通信约束下间歇量测的多自主体系统趋同控制研究

Due to influence of bandwidth constraints, communication costs or equipment failure, it is inevitable that communication constraints and intermittent measurements of information occur in multi-agent network systems. This results in that the controller analysis and design is more difficult. The project consider there exists communication constraints such as communication delay, the intermittent measurements is modeled as stochastic sampling, then event-triggered communication strategy is constructed to quantify sampled information selectively, design consensus controller and establish the overall work structure for consensus control of multi-agent systems with communication constraints and intermittent measurements. Then, for the case that state information is measured intermittently, quantified state-based consensus control law is designed and analyzed. And observer-based quantified dynamic output feedback consensus control law and the stability of system is researched for only measured output information. Furthermore, considering multi-agent systems with nonlinear dynamics, the design and stability problems for the consensus controller based on quantified state feedback and based on quantified output feedback with observer are studied, respectively. Finally, a unified and effective consensus control scheme is formed for multi-agent systems with communication constraints and intermittent measurements, and is verified, improved and perfected in the multi-robot systems. Research results will further improve the theoretical basis of consensus control for multi-agent systems, and provide a theoretical basis and guidance for popularization and application of consensus control methods for multi-agent systems under communication constraints and intermittent measurements.

多自主体网络系统因带宽限制、通信成本、设备故障等影响不可避免会导致通信约束及信息间歇量测的发生，使趋同控制器分析和设计愈加困难。本项目在通信时滞等通信约束的存在下，将间歇量测建模成随机采样方式，建立事件触发通信策略对采样信息有选择地量化并设计趋同控制器，构建通信约束下间歇量测多自主体系统趋同控制的整体工作结构。然后，针对状态信息可间歇量测情况，设计并分析基于量化状态的趋同控制律，仅输出信息可测时，提出基于观测器的量化动态输出反馈趋同控制律并分析稳定性。考虑具有非线性动态的多自主体系统，研究基于量化状态反馈的趋同控制器和基于观测器的量化输出反馈趋同控制器的设计和稳定性问题。最后，形成统一、有效的通信约束下间歇量测多自主体系统趋同控制方案，并在多机器人系统中验证、改进和完善。研究成果将进一步完善多自主体趋同控制的理论基础，为通信约束下间歇量测的多自主体趋同控制方法的推广应用提供理论依据和指导。

本项目主要针对通信约束下间歇量测的多自主体系统的趋同问题，对多自主体系统的分布式趋同控制方法及其在多单摆网络系统和多机器人/机械臂系统中的实验验证和分析、基于趋同性的分布式估计和滤波及其在分布式故障检测中的应用等进行了研究。深入考虑并探讨了资源受限、带宽限制、通信约束、系统状态信息不可测及通信网络拓扑结构发生变化等实际因素对多自主体系统稳定性的影响，分别提出了基于相对输出反馈的趋同性控制、基于动态输出反馈的一致性跟踪控制、基于间歇通信的一致性跟踪控制方法等，并将提出的控制方法应用于多单摆网络系统，实现了系统的状态，即单摆的角度、角速度和电机电枢电流分别达到趋同控制目标；研究切换通信网络环境下，带有通信时滞的分布式自适应趋同性控制及其在多机器人/机械臂系统的应用问题；考虑有损传感器网络下，分别针对一类带有时变时滞的线性离散系统和不确定性的T-S模糊系统，分别提出一种新的基于事件触发机制的分布式H∞一致性滤波和分布式H∞故障检测策略，形成统一、有效的通信约束下间歇量测多自主体系统趋同控制方案，为多自主体系统的推广应用提供一定的理论基础和实际价值。