基于未知输入观测器的多智能体一致性控制

Consensus control for multi-agent based on unknown input observer

现有研究成果大多仅考虑随机噪声或未知输入干扰单一存在的情况,实际工程中两者往往同时存在.在此背景下,本文针对一类含有未知输入干扰和随机噪声的非线性多智能体系统,提出了一种一致性控制协议方法.首先,针对单个智能体系统设计未知输入观测器以消除干扰项对状态估计的影响.参考Kalman滤波器算法来求解状态反馈矩阵,使得输出残差信号的协方差最小,从而增强系统对随机噪声的鲁棒性.然后,基于观测器的状态估计信息,设计了鲁棒一致性控制协议,并将其转化为线性矩阵不等式求解问题.最后,通过一个数值仿真证明了所提方法的正确性和有效性.

Most of the existing research results only consider the single existence of random noise or unknown input interference.In actual engineering,both often exist at the same time.In the context,This paper proposes a consensus control protocol method for a class of nonlinear multi-agent systems with unknown input interference and random noise.First,an unknown input observer is designed for a single agent system to eliminate the influence of interference items on state estimation.Refer to the Kalman filter algorithm to solve the state feedback matrix to minimize the covariance of the output residual signal,thereby enhancing the system’s robustness to random noise.Then,based on the state estimation information of the observer,a robust consensus control protocol is designed and convert the problem into a linear matrix inequality problem to solve it.Finally,a numerical simulation proves the correctness and effectiveness of the proposed method.