DNAct: Diffusion Guided Multi-Task 3D Policy Learning

TL;DR

DNAct is a multi-task 3D policy learning framework that combines neural rendering and diffusion training to enhance semantic understanding and robustness in robotic tasks, achieving over 30% success rate improvement.

Contribution

The paper introduces DNAct, integrating neural rendering pre-training with diffusion training for multi-modality learning in 3D action spaces, advancing robotic multi-task policy generalization.

Findings

Over 30% improvement in success rate over SOTA NeRF-based methods

Effective distillation of 2D semantics into 3D space using foundation models

Enhanced robustness and generalization in multi-task robotic manipulation

Abstract

This paper presents DNAct, a language-conditioned multi-task policy framework that integrates neural rendering pre-training and diffusion training to enforce multi-modality learning in action sequence spaces. To learn a generalizable multi-task policy with few demonstrations, the pre-training phase of DNAct leverages neural rendering to distill 2D semantic features from foundation models such as Stable Diffusion to a 3D space, which provides a comprehensive semantic understanding regarding the scene. Consequently, it allows various applications to challenging robotic tasks requiring rich 3D semantics and accurate geometry. Furthermore, we introduce a novel approach utilizing diffusion training to learn a vision and language feature that encapsulates the inherent multi-modality in the multi-task demonstrations. By reconstructing the action sequences from different tasks via the diffusion process, the model is capable of distinguishing different modalities and thus improving the robustness and the generalizability of the learned representation. DNAct significantly surpasses SOTA NeRF-based multi-task manipulation approaches with over 30% improvement in success rate. Project website: dnact.github.io.