仿人机器人复杂动作设计中人体运动数据提取及分析方法

Human Motion Data Acquiring and Analyzing Method for Humanoid Robot Motion Designing

提出了仿人机器人复杂动作设计中人体运动数据提取及分析方法.首先,通过运动捕捉系统获取人体运动数据,并采用运动重定向技术,输出人-机简化模型的数据;然后,对运动数据进行分析和运动学解算,给出基于人体运动数据的仿人机器人逆运动学求解方法,得到仿人机器人模型的关节角数据;再经过运动学约束和稳定性调节后,生成能够应用于仿人机器人的运动轨迹.最终,通过在仿人机器人BHR-2上进行刀术实验验证了该方法的有效性.

The methods of human motion data acquiring and analyzing for humanoid robot based on human motion capture are presented in this paper. Firstly, human motion data are obtained through a motion capture system, and the data of simplified human-humanoid model are outputted using motion redirection technique. Secondly, inverse kinematic solving method of humanoid robot based on human motion data is presented, and the joint angle of humanoid robot is obtained. Thirdly, motion trajectory adapting to humanoid robot is obtained after kinematic constraints and stability adjustment. Finally, the effectiveness of the proposed method is illustrated by simulations and experiments of Chinese Kungfu ＂sword＂ using a 32 DOF （Degree of freedom） humanoid robot.