Abstract
With the development of .exible and highly integrated dexterous gripping devices (e.g. fig.6.1) the research results on grasp and manipulation planning can be applied to realize systems with autonomous grasping capabilities. The need for efficient methods to perform grasp analysis and planning for real world applications, therefore increases.
The so far proposed grasp planning and grasp evaluation methods made big contributions on the understanding of the structure of the grasping problem. However, not too many grasp planning systems are known that are able to cope with the constraints of planning grasps in reality, like short planning times, complex and incomplete object models and physical relevance of the planning results.
In this chapter we summarize different grasp qualification methods and out line a physically well motivated grasp quality measure using wrench spaces. We present an algorithm, based on a physically well motivated grasp quality measure to qualify a given grasp with negligible approximation errors. Justified by statistic evaluations for some real world objects based on this grasp quality measure, we suggest a very effective generate and test grasp planner architecture. The proposed planner allows for planning high quality grasps for realistic object models extremely fast a nd thus can be used for online autonomous grasping systems.
Access provided by Autonomous University of Puebla. Download to read the full chapter text
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Author information
Authors and Affiliations
Editor information
Rights and permissions
About this chapter
Cite this chapter
Borst, C., Fischer, M., Hirzinger, G. 6 Efficient and Precise Grasp Planning for Real World Objects. In: Barbagli, F., Prattichizzo, D., Salisbury, K. (eds) Multi-point Interaction with Real and Virtual Objects. Springer Tracts in Advanced Robotics, vol 18. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11429555_6
Download citation
DOI: https://doi.org/10.1007/11429555_6
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-26036-3
Online ISBN: 978-3-540-31503-2
eBook Packages: EngineeringEngineering (R0)