VersualRL: 視覚的実行フィードバックによる閉ループ言語強化学習を用いたタスクレベルロボット計画
VersualRL: Closed-Loop Verbal Reinforcement Learning with Visual Execution Feedback for Task-Level Robot Planning
視覚言語モデルが実行結果を批評し、言語モデルが行動木を更新する閉ループ枠組みを提案し、実機ロボットの多段階タスクで有効性を示した。
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: Dmitrii Plotnikov, Iaroslav Kolomiets, Dmitrii Maliukov, Dmitrij Kosenkov, Daniia Zinniatullina, Artem Trandofilov, Georgii Gazaryan, Kirill Bogatikov, Timofei Kozlov, Mikhail Konenkov, Miguel Altamirano Cabrera, Dzmitry Tsetserukou
分類: cs.RO
原文アブストラクト
We introduce VersualRL, a closed-loop framework for task-level robot planning that uses visual execution feedback to iteratively refine executable Behavior Trees through structured natural-language critique. VersualRL combines a vision-language model critic, which analyzes visual observations and Behavior Tree execution traces, with a large language model actor that performs discrete and interpretable policy updates. During physical deployment, both foundation models remain frozen, while the executed Behavior Tree is updated at the symbolic level without online gradient-based policy optimization. This enables transparent reasoning, explicit causal feedback, and human-interpretable policy evolution. We validate VersualRL on a real mobile robot performing a multi-stage navigation and manipulation task under execution uncertainty. Experimental results show that the framework supports explainable policy improvements, closed-loop adaptation to execution failures, and successful deployment on physical robotic systems.
