見たものに基づいて行動する:視覚言語行動モデルにおける安全な社会的ナビゲーションの実現
Act on What You See: Unlocking Safe Social Navigation in Vision-Language-Action Models
事前学習済みの視覚言語行動(VLA)モデルが歩行者と障害物の区別や衝突予測信号を内部表現に持つことを利用し、注釈不要の2段階ポストトレーニング手法SALSAを提案。反事実的なシーンと将来リスクの教師信号で行動生成と潜在表現を整合させ、近接衝突を86.4%削減した。
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: Qingzi Wang, Xiyang Wu, Guangyao Shi, Dianwei Chen, Xianfeng Yang, Dinesh Manocha
分類: cs.RO
原文アブストラクト
Safe social navigation requires robots to distinguish people from ordinary obstacles and to react before danger becomes imminent. We show that pretrained Vision-Language-Action (VLA) models already encode pedestrian-object distinctions and future collision signals in their internal representations, but behavior cloning fails to translate these signals into socially appropriate actions. To address this mismatch, we propose SALSA, a two-stage annotation-free post-training framework: (1) social behavioral alignment bridges intermediate-layer social features to the action head and trains on counterfactual human-object scene pairs to break visual saliency shortcuts; (2) temporal safety alignment provides automatically generated future-risk supervision to enable anticipatory collision avoidance. On SCAND and real-world deployment, SALSA reduces near-collisions by 86.4% and improves social counterfactual accuracy from 53% to 93%, demonstrating that safer social navigation can be achieved by teaching VLA policies to act on representations they already possess. These results show that pretrained VLA policies can be adapted for safer social navigation by better aligning their latent representations with action generation.
