线性多智能体系统一致性的自适应动态规划求解方法
ADAPTIVE DYNAMIC PROGRAMMING TECHNIQUE FOR THE CONSENSUS OF LINEAR MULTI-AGENT SYSTEM
利用自适应动态规划的在线迭代算法来研究线性多智能体系统的一致性问题. 所研究的多智能体系统的状态矩阵和输入矩阵可以是已知的或未知的. 首先, 给出多智能体系统依赖初始时刻、终端时刻的性能指标; 然后, 将由初始时刻和终端时刻确定的时间段进行划分; 接着, 结合代数~Riccati 方程推导出迭代方程, 并在划分后的时间段上重复地利用系统的状态信息和输入信息进行迭代计算, 直至算法收敛为止; 最后,利用仿真试验验证了该算法的有效性.
In this paper, an on-line iterative algorithm, called adaptive dynamic programming technique is proposed to study the consensus problem of linear multi-agent systems. The state matrix and input matrix of the system can be known or unknown. Firstly, the performance index relying on the initial time and the terminal time is given. Secondly, the time interval determined by the initial time and the terminal time is divided. Then, an iterative equation is deduced based on the system equation and the algobraic Riccati equation. The state and the input are used for iterative calculation on these divided time intervals repeatedly till the algorithm converges. Finally, simulation results are given to show that the method is valid.
线性多智能体系统 / 一致性 / 自适应动态规划 / 代数~Riccati 方程. {{custom_keyword}} /
/
〈 | 〉 |