Firefly Optimized Neuro-Dynamic Programming for Optimal Tracking Control of Multi-Player Nonlinear Systems

Qiuye Wu1; Bo Zhao2; Derong Liu3

10.23919/CCC58697.2023.10241103

2023

1934-1768

979-8-3503-4259-8

2023 42nd Chinese Control Conference (CCC)

2023-July

1659-1663

24-26 July 2023

Tianjin, China

This paper develops a novel optimal tracking control (OTC) approach via firefly optimized neuro-dynamic programming for multi-player nonlinear systems. By constructing an augmented system composed of the tracking error and the reference trajectory, the OTC problem is transformed into an optimal regulation problem. To obtain the OTC policy of each player, the solution of the coupled Hamilton-Jacobi equation is approximated by a critic neural network, whose weight vector is tuned by the firefly algorithm instead of the gradient descent algorithm to avoid the difficulty in manually selecting the initial weight values. According to the Lyapunov's direct method, the tracking error of the closed-loop system is ensured to be uniformly ultimately bounded. The effectiveness of the developed approach is demonstrated by numerical simulation.

Adaptive Dynamic Programming Neuro-Dynamic Programming Optimal Tracking Control Firefly Algorithm Neural Networks Multi-Player Systems

其他

EI

20234515016044

Adaptive control systems ; Bioluminescence ; Closed loop control systems ; Dynamic programming ; Gradient methods ; Navigation ; Neural networks

Control Systems:731.1 ; Light/Optics:741.1 ; Optimization Techniques:921.5 ; Numerical Methods:921.6 ; Systems Science:961

IEEE

1.School of Automation, Guangdong University of Technology, Guangzhou, China
2.School of Systems Science, Beijing Normal University, Beijing, China
3.School of System Design and Intelligent Manufacturing, Southern University of Science and Technology, Shenzhen, China

Qiuye Wu,Bo Zhao,Derong Liu. Firefly Optimized Neuro-Dynamic Programming for Optimal Tracking Control of Multi-Player Nonlinear Systems[C],2023:1659-1663.