TY - GEN
T1 - Adaptive Neural Network Sliding Mode Controller for Welding Robot with Uncertain Model in Nuclear Power Plant
AU - Huan, Jiale
AU - Shi, Qingxin
AU - Chen, Pu
AU - Duan, Xingguang
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
PY - 2025
Y1 - 2025
N2 - The welding robot in nuclear power plant is used for repairing tasks of control rod guide tubes, thermocouple guide tubes, and baffle screws to prevent loosening welds. It can effectively replace manual operations in high radiation underwater environments. Due to the model uncertainty of the robot system itself and external disturbances, it is difficult for the robot to accurately track operational trajectories. This paper proposes an Adaptive Neural Network Sliding Mode Controller (ANNSMC) applied to welding robots in nuclear power plants, which includes three parts: sliding mode controller (SMC), neural network (NN)controller, and barrier Lyapunov function (BLF) output-constrained controller. The controller uses NN to improve adaptability to uncertain parts of the system and utilizes BLF to ensure tracking capability of trajectories when subjected to external disturbances. A simulation was conducted on a 2-link robot to evaluate the performance of this controller. The steady-state position error at the end of the robot is better than 0.01 rad.
AB - The welding robot in nuclear power plant is used for repairing tasks of control rod guide tubes, thermocouple guide tubes, and baffle screws to prevent loosening welds. It can effectively replace manual operations in high radiation underwater environments. Due to the model uncertainty of the robot system itself and external disturbances, it is difficult for the robot to accurately track operational trajectories. This paper proposes an Adaptive Neural Network Sliding Mode Controller (ANNSMC) applied to welding robots in nuclear power plants, which includes three parts: sliding mode controller (SMC), neural network (NN)controller, and barrier Lyapunov function (BLF) output-constrained controller. The controller uses NN to improve adaptability to uncertain parts of the system and utilizes BLF to ensure tracking capability of trajectories when subjected to external disturbances. A simulation was conducted on a 2-link robot to evaluate the performance of this controller. The steady-state position error at the end of the robot is better than 0.01 rad.
KW - Neural network control
KW - Output-constrained control
KW - Sliding mode control
UR - http://www.scopus.com/inward/record.url?scp=85218454754&partnerID=8YFLogxK
U2 - 10.1007/978-981-96-0771-6_32
DO - 10.1007/978-981-96-0771-6_32
M3 - Conference contribution
AN - SCOPUS:85218454754
SN - 9789819607709
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 437
EP - 448
BT - Intelligent Robotics and Applications - 17th International Conference, ICIRA 2024, Proceedings
A2 - Lan, Xuguang
A2 - Mei, Xuesong
A2 - Jiang, Caigui
A2 - Zhao, Fei
A2 - Tian, Zhiqiang
PB - Springer Science and Business Media Deutschland GmbH
T2 - 17th International Conference on Intelligent Robotics and Applications, ICIRA 2024
Y2 - 31 July 2024 through 2 August 2024
ER -