Obstacle avoidance for kinematically redundant manipulators using the deterministic annealing neural network

With the wide deployment of redundant manipulators in complex working environments, obstacle avoidance emerges as an important issue to be addressed in robot motion planning. In this paper, a new obstacle avoidance scheme is presented for redundant manipulators. In this scheme, obstacle avoidance is mathematically formulated as a time-varying linearly constrained quadratic programming problem. To solve this problem effectively in real time, the deterministic annealing neural network is adopted, which has the property of low structural complexity. The effectiveness of this scheme and the real time solution capability of the deterministic neural network is demonstrated by using a simulation example based on the Mitsubishi PA10-7C manipulator.