Data-driven optimal tracking control of discrete-time linear systems with multiple delays via the value iteration algorithm

Abstract

In this paper, the optimal tracking problem for discrete-time linear systems with multiple delays is studied without system dynamics. It is known that the total state of a system without specific dynamic characteristics is hard to be measured, unless such a system is equipped with massive sensors, which, however, may lead to an increase in cost and complexity for analysing. To deal with this problem and avoid adverse effects caused by using system state information, a new data-driven value iteration (DDVI) algorithm is proposed by considering three factors: past control inputs, system outputs, and external reference trajectories. Before the algorithm is proposed, a transformation is made to the original system according to the characteristics of the time-delay system, so that the time-delay number can be reduced or become a delay-free system. A novel data-driven state equation is derived from the historical data of the three factors, and then, it is adopted to solve the optimal control of multi-delay systems. Further results show that the proposed DDVI algorithm is convergent and the tracking error is asymptotically stable. Finally, simulations are provided to show the effectiveness of the controller.

Keywords:

• Optimal tracking control
• value iteration algorithm
• multiple delays system

Disclosure statement

No potential conflicts of interest were reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author [Chaoli Wang] upon reasonable request.

Additional information

Funding

This paper was partially supported by the National Defense Basic Research Program [JCKY2019413D001], Natural Science Foundation of China [6217023627, 62003214] and Shanghai Natural Science Foundation [19ZR1436000].