Multi-Agent System Anti-Interference Algorithm Based on Deep Reinforcement Learning

Abstract

Listen

Translate article

As multi-agent systems are increasingly applied in various environments, achieving stable control under communication interference has become a key challenge. This paper proposes a stability control algorithm based on a Markov model and reinforcement learning, which enables the multi-agent system to achieve its goals through a reward and punishment mechanism under interference. The results demonstrate that the algorithm effectively handles communication interference, converges well, and ultimately completes control tasks, showing promising application prospects for the proposed solution.