Subgoal Diffuser: Coarse-to-fine Subgoal Generation to Guide Model Predictive Control for Robot Manipulation

TL;DR

Subgoal Diffuser enhances long-horizon robot manipulation by dynamically generating coarse-to-fine subgoals, guiding MPC to overcome local minima and improve planning performance in complex tasks.

Contribution

Introduces a diffusion-based approach for dynamic subgoal generation that improves long-horizon manipulation planning with MPC.

Findings

Improves planning performance in robot manipulation tasks.

Outperforms prior diffusion-based subgoal methods.

Adapts subgoal density based on task difficulty and reachability.

Abstract

Manipulation of articulated and deformable objects can be difficult due to their compliant and under-actuated nature. Unexpected disturbances can cause the object to deviate from a predicted state, making it necessary to use Model-Predictive Control (MPC) methods to plan motion. However, these methods need a short planning horizon to be practical. Thus, MPC is ill-suited for long-horizon manipulation tasks due to local minima. In this paper, we present a diffusion-based method that guides an MPC method to accomplish long-horizon manipulation tasks by dynamically specifying sequences of subgoals for the MPC to follow. Our method, called Subgoal Diffuser, generates subgoals in a coarse-to-fine manner, producing sparse subgoals when the task is easily accomplished by MPC and more dense subgoals when the MPC method needs more guidance. The density of subgoals is determined dynamically based on a learned estimate of reachability, and subgoals are distributed to focus on challenging parts of the task. We evaluate our method on two robot manipulation tasks and find it improves the planning performance of an MPC method, and also outperforms prior diffusion-based methods.