Neural Elevation Models for Terrain Mapping and Path Planning

TL;DR

This paper presents Neural Elevation Models (NEMos), a novel terrain representation derived from imagery that enables high-quality terrain reconstruction and smooth, gradient-based path planning, improving over traditional discrete methods.

Contribution

Introduction of NEMos, a differentiable, continuous terrain model using a NeRF framework, and a gradient-based path planning algorithm leveraging this model.

Findings

NEMos accurately reconstructs terrain from imagery.

NEMos enables smoother, optimized paths.

Demonstrated effectiveness on real-world terrain data.

Abstract

This work introduces Neural Elevations Models (NEMos), which adapt Neural Radiance Fields to a 2.5D continuous and differentiable terrain model. In contrast to traditional terrain representations such as digital elevation models, NEMos can be readily generated from imagery, a low-cost data source, and provide a lightweight representation of terrain through an implicit continuous and differentiable height field. We propose a novel method for jointly training a height field and radiance field within a NeRF framework, leveraging quantile regression. Additionally, we introduce a path planning algorithm that performs gradient-based optimization of a continuous cost function for minimizing distance, slope changes, and control effort, enabled by differentiability of the height field. We perform experiments on simulated and real-world terrain imagery, demonstrating NEMos ability to generate high-quality reconstructions and produce smoother paths compared to discrete path planning methods. Future work will explore the incorporation of features and semantics into the height field, creating a generalized terrain model.