Practice Makes Policies: Bootstrapping and Consolidating Robotic Capabilities from Zero Human Demonstrations

2026-07-29

AI Summary

This work addresses the challenge of enabling general-purpose robots to autonomously evolve manipulation skills in open-world settings without human demonstrations. To this end, we propose HERO, a novel framework that achieves, for the first time, fully demonstration-free bootstrapping and continual consolidation of robotic manipulation capabilities. Built upon a hierarchical embodied intelligence architecture, HERO tightly couples heuristic reasoning, exemplar reuse, and reflexive execution to allow robots to autonomously collect experiences, transfer behaviors, and progressively crystallize them into efficient closed-loop visuomotor policies. Experimental results demonstrate that the system significantly reduces human intervention across a diverse range of real-world tasks while exhibiting robustness and sustained self-improvement.

Abstract

General-purpose robotic manipulation requires robots to perform diverse tasks in open-world environments while improving their skills over time. Despite recent progress in robotic manipulation, existing systems still primarily acquire manipulation skills in a static manner, where capabilities are learned for specific tasks or settings rather than adaptively evolving through physical interaction. Resembling how repeated practice enables humans to develop muscle memory, advanced manipulation proficiency requires an autonomous capability evolution mechanism that allows robots to progressively transform interaction experiences into increasingly effective manipulation abilities. To this end, we propose HERO, a self-improving hierarchical embodied agent that enables autonomous capability evolution from zero human demonstrations. HERO organizes heuristic reasoning, exemplar reuse, and reflexive execution into a unified orchestration framework, allowing robots to autonomously bootstrap manipulation experience, rapidly accumulate reusable behaviors through experience transfer, and progressively consolidate recurring interactions into efficient closed-loop visuomotor policies. By tightly coupling autonomous data collection with task execution, HERO continuously expands and dynamically schedules manipulation capabilities according to different stages of experience accumulation and execution requirements. Extensive experiments demonstrate that HERO substantially reduces human intervention during robotic data collection while achieving robust manipulation across diverse tasks, providing a promising path toward self-improving robotic systems.

Problem

• robotic manipulation
• autonomous capability evolution
• self-improving robotics
• open-world environments
• zero human demonstrations

Innovation

• self-improving robotics
• zero human demonstrations
• hierarchical embodied agent
• autonomous capability evolution
• visuomotor policy consolidation