ATDN vSLAM: An all-through Deep Learning-Based Solution for Visual Simultaneous Localization and Mapping

TL;DR

This paper presents ATDN vSLAM, a modular deep learning-based visual SLAM system that integrates state-of-the-art components to achieve accurate localization and mapping for autonomous driving applications.

Contribution

It introduces a novel modular deep learning architecture for vSLAM, including a new Embedding Distance Loss, achieving competitive accuracy.

Findings

Achieved 4.4% translation error on KITTI dataset

Achieved 0.0176 deg/m rotational error on KITTI dataset

Demonstrated suitability for autonomous driving tasks

Abstract

In this paper, a novel solution is introduced for visual Simultaneous Localization and Mapping (vSLAM) that is built up of Deep Learning components. The proposed architecture is a highly modular framework in which each component offers state of the art results in their respective fields of vision-based deep learning solutions. The paper shows that with the synergic integration of these individual building blocks, a functioning and efficient all-through deep neural (ATDN) vSLAM system can be created. The Embedding Distance Loss function is introduced and using it the ATDN architecture is trained. The resulting system managed to achieve 4.4% translation and 0.0176 deg/m rotational error on a subset of the KITTI dataset. The proposed architecture can be used for efficient and low-latency autonomous driving (AD) aiding database creation as well as a basis for autonomous vehicle (AV) control.