张国腾, 荣学文, 李贻斌, 柴汇, 李彬. 基于虚拟模型的四足机器人对角小跑步态控制方法[J]. 机器人, 2016, 38(1): 64-74. DOI: 10.13973/j.cnki.robot.2016.0064
引用本文: 张国腾, 荣学文, 李贻斌, 柴汇, 李彬. 基于虚拟模型的四足机器人对角小跑步态控制方法[J]. 机器人, 2016, 38(1): 64-74. DOI: 10.13973/j.cnki.robot.2016.0064
ZHANG Guoteng, RONG Xuewen, LI Yibin, CHAI Hui, LI Bin. Control of the Quadrupedal Trotting Based on Virtual Model[J]. ROBOT, 2016, 38(1): 64-74. DOI: 10.13973/j.cnki.robot.2016.0064
Citation: ZHANG Guoteng, RONG Xuewen, LI Yibin, CHAI Hui, LI Bin. Control of the Quadrupedal Trotting Based on Virtual Model[J]. ROBOT, 2016, 38(1): 64-74. DOI: 10.13973/j.cnki.robot.2016.0064

基于虚拟模型的四足机器人对角小跑步态控制方法

Control of the Quadrupedal Trotting Based on Virtual Model

  • 摘要: 为提高四足机器人对角小跑运动的稳定性,实现机器人躯干 6 维运动方向控制的解耦,提出了一种基于虚拟模型的对角小跑步态控制方法.控制器主要包括支撑相虚拟模型控制和摆动相虚拟模型控制.在支撑相,建立了作用于躯干质心的虚拟力与对角支撑腿关节扭矩之间的数学关系,通过调整躯干虚拟力的大小控制躯干的高度与姿态,控制机器人前进速度和自转角速度.在摆动相,将机器人侧向速度控制引入到足端轨迹规划中,并通过虚拟的“弹簧-阻尼”元件驱动摆动足沿给定轨迹运动.此外,在控制器设计过程中,引入了状态机,用于监控机器人各腿的状态,并输出对角小跑步态相位切换指令.仿真实验结果表明,机器人能够以对角小跑步态在平地上进行全方位移动,跨越不平坦地形,并能够抵抗外部冲击,证明了文中控制方法的有效性和鲁棒性.

     

    Abstract: In order to improve the stability of the trotting quadruped robot and to decouple the control of the robot torso motion along six directions, an approach based on virtual model is presented for trot gait control. The controller mainly consists of two main modules: the virtual model control at support phase and the virtual model control at flight phase. During the support phase, the mathematical relationship are mapped between the joint torques of diagonal support legs and the virtual forces acted on the center-of-mass of the torso. And the values of virtual torso forces are regulated to control the torso attitude and height, as well as the forward velocity and the yaw angular velocity of the robot. During the flight phase, lateral velocity is introduced to plan the toe trajectory. And virtual spring-damper sections are implemented to drive the flight toes to track the planned trajectories. In addition, while designing the controller, a state machine is introduced to monitor the legs' states and output phase switching commands for trot gait regulation. The simulations show that the robot is able to trot omni-directionally on flat ground as well as uneven terrains, even suffering from external impacts. And thus the effectiveness and robustness of the controller are proved.

     

/

返回文章
返回