Obstacle avoidance of redundant manipulators on visual servo system using estimated image features

Proposes a visual servo system using estimated image features and its application to obstacle avoidance. A major problem in a visual servo system arises because of the longer sampling period of the image processing compared to that of the manipulator control. To compensate for the delay time of visual information, a method is introduced using estimated image features. In obstacle avoidance of manipulators, not only the position of the end-effector but also the configuration of arms must be controlled. A criterion function using image features is adopted so that the redundant manipulators can avoid obstacles using redundancy. Here, a strategy which enables less computational time and easy programming by using the image Jacobian matrix is proposed. Moreover, the estimated image features can prevent manipulators from colliding with obstacles while the image is being processed. Experimental results show that obstacle avoidance can be realized by using the proposed method