Walk Like Dogs: Learning Steerable Imitation Controllers for Legged Robots from Unlabeled Motion Data

TL;DR

This paper introduces a framework for learning steerable, stylistically consistent quadrupedal locomotion controllers from unlabeled real-world motion data, enabling user-directed gait transitions without manual annotations.

Contribution

It presents a novel imitation learning approach that automatically discovers behavioral modes and maps user commands to generate realistic, style-preserving locomotion on robots.

Findings

Controller trained on dog data exhibits natural gait patterns.

Emergent gait transitions in response to velocity commands.

No manual labeling or predefined modes needed.

Abstract

We present an imitation learning framework that extracts distinctive legged locomotion behaviors and transitions between them from unlabeled real-world motion data. By automatically discovering behavioral modes and mapping user steering commands to them, the framework enables user-steerable and stylistically consistent motion imitation. Our approach first bridges the morphological and physical gap between the motion source and the robot by transforming raw data into a physically consistent, robot-compatible dataset using a kino-dynamic motion retargeting strategy. This data is used to train a steerable motion synthesis module that generates stylistic, multi-modal kinematic targets from high-level user commands. These targets serve as a reference for a reinforcement learning controller, which reliably executes them on the robot hardware. In our experiments, a controller trained on dog motion data demonstrated distinctive quadrupedal gait patterns and emergent gait transitions in response to varying velocity commands. These behaviors were achieved without manual labeling, predefined mode counts, or explicit switching rules, maintaining the stylistic coherence of the data.