The composite optimal outputtrackingcontrol problem of time-delayuncertainsystems is investigated in this paper. To improve the robustness of the uncertainsystems, a disturbance observer-based(DOB) compensation s...
详细信息
ISBN:
(数字)9789887581581
ISBN:
(纸本)9798350366907
The composite optimal outputtrackingcontrol problem of time-delayuncertainsystems is investigated in this paper. To improve the robustness of the uncertainsystems, a disturbance observer-based(DOB) compensation scheme is proposed to reduce the influence of the *** further enhance the control efficiency, we incorporate a Smith predictor into the control framework to precisely predict future states of the delayed system. Subsequently, a composite actor-critic reinforcementlearning(RL) scheme is built for approximating both the ideal value function associated with the compensated system and the optimal control policy in real-time, by augmenting the system states with the outputtracking error. To counteract the impact of the probing signal, concurrent learning is employed in this paper with the data collected from the known system *** results based on an experimental platform of slot die coating machines demonstrate the effectiveness of the proposed scheme.
暂无评论