Online Adaptation of Terrain-Aware Dynamics for Planning in Unstructured Environments

TL;DR

This paper introduces an online, neural network-based method for robots to adapt their dynamics models to new terrains in real-time, improving navigation safety and accuracy in unstructured environments.

Contribution

The authors propose a novel online adaptation technique using neural basis functions that enables rapid terrain-aware dynamics modeling without retraining.

Findings

Fewer collisions in cluttered environments compared to neural ODE baseline.

Improved navigation accuracy due to better dynamics modeling.

Effective real-time adaptation in a Unity-based simulator.

Abstract

Autonomous mobile robots operating in remote, unstructured environments must adapt to new, unpredictable terrains that can change rapidly during operation. In such scenarios, a critical challenge becomes estimating the robot's dynamics on changing terrain in order to enable reliable, accurate navigation and planning. We present a novel online adaptation approach for terrain-aware dynamics modeling and planning using function encoders. Our approach efficiently adapts to new terrains at runtime using limited online data without retraining or fine-tuning. By learning a set of neural network basis functions that span the robot dynamics on diverse terrains, we enable rapid online adaptation to new, unseen terrains and environments as a simple least-squares calculation. We demonstrate our approach for terrain adaptation in a Unity-based robotics simulator and show that the downstream controller has better empirical performance due to higher accuracy of the learned model. This leads to fewer collisions with obstacles while navigating in cluttered environments as compared to a neural ODE baseline.