单自由度有限循环行走机构运动控制及参数识别  

Speed Control and Parameter Identification of 1-DOF Limit Cycle Walker

作　　者：危清清 林云成[1,2] 肖轩 陈磊[1,2] 刘宾 王耀兵 WEI Qingqing;LIN Yunchen;XIAO Xuan;CHEN Lei;LIU Bin;WANG Yaobing(Beijing Key laboratory of Intelligent Space Robotic Systems Technology and Applications,Beijing,100094,China;Beijing Institute of Spacecraft System Engineering,China Academy of Space Technology,Beijing 100094,China;School of Aerospace Engineering,Tsinghua University,Beijing 100084,China)

机构地区：[1]空间智能机器人系统技术与应用北京重点实验室,北京100094 [2]中国空间技术研究院北京空间飞行器总体设计部,北京100094 [3]清华大学航天学院,北京100084

出　　处：《载人航天》2019年第5期673-679,共7页Manned Spaceflight

基　　金：载人航天预先研究项目(030601)

摘　　要：针对机器人物理参数的辨识会对目标状态控制产生影响的问题,设计了一种基于模型的反馈控制器,以克服物理参数辨识误差从而实现目标速度的有限循环行走。首先,设计一种组合式无框轮,通过动态规划运动方程进行反馈控制。其次,当假定物理参数未知并改为使用预测参数时,进行数值模拟以分析目标的行走速度和其他属性。控制器具有一定的预测误差适应性,所实现的运动速度与目标速度的误差可以控制在0.001%,通过提出的神经网络模型预测物理参数,预测参数的平均误差为1.1%。A model-based feedback controller which could generate limit cycle walking at target walking speed was studied and the physical parameters were identified through neural network.First,a combined rimless wheel(CRW)was developed and the feedback control was realized by dynamic planning of the motion equation.Second,the numerical simulations were conducted to analyze the walking speed and other properties when the physical parameters were assumed unknown and the prediction parameters were used instead.The controller had a certain adaptability to the prediction error(0.001%).The physical parameters could be predicted through a proposed neural network model with an average error of 1.1%.

关 键 词：步行机器人 神经网络 有限循环行走 速度控制 参数辨识 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]