VertiAdaptor: Online Kinodynamics Adaptation for Vertically Challenging Terrain

TL;DR

VertiAdaptor introduces an online adaptation framework that enhances kinodynamic modeling for autonomous vehicles navigating unpredictable off-road terrains by integrating elevation and semantic data, leading to improved accuracy and faster adaptation.

Contribution

This work presents a novel online adaptation method using neural ODE basis functions for real-time terrain-aware kinodynamic modeling in off-road autonomous driving.

Findings

Improves prediction accuracy by up to 23.9%.

Achieves 5X faster adaptation time.

Validated in simulation and on physical platform.

Abstract

Autonomous driving in off-road environments presents significant challenges due to the dynamic and unpredictable nature of unstructured terrain. Traditional kinodynamic models often struggle to generalize across diverse geometric and semantic terrain types, underscoring the need for real-time adaptation to ensure safe and reliable navigation. We propose VertiAdaptor (VA), a novel online adaptation framework that efficiently integrates elevation with semantic embeddings to enable terrain-aware kinodynamic modeling and planning via function encoders. VA learns a kinodynamic space spanned by a set of neural ordinary differential equation basis functions, capturing complex vehicle-terrain interactions across varied environments. After offline training, the proposed approach can rapidly adapt to new, unseen environments by identifying kinodynamics in the learned space through a computationally efficient least-squares calculation. We evaluate VA within the Verti-Bench simulator, built on the Chrono multi-physics engine, and validate its performance both in simulation and on a physical Verti-4-Wheeler platform. Our results demonstrate that VA improves prediction accuracy by up to 23.9% and achieves a 5X faster adaptation time, advancing the robustness and reliability of autonomous robots in complex and evolving off-road environments.