Human Imitated Bipedal Locomotion with Frequency Based Gait Generator Network

TL;DR

This paper introduces a lightweight framework combining a human motion-based gait generator with reinforcement learning, enabling robust and natural bipedal walking across various terrains with minimal training.

Contribution

It presents a novel integration of spectral motion priors with DRL for efficient, generalizable bipedal locomotion control.

Findings

Policies trained on flat terrain generalize to rough surfaces

The framework achieves natural-looking gait with modest training

Pairing spectral priors with DRL improves robustness

Abstract

Learning human-like, robust bipedal walking remains difficult due to hybrid dynamics and terrain variability. We propose a lightweight framework that combines a gait generator network learned from human motion with Proximal Policy Optimization (PPO) controller for torque control. Despite being trained only on flat or mildly sloped ground, the learned policies generalize to steeper ramps and rough surfaces. Results suggest that pairing spectral motion priors with Deep Reinforcement Learning (DRL) offers a practical path toward natural and robust bipedal locomotion with modest training cost.