ROBOT PATH PLANNING BASED ON MODIFIED GREY RELATIONAL ANALYSIS

Abstract

In this paper, a simple searching approach integrated the nonlinear programming problem and the modified grey relational analysis is proposed to find the near-optimal and collision-free path for a robot, from an initial position to a goal position, in which the robot can be acted in the known or unknown workspace with multiple circular obstacles. The proposed find-path procedure consists of at least one trial. Each trial includes three main stages, i.e., a forward search stage, a backward search stage, and an inference stage. After all trials are completed, a decision-making stage is introduced to determine the near-optimal and collision-free path which is guaranteed to reach the goal. Besides, the presented method is applicable for the on-line applications and, furthermore, can solve the local minimal problems. Simulation results for circular obstacles demonstrate the performances of the proposed approach and its potential as an on-line path planner.