摘要
针对可重构机械臂系统位置传感器和速度传感器多故障,提出一种主动取代分散容错控制方法.基于可重构机械臂的模块化属性,设计正常工作模式下的分散神经网络控制器.利用微分同胚原理将子系统结构进行非线性变换,将传感器故障转化成伪执行器故障,设计分散滑模观测器以对多传感器故障进行实时检测,并利用其输出信号取代故障传感器信号,实现了多传感器故障情形下可重构机械臂的主动容错控制.仿真结果表明了所设计的容错控制方法的有效性.
An active substituting decentralized fault tolerant control scheme is established for reconfigurable manipulators with position and velocity multi-sensor failures. A decentralized neural network controller is presented at the normal working state based on the reconfignrable manipulators＇ modularity property. The sensor fault is transformed into the pseudo- actuator fault scenario by constructing a nonlinear transformation for subsystem structure with diffeomorphism theory, and decentralized sliding mode observers are designed for detecting multi-sensor failures in real time, then the fault signals are substituted with the output of the observers to realize active fault-tolerant control for reconfigurable manipulators under the multi-sensor fault state. Simulation results show the effectiveness of the proposed scheme.
出处
《控制与决策》
EI
CSCD
北大核心
2014年第2期226-230,共5页
Control and Decision
基金
国家自然科学基金面上项目(61374051
60974010)
吉林省科技发展计划项目(20110705)
关键词
可重构机械臂
多传感器故障
分散滑模观测器
分散容错控制
主动取代
reconfigurablemanipulators； multi-sensorfailures
decentralized sliding mode observer
decentralized fault-tolerant control
active substituting
作者简介
赵博（1987-），男，博士生，从事可重构机械臂故障诊断与容错控制的研究；
李元春（1962-），男，教授，博士生导师，从事复杂系统建模、智能机械与机器人控制等研究．
