Diffusion Trajectory-guided Policy for Long-horizon Robot Manipulation

TL;DR

This paper introduces DTP, a diffusion-based trajectory guidance framework that enhances long-horizon robot manipulation by reducing errors and improving success rates, outperforming existing methods without external pretraining.

Contribution

The paper proposes a novel diffusion trajectory-guided policy framework that leverages generative models for trajectory creation to improve long-horizon robot imitation learning.

Findings

DTP outperforms state-of-the-art baselines by 25% in success rate on CALVIN.

DTP improves real-world robot performance significantly.

The two-stage approach effectively reduces error accumulation in long-horizon tasks.

Abstract

Recently, Vision-Language-Action models (VLA) have advanced robot imitation learning, but high data collection costs and limited demonstrations hinder generalization and current imitation learning methods struggle in out-of-distribution scenarios, especially for long-horizon tasks. A key challenge is how to mitigate compounding errors in imitation learning, which lead to cascading failures over extended trajectories. To address these challenges, we propose the Diffusion Trajectory-guided Policy (DTP) framework, which generates 2D trajectories through a diffusion model to guide policy learning for long-horizon tasks. By leveraging task-relevant trajectories, DTP provides trajectory-level guidance to reduce error accumulation. Our two-stage approach first trains a generative vision-language model to create diffusion-based trajectories, then refines the imitation policy using them. Experiments on the CALVIN benchmark show that DTP outperforms state-of-the-art baselines by 25% in success rate, starting from scratch without external pretraining. Moreover, DTP significantly improves real-world robot performance.