Modelling and simulating obstacle avoidance for a wheeled mobile robot

Abstract

This research constructed an artificial environment by simulating the sensors’ information and the wheeled robot’s kinematics characteristics. In this research, the robot’s performance on tracking and obstacle avoidance is tested in four scenarios which are the combinations of the stationary/ dynamic target and the stationary/ dynamic obstacles. The results show that the robot can perform well in tracking and obstacle avoidance, but the path will not be the optimal if related parameters are not well selected. As for the relevant algorithms that are used in the research, a proposed segmented strategy based on STANLEY is used for tracking, and the artificial potential field method is used in obstacle avoidance tasks. In detail, a pure control method based on positional errors called “Staticfit” is merged with STANLEY being used at different stages. Experiments in this research presents the feasibility and effectiveness of using STANLEY and artificial potential field method. However, there are still some aspects that can be improved in future works.