PhyPush: One Push is All You Need for Sensorless Physical Property Estimation with Physics-Guided Transformers

2026-05-25

AI Summary

Accurately estimating an object’s mass and friction coefficient using only standard robotic manipulator kinematic data—without force/torque sensors—remains challenging. This work proposes PhyPush, a novel framework that, for the first time, embeds Newton’s second law and the Coulomb friction model directly into the loss function of a Transformer architecture. By leveraging the end-effector velocity sequence generated from a single push interaction, PhyPush enables physical property estimation solely from readily available kinematic data. In simulation, the method reduces estimation error by over 10% compared to baselines that utilize full force information; in real-world experiments, it significantly outperforms purely data-driven approaches, demonstrating strong cross-domain generalization capabilities.

Abstract

Accurately estimating object mass and friction is fundamental to achieving reliable and adaptive robotic manipulation. Although interactive perception provides a powerful mechanism for inferring such properties, most existing approaches depend on specialized hardware such as force/torque sensors, tactile arrays, or multi-camera motion-capture systems, limiting scalability and deployment. This paper presents PhyPush, a physics-guided Transformer framework that estimates an object's mass and friction coefficient using only kinematically derived end-effector velocity from a single push. This typically requires data available on standard robotic arms. The model incorporates constraints from Newton's second law and the Coulomb friction model through a physics-guided loss, improving physical consistency and generalization to unseen objects and surfaces. Across diverse simulation and real-world setups, PhyPush consistently achieves more accurate mass and friction estimation in challenging out-of-domain conditions. In simulation, it reduces error by over 10% compared with a baseline that has privileged access to full force information, while in real-world experiments, it outperforms a data-driven loss approach. Overall, the results demonstrate that physics-guided learning can enable low-cost, sensor-efficient estimation of physical properties, relying solely on a single push and readily available kinematic data.

Problem

• physical property estimation
• sensorless manipulation
• mass estimation
• friction estimation
• interactive perception

Innovation

• physics-guided learning
• sensorless estimation
• Transformer
• interactive perception
• physical property estimation