A Robust Stereo Camera Localization Method with Prior LiDAR Map Constrains

TL;DR

This paper introduces a stereo visual localization method that leverages prior LiDAR maps and coplanarity constraints to achieve robust and accurate 6-DOF pose estimation in GPS-denied environments, outperforming existing SLAM methods.

Contribution

It presents a novel visual optimization model using LiDAR map information and coplanarity constraints, enhancing robustness and accuracy in visual localization.

Findings

More robust than ORB-SLAM2 in KITTI dataset tests

Achieves full 6-DOF pose estimation without scale drift

Utilizes graph-based optimization with local window for efficiency

Abstract

In complex environments, low-cost and robust localization is a challenging problem. For example, in a GPSdenied environment, LiDAR can provide accurate position information, but the cost is high. In general, visual SLAM based localization methods become unreliable when the sunlight changes greatly. Therefore, inexpensive and reliable methods are required. In this paper, we propose a stereo visual localization method based on the prior LiDAR map. Different from the conventional visual localization system, we design a novel visual optimization model by matching planar information between the LiDAR map and visual image. Bundle adjustment is built by using coplanarity constraints. To solve the optimization problem, we use a graph-based optimization algorithm and a local window optimization method. Finally, we estimate a full six degrees of freedom (DOF) pose without scale drift. To validate the efficiency, the proposed method has been tested on the KITTI dataset. The results show that our method is more robust and accurate than the state-of-art ORB-SLAM2.