自主式水下机器人推进器的故障诊断与容错控制方法研究

项目名称： 自主式水下机器人推进器的故障诊断与容错控制方法研究

项目编号： No.51509150

项目类型： 青年科学基金项目

立项/批准年度： 2016

项目学科： 其他

项目作者： 褚振忠

作者单位： 上海海事大学

项目金额： 20万元

中文摘要： 由于海洋环境的复杂性与不可预测性，自主式水下机器人(AUV: Autonomous Underwater Vehicle)一旦出现故障，不仅无法完成水下作业任务，甚至自身也无法回收，损失极大。推进器作为AUV主要的动力装置，构造复杂且直接与海水接触，是AUV最主要的故障来源之一。本项目研究海流干扰下的AUV推进器故障诊断与容错控制方法。首先，基于局部回归型神经网络建立以状态空间方程形式描述的AUV动力学模型，提出基于终端滑模观测器的推进器故障诊断方法，解决AUV速度状态不可直接测量时的状态估计误差有限时间收敛性问题以及海流干扰下的推进器故障重构问题；在此基础上，研究基于PD（比例微分）滑模的AUV区域跟踪控制方法以及基于加权最小范数的推力分配方法，解决AUV区域跟踪控制中的目标区域收敛性问题以及推进器主动容错控制问题；最后，通过水池实验验证AUV推进器故障诊断与容错控制方法的有效性。

中文关键词： 水下机器人；故障诊断与容错控制；终端滑模观测器；区域跟踪控制

英文摘要： Since the complexity and the unpredictability of marine environment, once a fault happens to AUV (Autonomous Underwater Vehicle), not only the underwater task could not be achieved, but also the AUV sometimes could not be reclaimed, which would cause economic loss. As the main power plant of AUV, the configuration of thruster is complex and it is contacted with the sea water in direct. So, the thruster is one of the most important fault sources for AUV. This project researches the fault diagnosis and fault tolerant control method under the current disturbance. First, an AUV dynamic model is built in the form of state space equation based on local recurrent neural network. And a fault diagnosis method based on Terminal sliding mode observer is proposed, which can make state estimation error converge to zero in a finite time when the speed state of AUV cannot be measured and achieve fault reconstruction under the current disturbance. On this basis, a PD-sliding mode based region tracking control method and a weighted minimum norm based thrust allocation method are proposed, which can be used to solve the problem of target region convergence in region tracking control and the problem of active fault tolerant control of thruster. Finally, the pool experiments are taken to verify the effectiveness of the proposed fault diagnosis and fault tolerant control method.

英文关键词： underwater vehicle;fault diagnosis and fault tolerant control;Terminal sliding mode observer;region tracking control