SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable Object Manipulation

Paper Overview

Field: Computer Vision / Embodied Learning Authors: Yunsong Zhou, Hangxu Liu, Xuekun Jiang Published: 2025-04-10 arXiv: 2504.07903

Summary

Robotic manipulation with deformable objects represents a data-intensive regime in embodied learning, where shape, contact, and topology co-evolve in ways that far exceed the variability of rigid objects. Although simulation promises relief from the cost of real-world data acquisition, prevailing sim-to-real pipelines remain rooted in rigid-body abstractions, producing mismatched geometry, fragile soft dynamics, and motion primitives poorly suited for cloth interaction. The authors posit that simulation fails not for being synthetic, but for being ungrounded.

Approach

SIM1 is a physics-aligned real-to-sim-to-real data engine that grounds simulation in the physical world. Given limited demonstrations, the system:

• Digitizes scenes into metric-consistent digital twins
• Calibrates deformable dynamics through elasticity modeling
• Scales behavior via diffusion-based trajectory generation with quality filtering

This pipeline converts sparse observations into large-scale synthetic supervision with near-demonstration fidelity.

Results

• Policies trained only on synthetic data match real-data baselines at an equivalent data ratio of 1:15
• 90% zero-shot success rate in real-world deployment
• 50% generalization gain over baselines

These results validate physics-aligned simulation as a practical path toward scalable supervision and data-efficient policy learning for deformable object manipulation.

--- *Auto-collected on 2026-04-12*