No Detail Found!

Vision based motion planning of humanoid robot arm for grasping task.

Author(s)
Altaf Hussain Rajpar Department of Mechanical Engineering, Quaid-e-Awam University of Engineering, Science & Technology, Shaheed Benazir Bhuttoabad, Nawabshah, Pakistan
Abdul Latif Manganhar Department of Mechanical Engineering, Quaid-e-Awam University of Engineering, Science & Technology, Shaheed Benazir Bhuttoabad, Nawabshah, Pakistan
Abdul Fattah Chandio Department of Electrical Engineering, Quaid-e-Awam University of Engineering, Science & Technology, Shaheed Benazir Bhuttoabad, Nawabshah, Pakistan
Abstract
This paper presents the robust and efficient motion control approach for a redundant humanoid arm of BHR-02 (Beijing Humanoid Robot-02). This work is based on the numerical algorithm for computing the Inverse Kinematics of redundant arm and stereo based vision system to detect the pose and position of target object. Forward recursion formula with backward cycle computation method is used to reach in the vicinity of target object. Forward recursion formula with backward cycle computation method is used to reach in the vicinity of target object. Then BFS (Broyden Fletecher Shanno) method is then used to grasp object in real-time. Vision is a very important non contact sensor for humanoid robots. A method combining visual feed forward and visual feedback is proposed. Visual feed forward facilitates the reach in the vicinity of the object and reduces the manipulation time. Visual feedback increases the robustness by compensating the weak calibration errors. Effectiveness of the proposed strategy is proved through experiments.
Publication Details
Page(s) 119-128
DOI DOI not available
Published Journal: Mehran University Research Journal of Engineering and Technology, Volume: 28, Issue: 2, Year: 2009

Request the full text from PASTIC

0

Citations

0

Downloads

85

Views