具有通信约束和干扰的不确定网络化控制系统的鲁棒控制研究

Abstract: Networked control systems (NCSs) have been widely used in industry, national defense, aerospace, and so on, and have important theoretical and practical value for their research. The problem of communication constraints (such as random delay, packet dropout and network attacks) and external disturbance in NCSs can reduce control performance, damage stability, and even cause the entire system to crash. This project studies the robust control of uncertain networked control systems with communication constraints and disturbances. Mainly including: For the uncertain networked control systems with random delay and packet dropout, a network predictive control algorithm based on robust sliding mode control is proposed to compensate for them. For the uncertain networked control systems with random delay，packet dropout and external disturbances, a robust H∞ control algorithm is designed to compensate the network and damp out the external disturbance. For the uncertain networked control systems with random delay，packet dropout, external disturbances and false data injection (FDI) attack, a robust adaptive sliding mode control is proposed to compensate communication constraints and damp out the disturbance. Finally, taking the networked robot teleoperation system as an example, the effectiveness of the algorithm is verified by semi-physical simulation.

摘要：网络化控制系统已被广泛应用在工业、国防、航空航天等各个领域，对其研究具有重要的理论意义和实用价值。然而网络中的通信约束(如随机延时、数据丢包和网络攻击)及外界干扰问题不仅会降低系统的性能，而且会破坏系统的稳定性，甚至导致整个系统崩溃。本项目研究具有通信约束和干扰的不确定网络化控制系统的鲁棒控制。主要研究内容为：针对不确定网络化控制系统中的随机时延和数据丢包问题，提出一种基于鲁棒滑模控制的网络预测控制算法；针对不确定网络化控制系统中随机延时、数据丢包和外界干扰问题，提出一种基于鲁棒H∞控制的网络补偿和干扰抑制算法。针对不确定性网络化控制系统中的随机延时和丢包、外界输入干扰及虚假数据注入(false data injection，FDI)攻击问题，提出一种基于鲁棒自适应滑模控制的通信约束补偿和干扰抑制算法。最后，以网络化机器人遥操作系统为实例，通过半实物仿真验证算法的有效性。

网络化控制系统已被广泛应用在工业、国防、航空航天等各个领域，对其研究具有重要的理论意义和实用价值。然而网络中的通信约束和外界干扰问题不仅会降低系统的性能，而且会破坏系统的稳定性，甚至导致整个系统崩溃。本项目研究具有通信约束和干扰的不确定网络化控制系统的鲁棒控制。主要研究内容为：针对不确定网络化控制系统中的随机时延和数据丢包问题，提出一种基于鲁棒滑模控制的网络预测控制算法；针对不确定网络化控制系统中随机延时、数据丢包和外界干扰问题，提出一种基于鲁棒H∞控制的网络补偿和干扰抑制算法。针对不确定性网络化控制系统中的随机延时和丢包、外界输入干扰及虚假数据注入攻击问题，提出一种基于鲁棒自适应滑模控制的通信约束补偿和干扰抑制算法。最后，以网络化电机和汽车系统为实例，通过数值仿真和半实物仿真验证算法的有效性。