An Adaptive Dynamic Programming-Based Control Framework for Multi-Agent Systems with BP Neural Network Integration

This paper presents a novel adaptive control framework for multi-agent systems (MAS), integrating Adaptive Dynamic Programming (ADP) and Backpropagation (BP) neural networks. By leveraging the nonlinear approximation and online learning capabilities of BP networks, the proposed method effectively reduces computational complexity and adapts to model uncertainties. The framework was validated through simulations in intelligent transportation scenarios, demonstrating its robustness and superior real-time performance compared to traditional approaches. Future work will explore large-scale MAS and real-world applications to enhance its scalability and adaptability.