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

TL;DR

This paper introduces SIM1, a physics-aligned simulation system that creates high-fidelity synthetic data for deformable object manipulation, enabling effective zero-shot policy transfer from simulation to real-world tasks.

Contribution

The paper presents a novel physics-grounded simulation pipeline that digitizes scenes, calibrates deformable dynamics, and generates behaviors, significantly improving sim-to-real transfer in deformable manipulation.

Findings

Policies trained on synthetic data match real-data baselines at 1:15 ratio.

Achieves 90% zero-shot success in real-world deformable manipulation tasks.

Demonstrates 50% improvement in generalization over prior methods.

Abstract

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 rigids. 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. We posit that simulation fails not for being synthetic, but for being ungrounded. To address this, we introduce SIM1, 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 twins, calibrates deformable dynamics through elastic modeling, and expands behaviors via diffusion-based trajectory generation with quality filtering. This pipeline transforms sparse observations into scaled synthetic supervision with near-demonstration fidelity. Experiments show that policies trained on purely synthetic data achieve parity with real-data baselines at a 1:15 equivalence ratio, while delivering 90% zero-shot success and 50% generalization gains in real-world deployment. These results validate physics-aligned simulation as scalable supervision for deformable manipulation and a practical pathway for data-efficient policy learning.