基于自适应动态规划的量化通信下协同最优输出调节  

Cooperative Optimal Output Regulation Under Quantized Communication Based on Adaptive Dynamic Programming

作　　者：王冰洁 徐磊 林宗利[2] 施阳 杨涛 WANG Bingjie;XU Lei;LIN Zongli;SHI Yang;YANG Tao(State Key Laboratory of Synthetical Automation for Process Industries,Northeastern University,Shenyang 110819,China;Department of Electrical and Computer Engineering,University of Virginia,Charlottesville 22904,USA;Department of Mechanical Engineering,University of Victoria,Victoria V8P5C2,Canada)

机构地区：[1]东北大学流程工业综合自动化国家重点实验室,沈阳110819 [2]弗吉尼亚大学电气和计算机工程系,美国夏洛茨维尔22904 [3]维多利亚大学机械工程系,加拿大维多利亚V8P5C2

出　　处：《自动化学报》2025年第4期813-823,共11页Acta Automatica Sinica

基　　金：国家自然科学基金(62133003,61991403);国家重点研发计划(2022YFB3305904)资助。

摘　　要：考虑了量化通信下多智能体系统的协同最优输出调节问题.为降低通信负担,利用取整量化器将智能体之间传输的浮点数数据转化为整数,从而减少通信信道中传输数据的比特数.通过将量化器引入编码−解码方案中,设计分布式量化观测器,保证在量化通信下,每个跟随者对外部系统状态的估计误差渐近收敛至零.在此基础上,在多智能体系统动态未知的情况下,提出基于自适应动态规划的数据驱动算法,在线学习次优控制策略,解决协同最优输出调节问题,保证每个跟随者的输出信号渐近跟踪参考信号,并抑制由外部系统产生的干扰信号.最后,在智能车联网自适应巡航控制系统上进行仿真实验并验证了所提方法的有效性.结果表明与精确通信相比,量化通信下比特数降低了58.33%.This paper considers the cooperative optimal output regulation problem of multi-agent systems under quantized communication.To reduce the communication burden,this paper uses the rounding quantizer to convert floating point data transmitted among agents into integers,reducing the number of bits of data transmitted in the communication channel.By introducing the quantizer into the encoder-decoder scheme,a distributed quantized observer for each follower agent is designed to ensure that the estimation error of the exosystem's state asymptotically converges to zero under quantized communication.On this basis,a data-driven algorithm based on adaptive dynamic programming is proposed to learn the suboptimal control strategy online with unknown multi-agent system dynamics.The algorithm solves the cooperative optimal output regulation problem,ensuring that each follower's output signal asymptotically tracks the reference signal and rejects disturbance signal generated by the exosystem.Finally,the simulation on the adaptive cruise control system of intelligent vehicle networking verifies the effectiveness of the proposed method.The results show that 58.33%of the bits are reduced under quantized communication compared with exact communication.

关 键 词：自适应动态规划 多智能体系统 输出调节 量化通信 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]