Real-to-Sim Robot Policy Evaluation with Gaussian Splatting Simulation of Soft-Body Interactions
This addresses the challenge of reproducible and efficient policy evaluation for robotic manipulation with deformable objects, which is incremental by combining existing techniques like physics-informed reconstruction and Gaussian splatting for a specific application.
The paper tackles the problem of costly and non-reproducible real-world evaluation of robotic manipulation policies for deformable objects by developing a real-to-sim framework that constructs soft-body digital twins from videos and uses 3D Gaussian Splatting for photorealistic rendering. It demonstrates strong correlation between simulated and real-world performance on tasks like plush toy packing and rope routing, enabling scalable and accurate policy evaluation.
Robotic manipulation policies are advancing rapidly, but their direct evaluation in the real world remains costly, time-consuming, and difficult to reproduce, particularly for tasks involving deformable objects. Simulation provides a scalable and systematic alternative, yet existing simulators often fail to capture the coupled visual and physical complexity of soft-body interactions. We present a real-to-sim policy evaluation framework that constructs soft-body digital twins from real-world videos and renders robots, objects, and environments with photorealistic fidelity using 3D Gaussian Splatting. We validate our approach on representative deformable manipulation tasks, including plush toy packing, rope routing, and T-block pushing, demonstrating that simulated rollouts correlate strongly with real-world execution performance and reveal key behavioral patterns of learned policies. Our results suggest that combining physics-informed reconstruction with high-quality rendering enables reproducible, scalable, and accurate evaluation of robotic manipulation policies. Website: https://real2sim-eval.github.io/