Self-triggered optimal fault-tolerant control for saturated-inputs zero-sum game nonlinear systems via particle swarm optimization-based reinforcement learning

•In this paper, PSOA is used for the first time to solve the optimal ZSG control problem for nonlinear systems. The algorithm can compensate for faults negative effects while minimizing the cost function.•Different from previous works on ETC, the design of the STC strategy avoids continuous monitori...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Communications in nonlinear science & numerical simulation Ročník 153; s. 109512
Hlavní autoři: Zhao, Junzheng, Niu, Ben, Xu, Ning, Zong, Guangdeng, Zhang, Liang
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.02.2026
Témata:
Optimal control
Particle swarm optimization algorithm
Saturated-inputs
Self-triggered control
Zero-sum games
Self-triggered control
Zero-sum games
Saturated-inputs
Optimal control
Particle swarm optimization algorithm
ISSN:1007-5704
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•In this paper, PSOA is used for the first time to solve the optimal ZSG control problem for nonlinear systems. The algorithm can compensate for faults negative effects while minimizing the cost function.•Different from previous works on ETC, the design of the STC strategy avoids continuous monitoring for trigger errors in this study. Moreover, a better compromise between system performance and communication load can be achieved.•The design of the STC is suitable for nonlinear systems with saturated-inputs, which not only adjusts trigger moments in real time to reduce computational burden, but also ensures that the system maintains stable operation and avoids over-excitation. This paper proposes a self-triggered fault-tolerant optimal control scheme for zero-sum game nonlinear systems with saturated-inputs by using particle swarm optimization-based adaptive dynamic programming. First, a new self-triggered control strategy is proposed to equivalently convert a fault-optimal control problem into finding an optimal zero-sum game control scheme. Then, a saturated-inputs problem is solved by constructing a novel value function. In addition, a self-triggered condition is designed to reduce the communication bandwidth and energy consumption. And, the particle swarm optimization algorithm is used instead of gradient descent to find the optimal neural network weights. Finally, the effectiveness of the developed control scheme is demonstrated via simulation results.
ISSN:1007-5704
DOI:10.1016/j.cnsns.2025.109512