行動生成はどこから始めるべきか？生成ロボットポリシーのための学習可能なソース事前分布
Where Should Action Generation Begin? A Learnable Source Prior for Generative Robot Policies
生成ロボットポリシーの初期ノイズ分布を、 proprioception に基づく学習可能な対角ガウス分布に置き換える LeaP を提案し、15の操作タスクで成功率を向上させた。
著者: Meipo Dai, Qiyuan Zhuang, He-Yang Xu, Ying-Jie Shuai, Yijun Wang, Qi Dou, Xiu-Shen Wei
分類: cs.RO, cs.CV, cs.LG
原文アブストラクト
Generative robot policies typically begin action generation from an observation-independent standard Gaussian distribution, leaving the choice of source distribution underexplored. This work asks a simple question: where should action generation begin? We propose LeaP, a Learnable source Prior that replaces the standard Gaussian with a proprioception-conditioned diagonal Gaussian over action chunks. Parameterized by a lightweight MLP, LeaP jointly predicts the mean and state-adaptive variance of the source distribution, while keeping the downstream generator architecture and inference solver unchanged. This design provides an observation-informed yet stochastic initialization, allowing the generator to focus on precise action refinement rather than transporting samples from an uninformed noise source. On 15 RoboTwin manipulation tasks, LeaP achieves an average success rate of 81.6%, outperforming four representative baselines -- including deterministic-source methods, a no-prior counterpart, and a diffusion-bridge policy -- by 6.5 to 25.5 percentage points. The same prior consistently improves both flow-matching and diffusion-bridge generators, while using fewer parameters and converging faster. The advantage carries over to real-world deployment, where LeaP attains the best performance. These results suggest that the source distribution is an independent and reusable design axis for generative robot policies, complementary to the choice of generative dynamics.