Robot system arranges objects using fast physics simulations
Test-Time Spatial Reasoning for Robot Manipulation Using Generative Real-to-Sim
RoboticsComputer Vision and Pattern Recognition
Summary
Robots need to understand how objects fit and interact to complete tasks like building towers. The authors created Simify, which lets a robot figure out how to arrange objects from a single picture by testing thousands of possibilities in a virtual simulation very quickly. Simify uses 3D models and language tools to turn real-world images into simulated scenes. Then it tries many object layouts to find the best one for a given goal, like making the tallest tower. Tests with real robots show that this approach works well even with objects the robot hasn't seen before.
What this means in practice
- •For robotics engineers: Enable robots to complete complex rearrangement tasks with new objects by quickly testing object setups in simulation.
- •For automation system developers: Improve automation workflows by integrating fast test-time spatial reasoning to optimize multi-object arrangements.
Authors
Ivan Kapelyukh, Yafei Hu, Ran Gong, Brandon May, Tushar Kusnur, Laura Herlant, Karl Schmeckpeper, Edward Johns, Xiaohan Zhang
Abstract
Spatial reasoning is fundamental to general robot intelligence, as it enables robots to complete long-horizon tasks involving multi-object interaction. We introduce Simify, a training-free, test-time framework that performs explicit spatial reasoning via massively parallel physics simulation. From a single RGB-D image of a scene, Simify reconstructs simulation-ready assets leveraging 3D generative models and vision-language models. Then given a task specified by a reward function (e.g., build the tallest tower), Simify launches thousands of parallel rollouts in simulation and performs an evolutionary search to optimize object arrangements, typically converging within seconds. We conduct quantitative experiments on real-robot hardware to demonstrate the ability of our framework to execute complex object rearrangement tasks end-to-end with previously unseen objects. Results show that our framework outperforms prior work on foundation models for spatial reasoning by effectively exploiting large-scale parallel simulation during inference, and also highlight the importance of complete and accurate geometry for successful sim-to-real transfer.