基于扩散式退火的力矩级运动控制全阶采样式MPC
Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealing
AI总结：
提出DIAL-MPC，一种基于扩散式退火的采样MPC框架，通过在线迭代优化实现全阶四足动力学实时控制，无需训练即显著降低跟踪误差并超越RL策略。
摘要：
由于高维和非凸性，使用全阶动力学模型进行腿足机器人的实时最优控制具有挑战性。因此，Nonlinear Model Predictive Control (NMPC) 方法常局限于降阶模型。基于采样的 MPC 在非凸甚至不连续问题中展现出潜力，但常产生高方差的次优解，限制了其在高维运动控制中的应用。本工作提出了 DIAL-MPC（Diffusion-Inspired Annealing for Legged MPC），一种具有新型扩散式退火过程的基于采样的 MPC 框架。该退火过程由 Model Predictive Path Integral Control (MPPI) 的理论景观分析及 MPPI 与单步扩散间的联系所支撑。在算法上，DIAL-MPC 在线迭代优化解，兼顾全局覆盖与局部收敛。在四足力矩级控制任务中，DIAL-MPC 将标准 MPPI 的跟踪误差降低了13.4倍，并在无训练情况下于挑战性攀爬任务中超越强化学习 (RL) 策略50%。此外，DIAL-MPC 实现了带负载的精确真实世界四足跳跃。据我们所知，DIAL-MPC 是首个实时优化全阶四足动力学的无训练方法。
英文摘要：
Due to high dimensionality and non-convexity, real-time optimal control using full-order dynamics models for legged robots is challenging. Therefore, Nonlinear Model Predictive Control (NMPC) approaches are often limited to reduced-order models. Sampling-based MPC has shown potential in nonconvex even discontinuous problems, but often yields suboptimal solutions with high variance, which limits its applications in high-dimensional locomotion. This work introduces DIAL-MPC (Diffusion-Inspired Annealing for Legged MPC), a sampling-based MPC framework with a novel diffusion-style annealing process. Such an annealing process is supported by the theoretical landscape analysis of Model Predictive Path Integral Control (MPPI) and the connection between MPPI and single-step diffusion. Algorithmically, DIAL-MPC iteratively refines solutions online and achieves both global coverage and local convergence. In quadrupedal torque-level control tasks, DIAL-MPC reduces the tracking error of standard MPPI by $13.4$ times and outperforms reinforcement learning (RL) policies by $50\%$ in challenging climbing tasks without any training. In particular, DIAL-MPC enables precise real-world quadrupedal jumping with payload. To the best of our knowledge, DIAL-MPC is the first training-free method that optimizes over full-order quadruped dynamics in real-time.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
