Cooperative fault-tolerant control of networked systems

容错控制是避免机器和过程因故障或故障的发生而发生安全危急情况和停机的重要手段。该项目的科学目标是阐述网络系统容错控制的新方法，以便在发生故障而子系统不再能够满足其目标时，通过在各子系统之间重新分配总体控制目标来保持整个系统的功能。所使用的体系结构不包括协调器，但子系统必须根据本地信息和它们通过数字网络接收的信息来做出决定。新的容错控制结构的组成应分三步进行阐述，包括网络化轨迹跟踪控制器的设计、分散式故障诊断和控制单元的协作重构。结果将进行实验评估。

Fault-tolerant control is an important means to avoid safety-critical situations and shut-downs of machines and processes due to the occurrence of faults or failures. The scientific aim of the project is to elaborate new methods of fault-tolerant control of networked systems such that the function of the overall system is retained by re-distributing the overall control aim among the subsystems after a failure has occurred and a subsystem is no longer able to satisfy its aim. The architecture used does not include a coordinator, but the subsystems have to make their decisions based on local information and information that they have received over a digital network. The components of the new fault-tolerant control architecture should be elaborated in three steps, which concern the design of networked trajectory-following controllers, the decentralised fault diagnosis and the cooperative reconfiguration of the control units. The results will be experimentally evaluated.