随机事件驱动的网络化控制系统研究及其应用

This project is concerned with networked control systems whose capacity is limited and nodes driven by stochastic events, with the aim of studying the basic properties, analysis and design methodology and its application to vehicular cooperation control. In summary, the main objectives of this project are threefold: (1) Understanding the basic properties and dynamic behavior of such type of systems. Especially, we want to make clear the relationship between the system behavior and characteristic parameters of the driving event, the connection of the system performance with the number of active nodes, and the effect of network constraint on stability and control performance. (2) Setting up the framework of analysis and design of such systems, by taking into account characteristic parameters of the driving event, protocol sequences and the capacity limitation. To this end, we plan to model the stochastic driving events as Markov chains or protocol sequences and consider two operation modes for the inactive nodes: zero-order hold and zero output. Then, based on the capacity limitation, the characteristic parameters of the driving events, network-induced delays and random packet dropouts, we aim to set up a theoretical framework for system analysis, node assignment and controller codesign. (3) Applying the theoretical results obtained to laboratory scale vehicular cooperation control system, with the aim of acquiring practical networked control algorithms and technology. The research work outlined in this project has important scientific value in shaping the control theory of stochastic event-driven systems.

本项目旨在针对容量有限且节点受随机事件驱动的网络化控制系统，研究这种系统的基本特点、分析与设计方法及其在Vehicle协作控制中的应用。本项目的主要研究内容可概括为如下三点：（1）理解这种系统的基本特点及其运动规律。重点理解驱动事件特征参数与系统属性的关系、活动节点数量与系统性能的关系以及网络限制因素与稳定性及控制性能的关系。（2）融合驱动事件的特征参数、通信协议及网络限制因素，建立这种系统分析和设计的方法体系。采用Markov链或协议序列描述随机驱动事件，不活动节点采用零阶保持器和"零"输出两种工作方式，结合信息容量限制、驱动事件特征参数、网络传输延时和随机丢包因素，建立关于系统分析、节点分配与控制器协同设计的理论体系。（3）将所得理论成果应用于实验室规模的Vehicle协作控制中，形成实用的网络化控制算法和技术。本项研究对形成随机事件驱动系统的控制理论体系具有较高的科学意义。

本项目研究了受事件驱动的控制系统的分析与设计方法，并针对智能交通应用研究了车辆协作控制问题。本项目的主要研究成果有：（1）对事件驱动系统的特点及其运动规律有了系统深入的理解，尤其是随机驱动事件的统计特征参数与系统性能的关系、以及网络信道限制与稳定性及控制性能的关系。（2）融合驱动事件的特征参数、通信协议及网络限制因素，建立了这种系统分析和设计的方法体系。将随机驱动事件用马科夫过程和伯努利过程来描述，确定性驱动事件用协议序列来描述，不活动节点采用零输出工作方式，结合信道容量限制、驱动事件特征参数等因素，建立了这种系统稳定性分析、任务分配与控制器协同设计理论体系。（3）将所得理论成果应用于智能车路控制系统之中，并用实验室小车加以验证，形成实用的网络化控制算法。