Humanoid-DART: Humanoid Loco-Manipulation using Diffusion-guided Augmentation through Relabeling and Tracking
For humanoid robotics researchers, this work reduces the need for costly human demonstrations and intervention in learning loco-manipulation policies.
The paper presents a self-supervised framework that bootstraps from sparse demonstrations to learn goal-conditioned humanoid loco-manipulation policies, combining diffusion-based trajectory generation with reinforcement learning. The method achieves effective performance across multiple skills with minimal expert supervision.
Imitating human demonstrations has emerged as a dominant paradigm for learning humanoid loco-manipulation policies. However, scaling these approaches remains challenging due to the high cost of collecting diverse demonstrations and the need for continual human intervention to correct policy failures. In this paper, we present a self-supervised framework that bootstraps from sparse demonstrations and progressively expands its behavioral repertoire, enabling the learning of a goal-conditioned policy that automatically explores the goal space with minimal expert supervision. Our approach combines diffusion-based trajectory generation with reinforcement learning, where the latter is used to track goal-conditioned trajectories produced by the diffusion model for a range of loco-manipulation skills. Through extensive ablation studies and comparisons with state-of-the-art methods, we demonstrate the effectiveness of our framework on multiple humanoid loco-manipulation skills.