基于自适应神经网络的异构多智能体系统一致性控制

Consensus control for heterogeneous multi-agent systems based on adaptive neural network approach

针对由一阶线性和二阶非线性组成的异构多智能体系统,提出一种基于自适应神经网络的分布式控制协议.利用每个智能体之间的通信信息调整分布式控制协议中的参数,使径向基函数(radial basis function,RBF)神经网络逼近二阶多智能体系统中的未知非线性项,进而使得异构多智能体系统实现自适应一致,并通过Lyapunov稳定性理论给予证明.MATLAB/Simulink仿真实验结果验证了该方法的有效性.

A distributed control protocol based on adaptive neural networks is proposed for heterogeneous multi-agent systems composed of first-order linear and second-order nonlinearities.The communication information between each agent is used to adjust the parameters in the distributed control protocol,so that the radial basis function(RBF)neural network can approximate the unknown nonlinear term in the second-order multi-agent systems,so that the heterogeneous multi-agent system can achieve adaptive consistency and the result is proved via Lyapunov stability theory.Finally,the simulation results obtained by MATLAB/Simulink verify the effectiveness of the proposed method.