberto T, Mwaffo V, Singh S, Neri D, Porfiri M (2016) Zebrafish response to a robotic replica in three dimensions. R Soc Open Sci 3(10):160505
Ruberto T, Polverino G, Porfiri M (2017) How different is a 3D-printed replica from a conspecific in the eyes of a zebrafish? J Exp Anal Behav 107(2):279–293
Ruhland K, Peters CE, Andrist S, Badler JB, Badler NI, Gleicher M, Mutlu B, McDonnell R (2015) A review of eye gaze in virtual agents, social robotics and hci: behaviour generation, user interaction and perception. In: Computer graphics forum, vol 34, No. 6. Wiley, pp 299–326
Rundus AS, Owings DH, Joshi SS, Chinn E, Giannini N (2007) Ground squirrels use an infrared signal to deter rattlesnake predation. Proc Natl Acad Sci 104(36):14372–14376
Sanchez CJ, Chiu CW, Zhou Y, González JM, Vinson SB, Liang H (2015) Locomotion control of hybrid cockroach robots. J R Soc Interface 12(105):20141363
Sato H, Maharbiz MM (2010) Recent developments in the remote radio control of insect flight. Front Neurosci 4:199
Sato H, Berry CW, Casey BE, Lavella G, Yao Y, VandenBrooks JM, Maharbiz MM (2008) A cyborg beetle: insect flight control through an implantable, tetherless microsystem. In: IEEE 21st international conference on micro electro mechanical systems, 2008. MEMS 2008. IEEE, pp 164–167
Sato H, Berry CW, Peeri Y, Baghoomian E, Casey BE, Lavella G et al (2009) Remote radio control of insect flight. Front Integr Neurosci 3:24
Sato H, Doan TTV, Kolev S, Huynh NA, Zhang C, Massey TL et al (2015) Deciphering the role of a coleopteran steering muscle via free flight stimulation. Curr Biol 25(6):798–803
Savard JF, Keagy J, Borgia G (2011) Blue, not UV, plumage color is important in satin bowerbird Ptilonorhynchus violaceus display. J Avian Biol 42(1):80–84
Schmickl T, Thenius R, Moslinger C, Timmis J, Tyrrell A, Read M, Hilder J, Halloy J, Campo A, Stefanini C, Manfredi L, Orofino S, Kernbach S, Dipper T, Sutantyo D (2011) CoCoRo–the self-aware underwater swarm. In: 2011 fifth IEEE conference on self-adaptive and self-organizing systems workshops (SASOW). IEEE, pp 120–126
Schwefel J, Ritzmann RE, Lee IN, Pollack A, Weeman W, Garverick S et al (2015) Wireless communication by an autonomous self-powered cyborg insect. J Electrochem Soc 161(13):H3113–H3116
Seo J, Choi GJ, Park S, Lee J, Baek C, Jang J et al (2017) Wireless navigation of pigeons using polymer-based fully implantable stimulator: a pilot study using depth electrodes. In: 2017 39th annual international conference of the IEEE engineering in medicine and biology society (EMBC). IEEE, pp 917–920
Sherman PW, Lacey EA, Reeve HK, Keller L (1995) Forum the eusociality continuum. Behav Ecol 6(1):102–108
Shi Q, Miyagishima S, Konno S, Fumino S, Ishii H, Takanishii A et al (2010) Development of the hybrid wheel-legged mobile robot WR-3 designed to interact with rats. In: 2010 3rd IEEE RAS and EMBS international conference on biomedical robotics and biomechatronics (BioRob). IEEE, pp 887–892
Shi Q, Ishii H, Kinoshita S, Takanishi A, Okabayashi S, Iida N et al (2013) Modulation of rat behaviour by using a rat-like robot. Bioinspir Biomim 8(4):046002
Shi Q, Ishii H, Tanaka K, Sugahara Y, Takanishi A, Okabayashi S et al (2015) Behaviour modulation of rats to a robotic rat in multi-rat interaction. Bioinspir Biomim 10(5):056011
Shoji K, Morishima K, Akiyama Y, Nakamura N, Ohno H (2016) Autonomous environmental monitoring by self-powered biohybrid robot. In: 2016 IEEE international conference on mechatronics and automation (ICMA). IEEE, pp 629–634
Spinello C, Macrì S, Porfiri M (2013) Acute ethanol administration affects zebrafish preference for a biologically inspired robot. Alcohol 47(5):391–398
Stefanini C, Orofino S, Manfredi L, Mintchev S, Marrazza S, Assaf T et al (2012) A novel autonomous, bioinspired swimming robot developed by neuroscientists and bioengineers. Bioinspir Biomim 7(2):025001
Stojnić A (2017) Only cyborgs and cockroaches. Perform Res 22(2):123–128
Stowers JR, Hofbauer M, Bastien R, Griessner J, Higgins P, Farooqui S, Fischer RM, Nowikovsky K, Haubensak W, Couzin ID, Tessmar-Raible K, Straw AD (2017) Virtual reality for freely moving animals. Nat Methods 14(10):995
Swain DT, Couzin ID, Leonard NE (2012) Real-time feedback-controlled robotic fish for behavioural experiments with fish schools. Proc IEEE 100(1):150–163
Takanishi A, Aoki T, Ito M, Ohkawa Y, Yamaguchi JI (1998) Interaction between creature and robot: development of an experiment system for rat and rat robot interaction. In: 1998 IEEE/RSJ international conference on intelligent robots and systems, 1998. Proceedings. IEEE, vol 3, pp 1975–1980
Talwar SK, Xu S, Hawley ES, Weiss SA, Moxon KA, Chapin JK (2002) Behavioural neuroscience: rat navigation guided by remote control. Nature 417(6884):37–38
Taylor RC, Klein BA, Stein J, Ryan MJ (2008) Faux frogs: multimodal signalling and the value of robotics in animal behaviour. Anim Behav 76(3):1089–1097
