Adaptive Policy Switching of Two-Wheeled Differential Robots for Traversing over Diverse Terrains

TL;DR

This paper presents a method for two-wheeled robots to adaptively switch policies based on terrain type, using short-term orientation data to classify terrain with high accuracy, enabling autonomous traversal of diverse terrains.

Contribution

The study demonstrates effective terrain classification using posture data, facilitating adaptive policy switching for robots in complex environments.

Findings

Orientation data can distinguish terrain types with over 98% accuracy.

Short-term posture observations are sufficient for reliable terrain estimation.

Adaptive policy switching improves robot traversal over diverse terrains.

Abstract

Exploring lunar lava tubes requires robots to traverse without human intervention. Because pre-trained policies cannot fully cover all possible terrain conditions, our goal is to enable adaptive policy switching, where the robot selects an appropriate terrain-specialized model based on its current terrain features. This study investigates whether terrain types can be estimated effectively using posture-related observations collected during navigation. We fine-tuned a pre-trained policy using Proximal Policy Optimization (PPO), and then collected the robot's 3D orientation data as it moved across flat and rough terrain in a simulated lava-tube environment. Our analysis revealed that the standard deviation of the robot's pitch data shows a clear difference between these two terrain types. Using Gaussian mixture models (GMM), we evaluated terrain classification across various window sizes. An accuracy of more than 98% was achieved when using a 70-step window. The result suggests that short-term orientation data are sufficient for reliable terrain estimation, providing a foundation for adaptive policy switching.