Visual control of musculoskeletal arm using multiple reference frame including simple coordinate transformation

We propose visual control of musculoskeletal arm using multiple reference frame including simple coordinate transformation. In this system, visual feedback with time delay and position feedback with a proprioception sensor are integrated. The weight for these two feedback systems can be adjusted depending on situations. From simulation results, we can find the proprioception sensor feedback and visual feedback are effective for initial quick motion and final positioning, respectively. By adjusting weights for these two systems, the task is performed.

[1]  Noriaki Maru,et al.  Positioning control of the arm of the humanoid robot by linear visual servoing , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[2]  Yoichi Hori,et al.  Leg space observer on biarticular actuated two-link manipulator for realizing spring loaded inverted pendulum model , 2012, 2012 12th IEEE International Workshop on Advanced Motion Control (AMC).

[3]  Sandra Hirche,et al.  A Framework for Information Distribution, Task Execution and Decision Making in Multi-Robot Systems , 2010, IEICE Trans. Inf. Syst..

[4]  Philip N. Sabes,et al.  Sensory transformations and the use of multiple reference frames for reach planning , 2009, Nature Neuroscience.

[5]  Pattie Maes,et al.  Postural primitives: Interactive Behavior for a Humanoid Robot Arm , 1996 .

[6]  Michael Sorg,et al.  What can be learned from human reach-to-grasp movements for the design of robotic hand-eye systems? , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).