A Less Conservative Condition for Flexible Control Lyapunov Functions used in Networked Predictive Control Systems

摘要

Nowadays, the networked control systems (NCSs) are used in different industries, including the manufacturing plants, automotive systems, power production and distribution, tele-robotics and tele-medicine. The NCSs bring a lot of advantages when dealing with large plants, but they also come with disadvantages issued by the network-enhanced complexities, one of them being caused by data-packet dropouts. The disadvantages usually degrade the control systems performances and can even lead to instability. Thus, in this paper, the problem of compensating the negative effects induced by the data-packet dropouts is tackled. For this, a predictive controller with inherited input-to-state stability and robustness is designed; the controller is based on the so-called flexible control Lyapunov functions and makes use of a newly developed method to model the network-induced data-packet dropouts effect as a disturbance. The flexibility of the control Lyapunov function is constrained by means of a new less conservative solution. The method's efficiency was tested in simulation on an electric power-assisted steering system controlled through controller area network and the results are promising.