针对双足机器人动态步行中存在的控制器复杂问题,提出了一种基于仿人智能控制的动态步行运动控制器。建立了平面五杆双足机器人动力学模型,通过模仿人类步行运动并根据动态步行过程中双腿的姿态变化,将动态步行复杂任务分解为顺序执行的4个过程,结合相平面法定性定量的分析,设计了步行运动控制模态及相应阶段的动觉智能图式群。仿真实验表明,该方法实现了双足机器人连续动态步行运动,控制方式简单易实现。
A human simulated intelligent control scheme based on sensory-motor schema for dynamic walking of biped robot is presented to solve the complex controller design. The dynamics model of a five-link planar biped is established. By imitating human walking movement and apprehending the posture change of the legs, the complex process of dynamic walking for biped robot is divided by four sequence phases. The multi-mode control law and the group of sensory-motor schema are designed based on the analysis of quality-quantity combination to the phase plane. Simulation results show that the proposed control strategy is effective to ensure continuous dynamic walking of biped robot and the control method is simple and easy to be realized.