Accurate Prior-centric Monocular Positioning with Offline LiDAR Fusion

TL;DR

This paper presents a monocular camera-based localization method that leverages a LiDAR-enhanced visual prior map to achieve high-precision positioning, reducing reliance on expensive GPS and LiDAR sensors for cost-effective autonomous navigation.

Contribution

The authors introduce a novel prior-centric monocular positioning approach that fuses deep-learning visual features with LiDAR-enhanced maps for accurate onboard localization.

Findings

Achieves centimeter-level global positioning accuracy.

Provides a low-cost, easily integrated localization solution.

Demonstrates effectiveness in real-world scenarios.

Abstract

Unmanned vehicles usually rely on Global Positioning System (GPS) and Light Detection and Ranging (LiDAR) sensors to achieve high-precision localization results for navigation purpose. However, this combination with their associated costs and infrastructure demands, poses challenges for widespread adoption in mass-market applications. In this paper, we aim to use only a monocular camera to achieve comparable onboard localization performance by tracking deep-learning visual features on a LiDAR-enhanced visual prior map. Experiments show that the proposed algorithm can provide centimeter-level global positioning results with scale, which is effortlessly integrated and favorable for low-cost robot system deployment in real-world applications.