A heuristic obstacle avoidance algorithm using vanishing point and obstacle angle

Abstract

Although there exists a class of algorithms for coping with unknown obstacles in mobile robot navigation, most of them produce rather conservative paths because the varying density of obstacles is not directly considered in the real-time motion planning stage. In this paper, we develop a heuristic obstacle avoidance method in terms of the vanishing point and obstacle angle (VP---OA) to compromise through an adjustable weighting factor between the lane tracking and the obstacle avoidance performance depending on the frequency of emerging obstacles. The suggested algorithm has the advantage of generating smooth local paths close to a human's car driving. Comparison simulations and experiments with other popular algorithms validate the effectiveness of the proposed scheme.