Reconstruct, Practice, Go Real: Guided Self-Improvement for Embodied Agents
Automatically prepared brief
The RPG framework enables robots to autonomously improve their execution without updating model weights by reconstructing manipulation capabilities, practicing in simulation, and refining skills based on feedback.
The research question addresses how to build reliable robot capabilities across diverse tasks with minimal human intervention, focusing on reducing the need for manual skill development and reward design.
The proposed approach, RPG, involves identifying manipulation capabilities from offline datasets, constructing practice tasks in simulation, and using feedback to diagnose failures, develop new skills, and refine existing ones.
The authors report that RPG improved task success from 28.6% to 95.0% over 15 practice rounds on 22 manipulation tasks, outperforming other baselines. After calibration, the system succeeded in all 30 physical trials.
About this brief
This article summarizes a public preprint abstract. Its findings have not been independently reproduced for this publication. Consult the paper for methods, evaluation scope and limitations. Automated briefs may contain errors; the original source takes precedence.