中文摘要：

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。

英文摘要：

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.