Improving Map Re-localization with Deep 'Movable' Objects Segmentation on 3D LiDAR Point Clouds

TL;DR

This paper introduces a deep learning method to segment movable objects in 3D LiDAR point clouds, enabling the creation of more durable maps that improve re-localization accuracy in dynamic environments like parking lots.

Contribution

The authors develop a novel deep learning architecture for segmenting movable objects in 3D LiDAR data, enhancing map longevity and re-localization robustness in dynamic scenes.

Findings

Significant reduction in localization errors when using filtered maps.

Consistent improvement in re-localization accuracy over non-filtered maps.

Effective segmentation of movable objects in cluttered, dynamic environments.

Abstract

Localization and Mapping is an essential component to enable Autonomous Vehicles navigation, and requires an accuracy exceeding that of commercial GPS-based systems. Current odometry and mapping algorithms are able to provide this accurate information. However, the lack of robustness of these algorithms against dynamic obstacles and environmental changes, even for short time periods, forces the generation of new maps on every session without taking advantage of previously obtained ones. In this paper we propose the use of a deep learning architecture to segment movable objects from 3D LiDAR point clouds in order to obtain longer-lasting 3D maps. This will in turn allow for better, faster and more accurate re-localization and trajectoy estimation on subsequent days. We show the effectiveness of our approach in a very dynamic and cluttered scenario, a supermarket parking lot. For that, we record several sequences on different days and compare localization errors with and without our movable objects segmentation method. Results show that we are able to accurately re-locate over a filtered map, consistently reducing trajectory errors between an average of 35.1% with respect to a non-filtered map version and of 47.9% with respect to a standalone map created on the current session.