# Gaussian Splatting Visual MPC for Granular Media Manipulation

arXiv: https://arxiv.org/abs/2410.09740 · HTML: https://mlsanity.com/paper/2410.09740/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Oct 2024.

## Built on by (most cited first, 1 of 1)

- [Learning Action-Conditional and Object-Centric Gaussian Splatting World Models for Rigid Objects](https://mlsanity.com/paper/2606.01950/) (2026)

## Abstract (arXiv)

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.