Mixture of Flow Matching and Particle Filter for Proprioception-Based Online Contact Estimation

概要

In this article, we propose ContactFlowFilter, a novel approach for contact estimation that determines the number of active contacts, their locations, and the associated force vectors. We consider a robot manipulator equipped with joint torque sensors and a base force/torque sensor; such a proprioceptive-only setup presents a challenging inverse problem characterized by (i) multi-modality from non-unique solutions, (ii) temporal dependencies in state evolution, and (iii) discrete contact events involving contact addition/removal. Existing approaches, such as particle filters (PFs) and learning-based methods, often struggle to address these challenges simultaneously. To overcome this, ContactFlowFilter integrates a generative model into the PF framework. The recursive structure of the PF captures temporal dependencies, while the flow matching model serves as an auxiliary proposal distribution to track multi-modal beliefs and detect contact events. Extensive validation in both simulation and real-world experiments confirms the effectiveness of the proposed method. In real-world settings, ContactFlowFilter achieves the lowest dissimilarity from the target distribution, with errors of only 0.36 cm and 0.88 cm for single- and dual-contact scenarios, respectively, while operating at 0.69 ms and 2.69 ms.

参考リンク

• Paper ↗

リンク先のライセンス・再現性・データ条件をご確認ください。自動分類や実装の目安は保証ではありません。

実装を始めるプロンプト

参考リンクを確認したうえで、次の内容をコピーして利用できます。

# 論文・記事
Mixture of Flow Matching and Particle Filter for Proprioception-Based Online Contact Estimation

# 概要
In this article, we propose ContactFlowFilter, a novel approach for contact estimation that determines the number of active contacts, their locations, and the associated force vectors. We consider a…

# 原典
https://arxiv.org/abs/2610.29568

上記の手法を実装してください。実行前に依存関係とライセンスを確認してください。