Adaptive formation control of robot swarms using optimized potential field method

Abstract

Robot Swarm is widely used in many applications such as forest fire detection, Search and rescue missions. Swarm of Robots is supposed to move together without collision and avoid obstacles while performing its target task. Therefore, the formation control of robot swarm is required to achieve the swarm robot target. In this paper, we present an adaptive formation control algorithm for robot swarm based on the Potential Field Method. The algorithm has three tasks, to keep robot swarm in a particular formation, avoid collision with obstacles in the environment and track a certain trajectory. An artificial neural network is employed to improve the performance of the algorithm. The network optimizes the weights in each layer then updates the potential Field parameters. A simulation via MATLAB is implemented to verify the proposed adaptive formation control algorithm. The results show that robot swarm takes less time to maintain formation, less time to track a trajectory and less time to reform again after avoiding an obstacle compared with the time in [1].