PALM: 知覚アライメントによる局所視覚運動ポリシーの汎化性向上
PALM: Enhanced Generalizability for Local Visuomotor Policies via Perception Alignment
操作ポリシーを大域的な粗い部分と局所的な細かい部分に分け、局所レベルで視覚的注目と固有受容感覚表現を揃えることで、追加の入力やデータ収集なしに分布外環境への汎化性能を高めた研究。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Ruiyu Wang, Zheyu Zhuang, Danica Kragic, Florian T. Pokorny
分類: cs.RO
原文アブストラクト
Generalizing beyond the training domain in image-based behavior cloning remains challenging. Existing methods address individual axes of generalization, workspace shifts, viewpoint changes, and cross-embodiment transfer, yet they are typically developed in isolation and often rely on complex pipelines. We introduce PALM (Perception Alignment for Local Manipulation), which leverages the invariance of local action distributions between out-of-distribution (OOD) and demonstrated domains to address these OOD shifts concurrently, without additional input modalities, model changes, or data collection. PALM modularizes the manipulation policy into coarse global components and a local policy for fine-grained actions. We reduce the discrepancy between in-domain and OOD inputs at the local policy level by enforcing local visual focus and consistent proprioceptive representation, allowing the policy to retrieve invariant local actions under OOD conditions. Experiments show that PALM limits OOD performance drops to 8% in simulation and 24% in the real world, compared to 45% and 77% for baselines.
