Scholarlink

This paper introduces a new adaptive integration method for solving the multi-contact problems in MuJoCo (Multi-Joint Dynamics with Contact). The proposed method is based on the classical embedded Runge-Kutta formulas and automatic step-size control. In addition, the transitions of contact states are carefully handled and the high-order convergence of solutions is achieved. We tested our algorithm on an 18-DOF quadruped robot with at most four contacts. The simulation results show an improvement in computational accuracy of one to two orders of magnitude, compared to the officially implemented semi-implicit Euler method and the Runge-Kutta method of order four.

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

High-Order Adaptive Integration of Contact Dynamics in MuJoCo

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Hongchen Li,
  • Ye Ding

摘要

This paper introduces a new adaptive integration method for solving the multi-contact problems in MuJoCo (Multi-Joint Dynamics with Contact). The proposed method is based on the classical embedded Runge-Kutta formulas and automatic step-size control. In addition, the transitions of contact states are carefully handled and the high-order convergence of solutions is achieved. We tested our algorithm on an 18-DOF quadruped robot with at most four contacts. The simulation results show an improvement in computational accuracy of one to two orders of magnitude, compared to the officially implemented semi-implicit Euler method and the Runge-Kutta method of order four.