Abstract:Autonomous mobile vehicles are expected to perform persistent and accurate localization with low-cost equipment. To achieve this goal, we propose a stereo camera based visual localization method using a modified laser map, which takes the advantage of both the low cost of camera, and high geometric precision of laser data to achieve long-term performance. Considering that LiDAR and camera give measurements of the same environment in different modalities, the cross-modal invariance is investigated to modify the laser map for visual localization. Specifically, a map learning algorithm is introduced to sample the robust subsets in laser maps that are useful for visual localization using multi-session visual and laser data. Further, a generative map model is derived to describe this crossmodal invariance, based on which two types of measurements are defined to model the laser map points as appropriate visual observations. Tightly coupling these measurements within the local bundle adjustment during online sliding-window based visual odometry, the vehicle can achieve robust localization even one year after the map was built. The effectiveness of the proposed method is evaluated on both the public KITTI datasets and self-collected datasets in our campus, which include seasonal, illumination and object variations. On all experimental localization sessions, our method provides satisfactory results, even when the direction is opposite to that in the mapping session, verifying the superior performance of the laser map based visual localization method.

A Mobile System Combining Laser Scanners And Cameras For Urban Spatial Objects Extraction

Robust Mobile Robot Localization By Fusing Laser And Sonar

Persistent Stereo Visual Localization on Cross-Modal Invariant Map

An Efficient Extrinsic Calibration Of A Multiple Laser Scanners And Cameras' Sensor System On A Mobile Platform

A Vehicle-Borne Urban 3-D Acquisition System Using Single-Row Laser Range Scanners

Extraction of moving objects at an intersecton and application for safe driving using laser scannners

A Laser-Scanner-Based Approach Toward Driving Safety and Traffic Data Collection

Coarse-to-fine Hybrid 3D Mapping System with Co-calibrated Omnidirectional Camera and Non-repetitive LiDAR

THE IMPLEMENTATION OF SEMI-AUTOMATED ROAD SURFACE MARKINGS EXTRACTION SCHEMES UTILIZING MOBILE LASER SCANNED POINT CLOUDS FOR HD MAPS PRODUCTION

An Object Localization System Using Monocular Camera and Two-Axis-Controlled Laser Ranging Sensor for Mobile Robot

Updating a digital geographic database using vehicle-borne laser scanners and line cameras

Rapid Geometric Evaluation of Transportation Infrastructure Based on a Proposed Low-Cost Portable Mobile Laser Scanning System

LOW-COST WHEELED ROBOT-BORNE LASER SCANNING SYSTEM FOR INDOOR AND OUTDOOR 3D MAPPING APPLICATION

Moving Object Classification Using Horizontal Laser Scan Data

Sensing an Intersection Using a Network of Laser Scanners and Video Cameras.

Colorisation et texturation temps réel d'environnements urbains par système mobile avec scanner laser et caméra fish-eye

IMU and Multiple RGB-D Camera Fusion for Assisting Indoor Stop-and-Go 3D Terrestrial Laser Scanning

Combining Laser-Scanning Data and Images for Target Tracking and Scene Modeling

Reconstructing Urban 3D Model Using Vehicle-Borne Laser Range Scanners

Computing Object-Based Saliency in Urban Scenes Using Laser Sensing