pith. sign in

arxiv: 2410.22325 · v2 · pith:PIRXM6NWnew · submitted 2024-10-29 · 💻 cs.RO · cs.AI· cs.CV

Robots Pre-train Robots: Manipulation-Centric Robotic Representation from Large-Scale Robot Datasets

classification 💻 cs.RO cs.AIcs.CV
keywords manipulationrobotictasksvisualrepresentationrobotactionscentricity
0
0 comments X
read the original abstract

The pre-training of visual representations has enhanced the efficiency of robot learning. Due to the lack of large-scale in-domain robotic datasets, prior works utilize in-the-wild human videos to pre-train robotic visual representation. Despite their promising results, representations from human videos are inevitably subject to distribution shifts and lack the dynamics information crucial for task completion. We first evaluate various pre-trained representations in terms of their correlation to the downstream robotic manipulation tasks (i.e., manipulation centricity). Interestingly, we find that the "manipulation centricity" is a strong indicator of success rates when applied to downstream tasks. Drawing from these findings, we propose Manipulation Centric Representation (MCR), a foundation representation learning framework capturing both visual features and the dynamics information such as actions and proprioceptions of manipulation tasks to improve manipulation centricity. Specifically, we pre-train a visual encoder on the DROID robotic dataset and leverage motion-relevant data such as robot proprioceptive states and actions. We introduce a novel contrastive loss that aligns visual observations with the robot's proprioceptive state-action dynamics, combined with a behavior cloning (BC)-like actor loss to predict actions during pre-training, along with a time contrastive loss. Empirical results across 4 simulation domains with 20 tasks verify that MCR outperforms the strongest baseline method by 14.8%. Moreover, MCR boosts the performance of data-efficient learning with a UR5e arm on 3 real-world tasks by 76.9%. Project website: https://robots-pretrain-robots.github.io/.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Contrastive Action-Image Pre-training for Visuomotor Control

    cs.RO 2026-06 unverdicted novelty 6.0

    CAIP learns action-aligned visual representations via contrastive pre-training on human hand keypoints from egocentric video, outperforming DINOv2, SigLIP, MVP, and R3M with >30% gains on real dexterous manipulation tasks.

  2. Action-Effect Memory Pretraining for Robot Manipulation

    cs.RO 2026-06 unverdicted novelty 6.0

    AEM pretrains compact history representations via masked modeling on interleaved vision-action sequences to boost downstream robot manipulation in simulation and real settings.

  3. TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies

    cs.RO 2024-12 conditional novelty 6.0

    Visual trace prompting improves spatial-temporal awareness in VLA models, delivering 10% gains on SimplerEnv and 3.5x on real-robot tasks.

  4. ReFineVLA: Multimodal Reasoning-Aware Generalist Robotic Policies via Teacher-Guided Fine-Tuning

    cs.RO 2026-04 unverdicted novelty 5.0

    ReFineVLA adds teacher-generated reasoning steps to VLA training and reports state-of-the-art success rates on SimplerEnv WidowX and Google Robot benchmarks.

  5. A Visual Reinforcement Learning-Based Separate Primitive Policy for Peg-in-Hole Tasks

    cs.RO 2025-04 unverdicted novelty 5.0

    S2P learns separate location and insertion primitives simultaneously via visual RL for peg-in-hole tasks, improving sample efficiency and success rates across polygon benchmarks in simulation and real-world tests.

  6. Multimodal Fusion for Sim2real Transfer in Visual Reinforcement Learning

    cs.CV 2025-07 unverdicted novelty 4.0

    A multimodal RGB-depth fusion backbone with vision transformer, masked-token contrastive learning, and curriculum domain randomization outperforms baselines in simulation and enables zero-shot real-world robot manipulation.