学習と制御のためのソフト勾配による硬接触の微分可能シミュレーション
Differentiable Simulation of Hard Contacts with Soft Gradients for Learning and Control
ペナルティベースのシミュレータにおける硬接触時の勾配劣化を分析し、適応的時間積分と接触距離推定を組み合わせて勾配精度を改善するDiffMJXを提案した。
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: Anselm Paulus, A. René Geist, Pierre Schumacher, Vít Musil, Simon Rappenecker, Georg Martius
分類: cs.RO, cs.LG, cs.SY, eess.SY
原文アブストラクト
Contact forces introduce discontinuities into robot dynamics that severely limit the use of simulators for gradient-based optimization. Penalty-based simulators such as MuJoCo, soften contact resolution to enable gradient computation. However, realistically simulating hard contacts requires stiff solver settings, which leads to incorrect simulator gradients when using automatic differentiation. Contrarily, using non-stiff settings strongly increases the sim-to-real gap. We analyze penalty-based simulators to pinpoint why gradients degrade under hard contacts. Building on these insights, we propose DiffMJX, which couples adaptive time integration with penalty-based simulation to substantially improve gradient accuracy. A second challenge is that contact gradients vanish when bodies separate. To address this, we introduce contacts from distance (CFD) which combines penalty-based simulation with straight-through estimation. By applying CFD exclusively in the backward pass, we obtain informative pre-contact gradients while retaining physical realism.
