摘要
研究无速度反馈的不确定性自由漂浮空间机器人轨迹跟踪控制问题,为了提高控制性能,采用一种观测器神经网络自适应鲁棒控制方法。利用神经网络设计观测器对系统的速度信息进行估计;其次,利用神经网络对系统模型的非线性进行逼近,设计神经网络自适应鲁棒控制器对系统进行控制,无需建立复杂的数学模型,并且所设计的自适应律能够进行在线学习;最后,根据H∞理论设计的鲁棒控制器保证了系统稳定,并使系统的L2增益小于给定指标。仿真结果表明,上述方法在无速度信息反馈下仍能进行有效的跟踪控制。
Considering trajectory tracking of a class of uncertainty free-floating space robot without speed feedback,a neural network adaptive robust control strategy based on observer is used. Firstly,the angular velocity is estimated by neural network. Secondly,another neural network is used to approximate the nonlinear systems,the neural network adaptive robust controller is applied to control the system without creating complex mathematical model,and the adaptive law is learning online. Finally,the robust controller is proposed based on H∞theory,and ensures the stability of the whole system,the L2 gain is less than the given index. The simulation results show that the proposed strategy can work effectively without speed feedback.
出处
《计算机仿真》
CSCD
北大核心
2015年第3期370-374,共5页
Computer Simulation
