Collision avoidance algorithm based on fuzzy expert systems for multi-path planning

Abstract

In the multi-path planning, every autonomous vehicle normally receives her path from the server and sends her position to the server. If server estimates collision between two vehicles, then the path should be re-planned by an algorithm in the server. Path could be compensated by fuzzy expert systems (FESs) that is designed using heuristic method for collision avoidance in multi-path planning. The server calculates the assistance via point and send to each vehicle. The algorithm was evaluated it's stability by simulation test, and then experimented by real autonomous vehicle. The experimental result proved this collision avoidance algorithm is good for multi-path planning.