Adaptive NN Fault-Tolerant Control for Underactuated MSVs with Input Quantization

In this work, we have studied the problem of trajectory tracking control for underactuated marine surface vehicles (MSV s) subject to uncertain model parameters, unknown external disturbances, actuator undesirable faults, and input signal quantization. To effectively handle input signal quantization...

Full description

Saved in:
Bibliographic Details
Published inChinese Automation Congress (Online) pp. 8689 - 8694
Main Authors Zhang, Nanyan, Lai, Xiao
Format Conference Proceeding
LanguageEnglish
Published IEEE 17.11.2023
Subjects
Actuator fault
Actuators
Adaptation models
Artificial neural networks
input quantization
neural network
Quantization (signal)
Simulation
Trajectory tracking
Upper bound
Online AccessGet full text
ISSN2688-0938
DOI10.1109/CAC59555.2023.10451065

Cover

More Information
Summary:In this work, we have studied the problem of trajectory tracking control for underactuated marine surface vehicles (MSV s) subject to uncertain model parameters, unknown external disturbances, actuator undesirable faults, and input signal quantization. To effectively handle input signal quantization, a hysteresis quantizer is employed. The uncertain dynamics of the MSVs are approximated by using neural networks (NN). By resorting the adaptive technique to estimate an upper bound for the compound disturbances are lumped by time-varying external disturbances, actuator faults, and NN approximation errors. By introducing a continuous smooth function, the trajectory tracking control law was developed within backstepping frameworks. The theoretical deduce and simulation results confirm the effectiveness of the developed control strategy.
ISSN:2688-0938
DOI:10.1109/CAC59555.2023.10451065