Dynamic-Resolution Model Learning for Object Pile Manipulation

TL;DR

This paper introduces a dynamic-resolution particle representation for learning adaptive scene models in robotic object pile manipulation, improving efficiency and effectiveness over fixed-resolution methods in simulation and real-world tasks.

Contribution

It proposes a unified graph neural network-based dynamics model that adaptively selects scene resolution during manipulation tasks, optimizing performance across simple and complex scenarios.

Findings

Outperforms fixed-resolution baselines in object gathering, sorting, and redistribution tasks.

Effective in both simulation and real-world environments with various granular objects.

Achieves better efficiency and effectiveness in manipulation tasks.

Abstract

Dynamics models learned from visual observations have shown to be effective in various robotic manipulation tasks. One of the key questions for learning such dynamics models is what scene representation to use. Prior works typically assume representation at a fixed dimension or resolution, which may be inefficient for simple tasks and ineffective for more complicated tasks. In this work, we investigate how to learn dynamic and adaptive representations at different levels of abstraction to achieve the optimal trade-off between efficiency and effectiveness. Specifically, we construct dynamic-resolution particle representations of the environment and learn a unified dynamics model using graph neural networks (GNNs) that allows continuous selection of the abstraction level. During test time, the agent can adaptively determine the optimal resolution at each model-predictive control (MPC) step. We evaluate our method in object pile manipulation, a task we commonly encounter in cooking, agriculture, manufacturing, and pharmaceutical applications. Through comprehensive evaluations both in the simulation and the real world, we show that our method achieves significantly better performance than state-of-the-art fixed-resolution baselines at the gathering, sorting, and redistribution of granular object piles made with various instances like coffee beans, almonds, corn, etc.