Semantic 3D Grid Maps for Autonomous Driving

TL;DR

This paper emphasizes the importance of 3D semantic maps for autonomous driving, introduces a real-time hierarchical 3D grid mapping framework, and demonstrates its capabilities in complex spatial reasoning tasks.

Contribution

It presents UFOMap, a real-time hierarchical 3D grid mapping framework that supports semantic information and complex spatial functions for autonomous driving.

Findings

UFOMap operates within real-time constraints.

It efficiently supports semantic segmentation integration.

Enables calculation of occluded spaces in 3D maps.

Abstract

Maps play a key role in rapidly developing area of autonomous driving. We survey the literature for different map representations and find that while the world is three-dimensional, it is common to rely on 2D map representations in order to meet real-time constraints. We believe that high levels of situation awareness require a 3D representation as well as the inclusion of semantic information. We demonstrate that our recently presented hierarchical 3D grid mapping framework UFOMap meets the real-time constraints. Furthermore, we show how it can be used to efficiently support more complex functions such as calculating the occluded parts of space and accumulating the output from a semantic segmentation network.