发表机构
Carnegie Mellon University; University of Michigan, Ann Arbor（卡内基梅隆大学; 密歇根大学安娜堡分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对四足和人形机器人的约束运动规划难题，提出基于模型的几何感知生成优化（2GO），将主动约束几何转化为去噪算子并解耦生成传输与随机性，实验显示更高成功率、更少违规。
AI 中文摘要
四足机器人和人形机器人的约束运动规划（CLP）要求机器人满足碰撞避免、接触一致性、运动学可行性和支撑约束，在高维动力学和高度非凸环境下具有挑战性。最近的基于模型的扩散（MBD）方法将轨迹优化重新表述为对轨迹的后验采样，利用已知动力学和蒙特卡洛 rollout 来解析估计去噪得分函数，无需示范学习。虽然约束变体进一步将可行性纳入基于模型的得分 rollout 并显示出有前景的性能，但它们仍然受到以下限制：（1）缺乏任务调制的主动约束几何来塑造得分方向和反向随机性；（2）使用确定性的 DDPM 风格反向传输，而没有跨不同生成传输的自适应调度。因此，我们提出了用于约束运动的基于模型的几何感知生成优化（2GO），该方法通过法向诱导度量塑造、切空间随机滤波和基于 CFS 的收缩将主动约束几何转化为可执行的去噪算子。2GO 进一步通过自适应扩散和类流调度将生成传输与反向随机性解耦。在约束四足和人形运动上的实验表明，在离散落脚点选择和连续姿态规划方面具有强大性能，具有更高的成功率、更少的违规和改进的执行兼容性。
英文摘要
Constrained Locomotion Planning (CLP) for quadrupeds and humanoids, where robots must satisfy collision avoidance, contact consistency, kinematic feasibility, and support constraints, is challenging under high-dimensional dynamics and highly non-convex environments. Recent Model-Based Diffusion (MBD) approaches recast trajectory optimization as posterior sampling over trajectories, using known dynamics and Monte Carlo rollouts to analytically estimate the denoising score function without demonstration learning. While constrained variants further incorporate feasibility into model-based score rollouts and show promising performance, they are still limited by (1) lacking a task-modulated active constraint geometry that shapes the score direction and reverse stochasticity, and (2) using deterministic DDPM-style reverse transport without adaptive scheduling across different generative transports. Therefore, we introduce Model-Based Geometry-Aware Generative Optimization (2GO) for constrained locomotion, which turns active constraint geometry into executable denoising operators through normal- induced metric shaping, tangent-space stochastic filtering, and CFS-based retraction. 2GO further decouples generative transport from reverse stochasticity through an adaptive diffusion and flow-like schedule. Experiments on constrained quadruped and humanoid locomotion demonstrate strong performance in discrete foothold selection and continuous posture planning, with higher success rates, fewer violations, and improved execution compatibility.
