Autonomous navigation system based on collision danger-degree for unmanned ground vehicle

This paper proposes an autonomous navigation system for an unmanned ground vehicle (UGV) using a collision danger-degree on outdoor environment. The proposed autonomous navigation system mainly consists of three kinds of control module, a self-tuning fuzzy (STF) path tracking control module, an obstacle avoidance control module, and a two degree-of-freedom speed control module. The STF path tracking control module estimates a steering angle for reference path tracking. On the other hand, the obstacle avoidance control module estimates a steering angle for realizing effective avoidance action by considering the reference path tracking. To solve a tradeoff problem between these control modules, estimated two steering angles are weighted by the collision danger-degree, and final steering angle command is determined. Some experimental results using the developed UGV demonstrate the usefulness of the proposed autonomous navigation system.