Path planning of mobile robot using neural network

In this paper, the authors present an effective method to achieve both obstacle-avoidance and target-tracking for an autonomous mobile robot in an indoor environment. They employ a wall following algorithm using neural network pattern recognition to avoid obstacles. An autonomous mobile robot reaches a given goal target by tracking algorithm. In case obstacles are detected by sonar sensors, an autonomous mobile robot avoids collision with obstacles by wall following algorithm. They propose a simple making method to avoid being trapped in a local minima which was a serious problem in local path planning. Simulation results using mobile robot demonstrate that the proposed algorithms are well suited to the obstacle-avoidance using wall-following and the path planning task for target-tracking.