模糊CMAC网络的双足机器人跑步轨迹规划

摘要

To study the center of mass trajectories and the feet of a biped robot in single-leg support and flight phases,a fuzzy cerebellar model articulation controller (FCMAC) neural network was applied to program the center of mass and the feet of the biped robot while running.A running trajectory for the biped robot was generated by FCMAC neural network learning.The simulation shows that the angle values and joint torque values for each joint change smoothly;the biped robot was able to stably run;and the effectiveness of the method was verified satisfactorily.%为了研究双足机器人在单腿支撑阶段和飞行阶段质心轨迹和双足轨迹,采用模糊CMAC(fuzzy cerebellar model articulation controller,FCMAC)神经网络规划双足机器人跑步时质心和双足的方法.通过FCMAC神经网络学习生成跑步轨迹,仿真实验结果表明:双足机器人跑步时各关节角度和关节驱动力矩变化稳定,能够实现稳定的跑步,验证了方法的有效性.