Robot navigation in dynamic environments using fuzzy logic and trajectory prediction table

Abstract

Robot navigation method is to drive a robot in the unknown environment to a designate position, the target, without collision. In this paper, we propose a robot navigation approach for dynamic environment with static and moving obstacles. The proposed navigation method consists of static navigation method and dynamic path planning. The static navigation drives the robot to avoid the static obstacles by using fuzzy logic controller, which contains four input and two output variables. If the robot detects moving obstacles, the robot can recognize the velocity and moving direction of each obstacle and generate the corresponding Trajectory Prediction Table to predict the obstacles' future trajectory. If the table reveals that the robot will collide with an obstacle, the dynamic path planning will find a new path to avoid the obstacle by waiting strategy or detouring strategy. The detouring strategy is to detour the obstacle before colliding with the obstacle by using the method similar to A <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> algorithm. The simulation results have demonstrated the proposed method, and the robot can reach the target without collision successfully.