PhysCoRe: Physics-Corrected Residual World Models for Material-Aware Deformable Dynamics
cs.RO, cs.CV, cs.LG
Submitted: 2026-07-22
Updated: 2026-09-16
Comments: Website: https://lunarlab-gatech.github.io/PhysCoRe-website/
Project page: https://lunarlab-gatech.github.io/PhysCoRe-website
License: http://creativecommons.org/licenses/by/4.0/
The gist: Predicting how deformable objects evolve under robotic manipulation is a longstanding challenge.
Terminology
Abstract
Predicting how deformable objects evolve under robotic manipulation is a longstanding challenge. Existing approaches typically rely on per-object optimization to fit material parameters, which can be slow and cannot generalize, while end-to-end learned alternatives extrapolate poorly and often violate basic physical structure. We present PhysCoRe, a physics-corrected residual world model that couples a differentiable Material Point Method (MPM) simulator with two feed-forward neural networks. A material refinement module, Material from Motion (MfM), infers per-particle elasticity from visual observations, grounding the simulator in object-specific physics. A residual correction module, Residual from Dynamics (RfD), learns the discrepancy and predicts corrections to the simulator's internal dynamics, absorbing systematic biases that the analytical model cannot capture. This design also supports online material identification on novel objects. MfM adapts from limited interactions, and its predictive uncertainty steers further exploration toward the regions where its estimate is least confident. Experiments on real deformable-object manipulation sequences show that PhysCoRe outperforms state-of-the-art baselines in prediction accuracy, and that its predicted confidence forms a reliable distribution across the object's geometry, providing a natural signal for future confidence-guided exploration.
Sources
- Physics3D: Learning Physical Properties of 3D Gaussians via Video Diffusion
- Learning Equivariant Neural-Augmented Object Dynamics From Few Interactions
- PlayWorld: Learning Robot World Models from Autonomous Play
- Scaling Cross-Embodiment World Models for Dexterous Manipulation
- Real-is-Sim: Bridging the Sim-to-Real Gap with a Dynamic Digital Twin
- Learning Physics-Guided Residual Dynamics for Deformable Object Simulation
- Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
- Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving