From Tabula Rasa to Emergent Abilities: Discovering Robot Skills via Real-World Unsupervised Quality-Diversity

TL;DR

This paper introduces URSA, a novel method for autonomous real-world robot skill discovery that enables quadruped robots to learn diverse behaviors without supervision, improving adaptability and reducing human intervention.

Contribution

URSA extends QDAC to enable unsupervised, real-world skill discovery directly on robots, supporting both heuristic and fully unsupervised learning modes.

Findings

URSA successfully discovers diverse locomotion skills on a quadruped robot.

URSA outperforms baselines in damage adaptation scenarios.

The approach reduces human intervention in robot skill learning.

Abstract

Autonomous skill discovery aims to enable robots to acquire diverse behaviors without explicit supervision. Learning such behaviors directly on physical hardware remains challenging due to safety and data efficiency constraints. Existing methods, including Quality-Diversity Actor-Critic (QDAC), require manually defined skill spaces and carefully tuned heuristics, limiting real-world applicability. We propose Unsupervised Real-world Skill Acquisition (URSA), an extension of QDAC that enables robots to autonomously discover and master diverse, high-performing skills directly in the real world. We demonstrate that URSA successfully discovers diverse locomotion skills on a Unitree A1 quadruped in both simulation and the real world. Our approach supports both heuristic-driven skill discovery and fully unsupervised settings. We also show that the learned skill repertoire can be reused for downstream tasks such as real-world damage adaptation, where URSA outperforms all baselines in 5 out of 9 simulated and 3 out of 5 real-world damage scenarios. Our results establish a new framework for real-world robot learning that enables continuous skill discovery with limited human intervention, representing a significant step toward more autonomous and adaptable robotic systems. Demonstration videos are available at https://adaptive-intelligent-robotics.github.io/URSA.