DiffusionAnything: 統一ナビゲーションと把持前動作のためのエンドツーエンド・インコンテキスト拡散学習
DiffusionAnything: End-to-End In-context Diffusion Learning for Unified Navigation and Pre-Grasp Motion
RGB画像のみから、メートル規模のナビゲーションとセンチメートル規模のマニピュレーションを単一モデルで計画する拡散ポリシーを提案。タスクごとに5分の自己教師ありデータで学習し、未知シーンへのゼロショット汎化を実現する。
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: Iana Zhura, Yara Mahmoud, Jeffrin Sam, Hung Khang Nguyen, Didar Seyidov, Miguel Altamirano Cabrera, Dzmitry Tsetserukou
分類: cs.RO
原文アブストラクト
Efficiently predicting motion plans directly from vision remains a fundamental challenge in robotics, where planning typically requires explicit goal specification and task-specific design. Recent vision-language-action (VLA) models infer actions directly from visual input but demand massive computational resources, extensive training data, and fail zero-shot in novel scenes. We present a unified image-space diffusion policy handling both meter-scale navigation and centimeter-scale manipulation via multi-scale feature modulation, with only 5 minutes of self-supervised data per task. Three key innovations drive the framework: (1) Multi-scale FiLM conditioning on task mode, depth scale, and spatial attention enables task-appropriate behavior in a single model; (2) trajectory-aligned depth prediction focuses metric 3D reasoning along generated waypoints; (3) self-supervised attention from AnyTraverse enables goal-directed inference without vision-language models and depth sensors. Operating purely from RGB input (2.0 GB memory, 10 Hz), the model achieves robust zero-shot generalization to novel scenes while remaining suitable for onboard deployment.
