Abstract
In the potential field approach to path planning, artificial potential fields are utilized for path planning and obstacle avoidance. However, the development of such fields is computationally intensive. This paper describes a neural network called the Wave expansion neural network (WENN) that is capable of developing artificial potential fields over discretized representations of a moving object's environment. WENN dynamics are analyzed and techniques for efficient emulation in conventional processing systems are described. The process of artificial potential field development by the WENN is examined and the methodologies which use these fields for path planning are briefly discussed.