Gaussian Splatting Visual MPC for Granular Media Manipulation

TL;DR

This paper introduces a Gaussian splatting-based visual dynamics model for granular media manipulation, enabling efficient planning and generalization in complex, particle-based environments using Model-Predictive Control.

Contribution

It presents a novel visual dynamics modeling approach over Gaussian splatting representations specifically designed for granular media, improving manipulation and generalization capabilities.

Findings

Effective in simulated and real-world granular manipulation tasks

Outperforms existing methods in prediction accuracy and manipulation success

Demonstrates zero-shot transfer to new environments

Abstract

Recent advancements in learned 3D representations have enabled significant progress in solving complex robotic manipulation tasks, particularly for rigid-body objects. However, manipulating granular materials such as beans, nuts, and rice, remains challenging due to the intricate physics of particle interactions, high-dimensional and partially observable state, inability to visually track individual particles in a pile, and the computational demands of accurate dynamics prediction. Current deep latent dynamics models often struggle to generalize in granular material manipulation due to a lack of inductive biases. In this work, we propose a novel approach that learns a visual dynamics model over Gaussian splatting representations of scenes and leverages this model for manipulating granular media via Model-Predictive Control. Our method enables efficient optimization for complex manipulation tasks on piles of granular media. We evaluate our approach in both simulated and real-world settings, demonstrating its ability to solve unseen planning tasks and generalize to new environments in a zero-shot transfer. We also show significant prediction and manipulation performance improvements compared to existing granular media manipulation methods.