<p>Hexapod robots demonstrate remarkable adaptability to complex terrains and maintain superior mobility stability, rendering them particularly suitable for field exploration and payload transportation applications. The implementation of a single-leg terrain probing mechanism for walkability assessment can substantially improve the mobility safety of hexapod robotic systems. Nevertheless, the control of support legs becomes considerably challenging due to unknown disturbances, including uncertain robot dynamics and unpredictable robot-ground interactions, which may compromise system stability. This study presents a novel variable admittance control strategy based on virtual decomposition control (VDC) for support leg regulation. The proposed methodology introduces two fundamental innovations, featuring the design of specialized adaptive controllers to address joint flexibility and nonlinear friction compensation, coupled with the integration of real-time ground stiffness estimation to enable dynamic adaptation of admittance control parameters. By exploiting the distinctive characteristics of VDC, the system’s stability has been theoretically guaranteed via the demonstration of all subsystems’ <i>virtual stability</i>, encompassing both free-space motion and leg-ground contact phases. The proposed approach’s effectiveness in optimizing robot-ground dynamic interactions is validated through comprehensive simulations using the ROS framework and Gazebo simulation environment.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Variable admittance control of single-leg dynamics: A key to stable hexapod robot locomotion

  • Nan Li,
  • Hongjun Xing,
  • Ruixiang Huang,
  • Xinqiang Zou,
  • Weihua Li,
  • Yiqun Liu,
  • Haibo Gao,
  • Liang Ding

摘要

Hexapod robots demonstrate remarkable adaptability to complex terrains and maintain superior mobility stability, rendering them particularly suitable for field exploration and payload transportation applications. The implementation of a single-leg terrain probing mechanism for walkability assessment can substantially improve the mobility safety of hexapod robotic systems. Nevertheless, the control of support legs becomes considerably challenging due to unknown disturbances, including uncertain robot dynamics and unpredictable robot-ground interactions, which may compromise system stability. This study presents a novel variable admittance control strategy based on virtual decomposition control (VDC) for support leg regulation. The proposed methodology introduces two fundamental innovations, featuring the design of specialized adaptive controllers to address joint flexibility and nonlinear friction compensation, coupled with the integration of real-time ground stiffness estimation to enable dynamic adaptation of admittance control parameters. By exploiting the distinctive characteristics of VDC, the system’s stability has been theoretically guaranteed via the demonstration of all subsystems’ virtual stability, encompassing both free-space motion and leg-ground contact phases. The proposed approach’s effectiveness in optimizing robot-ground dynamic interactions is validated through comprehensive simulations using the ROS framework and Gazebo simulation environment.