Adaptive Neural Network Sliding Mode Controller for Welding Robot with Uncertain Model in Nuclear Power Plant

Jiale Huan, Qingxin Shi, Pu Chen, Xingguang Duan*

*此作品的通讯作者

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

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.

源语言英语
主期刊名Intelligent Robotics and Applications - 17th International Conference, ICIRA 2024, Proceedings
编辑Xuguang Lan, Xuesong Mei, Caigui Jiang, Fei Zhao, Zhiqiang Tian
出版商Springer Science and Business Media Deutschland GmbH
437-448
页数12
ISBN(印刷版)9789819607709
DOI
出版状态已出版 - 2025
活动17th International Conference on Intelligent Robotics and Applications, ICIRA 2024 - Xi'an, 中国
期限: 31 7月 20242 8月 2024

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
15201 LNAI
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议17th International Conference on Intelligent Robotics and Applications, ICIRA 2024
国家/地区中国
Xi'an
时期31/07/242/08/24

指纹

探究 'Adaptive Neural Network Sliding Mode Controller for Welding Robot with Uncertain Model in Nuclear Power Plant' 的科研主题。它们共同构成独一无二的指纹。

引用此