残差強化学習による視覚サーボ制御とゼロショットsim-to-real転移
Residual Reinforcement Learning for Visual Servo Control with Zero-Shot Sim-to-Real Transfer
不正確なデジタルツイン上で残差方策を強化学習し、遅延補償カルマンフィルタと組み合わせることで、コンベア上の移動物体追従誤差をベースラインの1.1-1.4mmから0.3-0.7mmへ低減し、ゼロショットで実機転移を実現した。
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: Mahdi Nobar, Alisa Rupenyan, John Lygeros, Jürg Keller
分類: cs.RO, eess.SY
原文アブストラクト
We address tracking a moving object on a conveyor with a Franka Emika Panda robot arm, despite an imprecise digital twin, measurement errors, and noise. Our novel approach augments a loop-shaped inverse-Jacobian PI controller with a residual policy learned via Soft Actor-Critic in the digital twin. A delay-compensated Kalman filter fuses out-of-sequence camera measurements to provide latency-aware moving target position estimates. We compare the achieved performance to a performance bound for a class of model-based controllers derived from a frequency-domain sensitivity analysis, quantifying the minimum achievable error imposed by uncertain kinematic model mismatch. Simulation and hardware experiments indicate that our residual policy drives tracking error below the bound. Across conveyor speeds (27.9-41.8 mm/s), our method reduces mean tracking errors from 1.1-1.4 mm (baseline inverse-Jacobian PI) to 0.3-0.7 mm. These results show that learning residual corrections in a limited-fidelity simulator, combined with delay compensation, can substantially improve model-based control under modeling and measurement uncertainties while maintaining reliable sim-to-real transfer.