Tinbergen N (1948) Social releasers and the experimental method required for their study. Wilson Bull 60:6–51
Tsang WM, Aldworth Z, Stone A, Permar A, Levine R, Hildebrand et al (2008) Insect flight control by neural stimulation of pupae-implanted flexible multisite electrodes. In: The proceeding of µ TAS, pp 1922–1924
Tye M (2016) Are insects sentient? Animal sentience: an interdisciplinary. J Anim Feel 1(9):5
Uematsu K, Todo T (1997) Identification of the midbrain locomotor nuclei and their descending pathways in the teleost carp, Cyprinus carpio. Brain Res 773(1):1–7
Vaughan R, Sumpter N, Henderson J, Frost A, Cameron S (2000) Experiments in automatic flock control. Robot Auton Syst 31(1):109–117
Wang H, Huai RT, Yang JQ, Su XC (2013) Review of research progress in biorobot. In: Advanced materials research. Trans Tech Publications, vol 655. Springer, pp 1061–1065
Wang Y, Lu M, Wu Z, Tian L, Xu K, Zheng X, Pan G (2015) Visual cue-guided rat cyborg for automatic navigation [research frontier]. IEEE Comput Intell Mag 10(2):42–52
Wang S, Shen L, Liu X, Liao H (2016) A wearable backpack chip for honeybee biorobot. In: Semiconductor technology international conference (CSTIC), 2016 China. IEEE, pp 1–3
Wang Y, Lu M, Wu Z, Zheng X, Pan G (2017) Visual cue-guided rat cyborg. In: Brain–computer interface research. Springer, Cham, pp 65–78
Wang H, Yang J, Lv C, Huai R, Li Y (2018) Intercollicular nucleus electric stimulation encoded “walk forward” commands in pigeons. Brill. https://doi.org/10.1163/15707563-17000053
Ward AJ, Sumpter DJ, Couzin ID, Hart PJ, Krause J (2008) Quorum decision-making facilitates information transfer in fish shoals. Proc Natl Acad Sci 105(19):6948–6953
Webb B (2000) What does robotics offer animal behaviour? Anim Behav 60(5):545–558
Wenbo W, Ce G, Jiurong S, Zhendong D (2009) Locomotion elicited by electrical stimulation in the midbrain of the lizard gekko gecko. Intelligent unmanned systems: theory and applications. Springer, Berlin, pp 145–153
Worm M, Landgraf T, Nguyen H, von der Emde G (2014) Electro-communicating dummy fish initiate group behaviour in the weakly electric fish Mormyrus rume. In: Duff A, Lepora NF, Mura A, Prescott TJ, Verschure PFMJ (eds) Biomimetic and biohybrid systems LNAI, vol 8608, pp 446–448
Worm M, Landgraf T, Prume J, Nguyen H, Kirschbaum F, von der Emde G (2018) Evidence for mutual allocation of social attention through interactive signaling in a mormyrid weakly electric fish. Proc Natl Acad Sci 24:245. https://doi.org/10.1073/pnas.1801283115
Wu Z, Pan G, Zheng N (2013) Cyborg intelligence. IEEE Intell Syst 28(5):31–33
Wu Z, Yang Y, Xia B, Zhang Z, Pan G (2014) Speech interaction with a rat. Chin Sci Bull 59(28):3579–3584
Wu Z, Zhou Y, Shi Z, Zhang C, Li G, Zheng X, Zheng N, Pan G (2016) Cyborg intelligence: recent progress and future directions. IEEE Intell Syst 31(6):44–50
Xu S, Talwar SK et al (2004) A multi-channel telemetry system for brain microstimulation in freely roaming animals. J Neurosci Methods 133(1–2):57–63
Xu K, Zhang J, Zhou H, Lee JCT, Zheng X (2016) A novel turning behaviour control method for rat-robot through the stimulation of ventral posteromedial thalamic nucleus. Behav Brain Res 298:150–157
Yang J, Huai R, Wang H, Lv C, Su X (2015) A robo-pigeon based on an innovative multi-mode telestimulation system. Biomed Mater Eng 26(s1):357–363
Yu Y, Pan G, Gong Y, Xu K, Zheng N, Hua W et al (2016a) Intelligence-augmented rat cyborgs in maze solving. PLoS ONE 11(2):e0147754
Yu Y, Wu Z, Xu K, Gong Y, Zheng N, Zheng X, Pan G (2016b) Automatic training of rat cyborgs for navigation. Comput Intell Neurosci. https://doi.org/10.1155/2016/6459251
Zahadat P, Hahshold S, Thenius R, Crailsheim K, Schmickl T (2015) From honeybees to robots and back: division of labour based on partitioning social inhibition. Bioinspir Biomim 10(6):066005
Zhang C, Cao F, Li Y, Sato H (2016) Fuzzy-controlled living insect legged actuator. Sens Actuators A Phys 242:182–194
Zheng N, Jin M, Hong H, Huang L, Gu Z, Li H (2017) Real-time and precise insect flight control system based on virtual reality. Electron Lett 53(6):387–389
Funding
This work was funded by the EU project subCULTron (submarine cultures perform long-term robotic exploration of unconventional environmental niches) number 640967. The funder had no role in the study design, data collection and analysis, decision to publish or preparation of the manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Jean-Marc Fellous.
Rights and permissions
About this article
Cite this article
Romano, D., Donati, E., Benelli, G. et al. A review on animal–robot interaction: from bio-hybrid organisms to mixed societies. Biol Cybern 113, 201–225 (2019). https://doi.org/10.1007/s00422-018-0787-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00422-018-0787-5