TY - JOUR
T1 - Fault-tolerant visual servo control for a robotic arm with actuator faults
AU - Li, Jiashuai
AU - Peng, Xiuyan
AU - Li, Bing
AU - Sreeram, Victor
AU - Wu, Jiawei
N1 - Publisher Copyright:
© The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.
PY - 2024/9
Y1 - 2024/9
N2 - The study targets uncertain coupling faults in robotic arm actuators and proposes a new fault-tolerant visual servo control strategy. Specifically, it considers both multiplicative and additive actuator faults within the dynamic of the robotic arm, treating the coupling faults and time-varying disturbances as an aggregate of concentrated uncertainties. A radial basis function neural network-based state observer is introduced to online approximate these concentrated uncertainties, which include fault information, eliminating the need for prior knowledge of faults. Furthermore, a fault-tolerant controller based on a non-singular fast terminal sliding mode is proposed, which separately decouples the nominal quantities and concentrated uncertainties and develops individual adaptive control laws for each. This effectively reduces the detrimental impact of coupled faults and disturbances on the system’s performance, facilitating image feature trajectory tracking control with minimal jitter, high precision, and strong transient response capabilities. The stability of the state observer and the fault-tolerant controller has been substantiated through Lyapunov’s theory. Lastly, numerical simulations validate the efficacy and robustness of the proposed fault-tolerant visual servo control approach.
AB - The study targets uncertain coupling faults in robotic arm actuators and proposes a new fault-tolerant visual servo control strategy. Specifically, it considers both multiplicative and additive actuator faults within the dynamic of the robotic arm, treating the coupling faults and time-varying disturbances as an aggregate of concentrated uncertainties. A radial basis function neural network-based state observer is introduced to online approximate these concentrated uncertainties, which include fault information, eliminating the need for prior knowledge of faults. Furthermore, a fault-tolerant controller based on a non-singular fast terminal sliding mode is proposed, which separately decouples the nominal quantities and concentrated uncertainties and develops individual adaptive control laws for each. This effectively reduces the detrimental impact of coupled faults and disturbances on the system’s performance, facilitating image feature trajectory tracking control with minimal jitter, high precision, and strong transient response capabilities. The stability of the state observer and the fault-tolerant controller has been substantiated through Lyapunov’s theory. Lastly, numerical simulations validate the efficacy and robustness of the proposed fault-tolerant visual servo control approach.
KW - Fault-tolerant control
KW - Image-based visual servoing
KW - Neural network observer
KW - Robotic arm
UR - http://www.scopus.com/inward/record.url?scp=85193499892&partnerID=8YFLogxK
U2 - 10.1007/s00521-024-09714-x
DO - 10.1007/s00521-024-09714-x
M3 - Article
AN - SCOPUS:85193499892
SN - 0941-0643
VL - 36
SP - 15815
EP - 15828
JO - Neural Computing and Applications
JF - Neural Computing and Applications
IS - 25
ER -