Fixed-time Fault Tolerant Control Based on Neural Network and Adaptive Sliding Mode Controller of Autonomous Underwater Vehicle with Actuator Saturation

Articles in Press

Document Type : Original Article

Authors

• Mostafa Taghi zadeh
• Majid Mokhtari
• mahmood mazare

Shahid Beheshti University - Faculty of Mechanics and Energy

10.22067/jacsm.2024.85527.1219

Abstract

The complexity of the sea environment, external disturbances and uncertainties in the system, as well as the failure and saturation of the actuators, are effective factors in the control of underwater vehicles. In this note, to deal with the mentioned problems, a robust and adaptive controller is proposed by combining the fast terminal dynamic sliding mode controller and the radial basis neural network. In the proposed approach, the problem of actuator saturation is considered and its stability is proved using Lyapunov theory. Fix-time convergence, estimation and dealing with external disturbances and uncertainties, active dealing with actuator fault, elimination of chattering phenomenon, and non-saturation of the actuator are the advantages of the proposed controller. The designed controller is applied on an autonomous underwater vehicle and the controller parameters are optimized to achieve the least error in tracking the reference trajectories as well as the shortest convergence time. To evaluate the performance of the proposed controller, it has been compared with a passive fault control method and PID controller, which has shown the superiority of the proposed control method in tracking the desired trajectories, the amount of control effort, dealing with actuators faults and external disturbances, as well as convergence in less time.

Keywords

• Autonomous underwater vehicle
• Adaptive dynamic fast terminal sliding mode controller
• Neural network
• Fault tolerant control
• Actuator saturation

Main Subjects

• Control

---