Learning to Localize Using a LiDAR Intensity Map

TL;DR

This paper introduces a real-time localization system for self-driving cars that uses deep embeddings of LiDAR data and intensity maps, achieving high accuracy across various sensors and environments.

Contribution

It presents a novel deep embedding approach for LiDAR-based localization that is calibration-agnostic and operates in real-time.

Findings

Operates at 15Hz in real-time

Achieves centimeter-level accuracy

Works across different LiDAR sensors and environments

Abstract

In this paper we propose a real-time, calibration-agnostic and effective localization system for self-driving cars. Our method learns to embed the online LiDAR sweeps and intensity map into a joint deep embedding space. Localization is then conducted through an efficient convolutional matching between the embeddings. Our full system can operate in real-time at 15Hz while achieving centimeter level accuracy across different LiDAR sensors and environments. Our experiments illustrate the performance of the proposed approach over a large-scale dataset consisting of over 4000km of driving.