系统工程与电子技术 ›› 2024, Vol. 46 ›› Issue (4): 1287-1296.doi: 10.12305/j.issn.1001-506X.2024.04.17

• 系统工程 • 上一篇    下一篇

基于自适应超螺旋观测器的空间机械臂鲁棒故障诊断

高升1,2, 张海龙3, 张伟1,2,*, 孔维国1,2   

  1. 1. 中国科学院沈阳自动化研究所机器人学国家重点实验室, 辽宁 沈阳 110016
    2. 中国科学院机器人与智能制造创新研究院, 辽宁 沈阳 110169
    3. 航天器在轨故障诊断与维修重点实验室, 陕西 西安 710043
  • 收稿日期:2023-04-20 出版日期:2024-03-25 发布日期:2024-03-25
  • 通讯作者: 张伟
  • 作者简介:高升 (1991—), 男, 助理研究员, 博士, 主要研究方向为故障诊断、空间机械臂控制
    张海龙 (1984—), 男, 高级工程师, 博士, 主要研究方向为航天测运控数据分析与应用、智能诊断与健康评估
    张伟 (1979—), 男, 研究员, 博士, 主要研究方向为卫星、深空探测、在轨服务
    孔维国 (1989—), 男, 助理研究员, 硕士, 主要研究方向为星载结构与机构及其可靠性设计
  • 基金资助:
    国家自然科学基金(61973100);中国科学院沈阳自动化研究所基础研究计划项目(2022JC3K03);中国科学院沈阳自动化研究所研究室发展基金(E155380107)

Robust fault diagnosis for space robot manipulator based on adaptive super-twisting observer

Sheng GAO1,2, Hailong ZHANG3, Wei ZHANG1,2,*, Weiguo KONG1,2   

  1. 1. National State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China
    2. Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110169, China
    3. Key Laboratory for Fault Diagnosis and Maintenance of Spacecraft in Orbit, Xi'an 710043, China
  • Received:2023-04-20 Online:2024-03-25 Published:2024-03-25
  • Contact: Wei ZHANG

摘要:

针对空间机械臂系统执行器鲁棒故障诊断问题, 提出一种基于改进的自适应超螺旋观测器故障诊断方法。为抑制复杂空间环境引起的外部干扰对故障诊断结果的影响, 在经典超螺旋观测器的基础上引入观测器参数自适应调整算法, 同时解决观测器参数过估计问题与噪声扩展问题。此外, 引入小于1的分数幂和线性比例项来提高观测器的平滑性和快速性, 进一步改善故障诊断效果。基于Moreno-Lyapunov函数算法分析了观测器的有限时间稳定性, 证明了观测器的估计误差可以在有限时间内收敛至包含原点的邻域内, 提出残差生成方法并设计了基于自适应阈值的故障诊断策略。最后, 结合小行星采样空间三连杆机械臂算例, 通过仿真验证了所提方法的有效性。

关键词: 空间机械臂, 执行器故障, 故障诊断, 残差信息, 自适应超螺旋观测器

Abstract:

In view of the problem of robust fault diagnosis of actuators in space robot manipulators a fault diagnosis method based on improved adaptive super-twisting observer is pro-posed. To attenuate the influence of external disturbances caused by the sophisticated space environment on the fault diagnosis results, an adaptive adjusting algorithm of the observer parameters is introduced based on the classical super-twisting observer, and the problem of over-estimation of the observer parameters and noise expansion are solved simultaneously. In addition, the smoothness and rapidity of the observer are improved by introducing additional fractional power less than 1 and linear term to further enhance the fault diagnosis effect. Then, the finite-time stability of the observer is analysed based on the Moreno-Lyapunov function algorithm, which proves that the estimation error of the observer can converge to a region of zero in finite time. Moreover, a residual generation method is proposed and a fault diagnosis scheme based on the adaptive threshold is further designed. Finally, the effectiveness of the proposed method is verified by simulations expriment of the asteroid sampling space three-link robot manipulator example.

Key words: space robot manipulator, actuator fault, fault diagnosis, residual imformation, adaptive super-twisting observer

中图分类号: