ON THE PATH PLANNING PROBLEM IN THE VICINITY OF OBSTACLES

Abstract

This paper presents a new approach to motion planning in the neighborhood of obstacles. The technique presented here, the configuration space vector path planner CSVPP , generates a collision-free path for a robot amongst unknown arbitrarily shaped obstacles. The CSVPP algorithm utilizes discrete vectors in the configuration space of the robot to generate a path between any two points in the robot's dynamic time-varying workspace. The calculation of the robot's path assumes interpolated joint control, and provides a computational speed that enables the algorithm to be implemented in real time. A number of simulations are provided for several varying environments.