Artificial Potential Field Method with Predicted State and Input Threshold for Multi-agent System

Dengxiu Yu, Zhize Du, Zhen Wang

Abstract: In this article, we consider the multi-agent control based on the artificial potential field (APF) method with predicted state and input threshold. APF is a very practical and efficient method for multi-agent control. However, the accuracy of APF is susceptible to communication delay. Hence, we introduce the predictive state model to reduce the impact of this delay when the agent is avoiding collisions and maintaining formation. Meanwhile, the input threshold is applied to ensure the safety of the system. The introduction of the predicted state and the input threshold leads to the failure of traditional APF. Therefore, we propose a new controller based on the improved APF. Then, the Lyapunov stability of the designed controller is analyzed. Simulation results show the effectiveness of the proposed controller and its superiority over the original method.