Discovering Unsupervised Behaviours from Full-State Trajectories

TL;DR

This paper introduces an autonomous Quality-Diversity algorithm that discovers behavioral characterizations from full-state trajectories, enabling robots to autonomously identify diverse skills without prior task knowledge.

Contribution

The work presents a novel method for autonomous behavioral discovery in robots, removing the need for hand-coded descriptors in Quality-Diversity algorithms.

Findings

The algorithm autonomously finds diverse movement policies.

It discovers varied ways for robots to utilize their legs.

It identifies multiple behaviors including navigation and half-rolls.

Abstract

Improving open-ended learning capabilities is a promising approach to enable robots to face the unbounded complexity of the real-world. Among existing methods, the ability of Quality-Diversity algorithms to generate large collections of diverse and high-performing skills is instrumental in this context. However, most of those algorithms rely on a hand-coded behavioural descriptor to characterise the diversity, hence requiring prior knowledge about the considered tasks. In this work, we propose an additional analysis of Autonomous Robots Realising their Abilities; a Quality-Diversity algorithm that autonomously finds behavioural characterisations. We evaluate this approach on a simulated robotic environment, where the robot has to autonomously discover its abilities from its full-state trajectories. All algorithms were applied to three tasks: navigation, moving forward with a high velocity, and performing half-rolls. The experimental results show that the algorithm under study discovers autonomously collections of solutions that are diverse with respect to all tasks. More specifically, the analysed approach autonomously finds policies that make the robot move to diverse positions, but also utilise its legs in diverse ways, and even perform half-rolls.