基于计算力矩的助行腿机器人神经网络补偿控制  

Neural-network Compensation Control for Exoskeleton Robot Based on Computed Torque Control

作　　者：冯治国[1] 

机构地区：[1]贵州大学,贵阳550025

出　　处：《中国机械工程》2013年第16期2173-2179,共7页China Mechanical Engineering

基　　金：国家高技术研究发展计划(863计划)资助项目(2006AA04Z224);国家自然科学基金资助项目(50975165)

摘　　要：在建立"机器主动"训练模式时助行腿机器人在跑步机上的步行动力学模型的基础上,设计了基于计算力矩加PD反馈的神经网络控制系统,并采用Lyapunov方法分析了控制系统的稳定性和收敛性。通过虚拟样机协同仿真平台进行了控制系统的仿真实验和样机系统测试验证,结果表明,该控制方法有效地消除了系统建模误差影响,提高了助行腿机器人轨迹跟踪能力。Based on the dynamics model of an exoskeleton robot walking on the treadmill for gait rehabilitation in robot--in--charge mode,the computed torque controller with a neural--compensator and portion--differential feedback was designed. According to the Lyapunov function, the stability and the conver- gence of the control system were proved. The simulation on the virtual prototype were carried out in a conllab- orative simulation platform. The results indicate that the proposed method can remove uncertain factors of the dynamic model effectively and increase the ability of the trajectory tracking.

关 键 词：步行训练 助行腿机器人 机器主动 计算力矩控制 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]