PGTT: Phase-Guided Terrain Traversal for Perceptive Legged Locomotion

TL;DR

This paper introduces PGTT, a perception-aware deep reinforcement learning approach for legged robots that improves terrain traversal by reducing bias and supporting various robot morphologies, validated in simulation and on real robots.

Contribution

PGTT is a novel perception-aware RL method that encodes gait phase with reward shaping, enabling robust, morphology-agnostic legged locomotion across diverse terrains.

Findings

PGTT outperforms baselines in success rate under disturbances and obstacles.

PGTT converges twice as fast as end-to-end baselines.

Successful real-world deployment on Unitree Go2 and preliminary tests on ANYmal-C.

Abstract

State-of-the-art perceptive Reinforcement Learning controllers for legged robots either (i) impose oscillator or IK-based gait priors that constrain the action space, add bias to the policy optimization and reduce adaptability across robot morphologies, or (ii) operate "blind", which struggle to anticipate hind-leg terrain, and are brittle to noise. In this paper, we propose Phase-Guided Terrain Traversal (PGTT), a perception-aware deep-RL approach that overcomes these limitations by enforcing gait structure purely through reward shaping, thereby reducing inductive bias in policy learning compared to oscillator/IK-conditioned action priors. PGTT encodes per-leg phase as a cubic Hermite spline that adapts swing height to local heightmap statistics and adds a swing-phase contact penalty, while the policy acts directly in joint space supporting morphology-agnostic deployment. Trained in MuJoCo (MJX) on procedurally generated stair-like terrains with curriculum and domain randomization, PGTT achieves the highest success under push disturbances (median +7.5% vs. the next best method) and on discrete obstacles (+9%), with comparable velocity tracking, and converging to an effective policy roughly 2x faster than strong end-to-end baselines. We validate PGTT on a Unitree Go2 using a real-time LiDAR elevation-to-heightmap pipeline, and we report preliminary results on ANYmal-C obtained with the same hyperparameters. These findings indicate that terrain-adaptive, phase-guided reward shaping is a simple and general mechanism for robust perceptive locomotion across platforms.