AutoIntervene: アクションチャンキング模倣学習ポリシーのための校正された介入
AutoIntervene: Calibrated Intervention for Action-Chunking Imitation Learning Policies
アクションチャンキング模倣学習ポリシーがデモ分布から逸脱した際に、オペレータへの制御移行を選択的に行うオンラインフレームワークを提案。視覚・行動サポートメモリと校正された閾値で介入を制御し、実世界の双腕操作タスクで成功率を向上させた。
著者: Jinhe Tang, Weiming Zhi
分類: cs.RO, cs.AI, cs.CV, cs.HC, cs.LG
原文アブストラクト
Action-chunking visuomotor policies learn from demonstrations and improve temporal consistency by predicting short action sequences rather than single-step commands. Yet perception errors and execution drift can move the robot outside the demonstration distribution, while the policy continues to produce smooth action chunks that are inconsistent with the observed state. We present AutoIntervene, an online framework that selectively transfers control between an action-chunking policy and an operator during deployment. AutoIntervene evaluates proposed chunks against a visual-action support memory built from successful task executions, combining visual similarity with consistency between proposed and reference actions. Phase-local support governs policy-to-operator transfer within the current task phase, whereas global support governs the return to policy control after operator recovery. We calibrate separate switching thresholds for the two directions from empirical quantiles of evaluation-level scores on held-out expert demonstrations, avoiding direct manual tuning of score cutoffs. Intervention segments retained from successful rollouts target learner-induced states and provide corrective supervision for subsequent policy updates. Experiments on real-world bimanual manipulation tasks show higher post-adaptation task success and lower operator-control time than manual intervention. Videos and additional results are available at https://aus.bot/research/autointervene/.