拡散モデルを用いた動的ニューラルクープマン蒸留によるリアルタイムロボット制御
Dynamic Neural Koopman Distillation for Real-Time Robot Control Using Diffusion Models
拡散モデルの反復ノイズ除去による遅延を解消するため、多段階推論を単一パスに蒸留しつつ多様性を保つフレームワークを提案し、D4RLベンチマークと実機で高速な閉ループ制御を実証した。
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: Lei Zheng, Peiqi Yu, Zengqi Peng, Changliu Liu, Armin Lederer
分類: cs.RO
原文アブストラクト
Diffusion models excel at generating diverse and multimodal trajectories for robotic planning, yet their iterative denoising process introduces latency that is incompatible with high-frequency closed-loop control. To address this problem, we propose Dynamic Neural Koopman Distillation, a framework that distills multistep diffusion inference into a single forward pass while retaining the multimodal expressivity of the teacher model. Specifically, we introduce a Factorized Dynamic Koopman layer that models the denoising process through a factorized latent transition with state-dependent modal gains. We evaluate the proposed method on standard D4RL MuJoCo locomotion benchmarks and a physical Kinova manipulator, comparing against one-step baselines. The results show that our method significantly outperforms existing one-step distillation approaches on the reported locomotion tasks, and reduces the inference latency to the millisecond regime compared with the teacher policy. Hardware experiments further demonstrate that our method enables smooth and fast closed-loop execution while maintaining task success and comparable accuracy. A project page is available at https://fdkoopman.github.io/.
