SDS:単一動画から四足歩行スキルを合成する「見て、やって、完了」パイプライン
SDS -- See it, Do it, Sorted: Quadruped Skill Synthesis from Single Video Demonstration
GPT-4oを用いて単一の動画から報酬関数を自動生成し、四足ロボットの歩容(トロット、バウンド、ペース、ホップ)を強化学習で獲得する手法を提案。シミュレーションと実機で100%の歩容一致を達成。
著者: Maria Stamatopoulou, Jeffrey Li, Dimitrios Kanoulas
分類: cs.RO
原文アブストラクト
Imagine a robot learning locomotion skills from any single video, without labels or reward engineering. We introduce SDS ("See it. Do it. Sorted."), an automated pipeline for skill acquisition from unstructured demonstrations. Using GPT-4o, SDS applies novel prompting techniques, in the form of spatio-temporal grid-based visual encoding ($G_{v}$) and structured input decomposition (SUS). These produce executable reward functions (RF) from the raw input videos. The RFs are used to train PPO policies and are optimized through closed-loop evolution, using training footage and performance metrics as self-supervised signals. SDS allows quadrupeds (e.g. Unitree Go1) to learn four gaits -- trot, bound, pace, and hop -- achieving 100% gait matching fidelity, Dynamic Time Warping (DTW) distance in the order of $10^{-6}$, and stable locomotion with zero failures, both in simulation and the real world. SDS generalizes to morphologically different quadrupeds (e.g. ANYmal) and outperforms prior work in data efficiency, training time and engineering effort. Further materials and the code are open-source under: https://rpl-cs-ucl.github.io/SDSweb/.
