| Abstract: | Planning methods for effective manipulation of single or multiple redundant arm systems must take account of DOF, the task, constraints, and joint drifts. Here, a new approach to redundancy resolution and obstacle avoidance for cooperative robot arms is proposed. In this development, a relative Jacobian and a relative dexterity measure for cooperative robot arms are derived. A nonlinear programming method is used to optimize the relative dexterity while satisfying cooperative task requirements, limits on joint angles, and obstacle avoidance. With this approach, it is not necessary to balance the weightings between the cost term to be optimized and the penalty from constraints. Configuration jumps over obstacles are avoided. Further, since globally optimal joint configurations are produced, drifts in joint configurations will be absent from the resulting configurations. This article includes several illustrative examples to demonstrate the effectiveness and usefulness of this approach. Results have indicated the benefits of both the relative dexterity and the sum of individual arm dexterities in planning of cooperative tasks. ©1999 John Wiley & Sons, Inc. |
