コンテキストは現実ギャップを埋められるか?コンテキスト認識ポリシーのSim-to-Real転送
Can Context Bridge the Reality Gap? Sim-to-Real Transfer of Context-Aware Policies
シミュレーションから実機への転送において、動力学パラメータの推定値(コンテキスト)をポリシーに条件付けることで性能が向上するかを検証し、実機の押し操作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: Marco Iannotta, Yuxuan Yang, Johannes A. Stork, Erik Schaffernicht, Todor Stoyanov
分類: cs.RO
原文アブストラクト
Sim-to-real transfer remains a major challenge in reinforcement learning (RL) for robotics, as policies trained in simulation often fail to generalize to the real world due to discrepancies in environment dynamics. Domain Randomization (DR) mitigates this issue by exposing the policy to a wide range of randomized dynamics during training, yet leading to a reduction in performance. While standard approaches typically train policies agnostic to these variations, we investigate whether sim-to-real transfer can be improved by conditioning the policy on an estimate of the dynamics parameters -- referred to as context. To this end, we integrate a context estimation module into a DR-based RL framework and systematically compare SOTA supervision strategies. We evaluate the resulting context-aware policies in both a canonical control benchmark and a real-world pushing task using a Franka Emika Panda robot. Results show that context-aware policies outperform the context-agnostic baseline across all settings, although the best supervision strategy depends on the task.
