Grounding Graph Network Simulators using Physical Sensor Observations

TL;DR

This paper introduces a method to incorporate real-world sensory data, specifically point clouds, into graph network simulators to improve long-term prediction accuracy of deformable object simulations.

Contribution

It presents a novel approach that grounds graph network simulators on real sensory observations, enabling more accurate and stable long-term predictions in physical simulations.

Findings

Enhanced long-term prediction accuracy with sensory grounding.

Successful integration of point cloud data into mesh-based simulations.

Improved stability in simulations under uncertain material properties.

Abstract

Physical simulations that accurately model reality are crucial for many engineering disciplines such as mechanical engineering and robotic motion planning. In recent years, learned Graph Network Simulators produced accurate mesh-based simulations while requiring only a fraction of the computational cost of traditional simulators. Yet, the resulting predictors are confined to learning from data generated by existing mesh-based simulators and thus cannot include real world sensory information such as point cloud data. As these predictors have to simulate complex physical systems from only an initial state, they exhibit a high error accumulation for long-term predictions. In this work, we integrate sensory information to ground Graph Network Simulators on real world observations. In particular, we predict the mesh state of deformable objects by utilizing point cloud data. The resulting model allows for accurate predictions over longer time horizons, even under uncertainties in the simulation, such as unknown material properties. Since point clouds are usually not available for every time step, especially in online settings, we employ an imputation-based model. The model can make use of such additional information only when provided, and resorts to a standard Graph Network Simulator, otherwise. We experimentally validate our approach on a suite of prediction tasks for mesh-based interactions between soft and rigid bodies. Our method results in utilization of additional point cloud information to accurately predict stable simulations where existing Graph Network Simulators fail.