事件驱动采样下随机系统的控制与估计及在传感器网络中的应用

With great advantage in using the sampling and communication resource, event-based control has became a heated subject in the system and control community. Due to the extreme nonlinear behavior of systems arising in event-driven sampling, it is quite hard to design event based control and analyze the closed-loop systems mathematically. So far, there are only few quantitative results on stochastic event based control and estimation, and most works are still confined to the first-order systems. In this project, we devote to the following work. (1) By using stochastic analysis and optimal filtering, we would seek the optimal sampling estimation scheme and the lower bound of estimation errors for a typical stochastic system with the constraint of communication rates. (2) By analyzing the stochastic Lyapunov function, we would design an implementable event based control scheme for the higher-dimensional Markov jump system. (3) Combined with the results above, we would investigate some hot topics in sensor networks, such as the online sensor scheduling and the consensus problem for multiagent systems under the uncertain environment. This research not only could promote the development of stochastic event based control and estimation theory, but also would play an guiding role in design and implementation of event based sampling in the future engineering practice (e.g. sensor scheduling).

事件驱动的采样控制在利用采样和通信资源等方面有着明显的优势,已成为系统工程界的热门研究专题.由于事件驱动采样给系统带来严重的非线性行为,控制设计和系统分析变得十分困难. 到目前为止, 对随机系统的事件驱动采样控制与估计的研究结果相对较少，大多工作局限于一阶系统. 本项目拟致力于如下方面的研究：(1) 利用随机分析和最优估计理论给出限制采样率下随机线性系统状态估计问题的最优采样方案和估计误差下界. (2) 通过分析随机Lyapunov函数设计高维Markov跳变系统镇定问题可行的事件驱动采样控制方案. (3) 结合上述结果研究传感器网络中热点问题，包括在线传感器调度方案设计和不确定环境下多自主体系统的趋同问题. 预期结果不仅可推进随机事件驱动采样控制与估计理论的发展，而且有望对未来工程实践中事件驱动采样方案(比如传感器调度)的设计实施产生重要指导意义.

本项目主要研究了无线传感网络背景下事件驱动控制、平均场博弈以及分布式估计问题. 具体包括: (1) 研究了具有Markov切换拓扑和非线性耦合的复杂网络的事件驱动牵引控制. (2)研究了具有Markov跳变参数的平均场模型的社会最优控制. (3) 发展了电力系统的分布式(动态)状态估计的新方法.(4) 首次提出了具有黏性价格的动态产量调节问题的平均场模型. 预期结果不仅可推进事件驱动控制、平均场博弈与分布式估计理论的发展，而且有望对未来工程实践中控制(估计)器的设计实施产生重要指导意义.