ORB-SLAM3 Evaluation

Introduction

1 The overview of ORB-SLAM3

ORB-SLAM3 is the first system able to perform visual, visual-inertial and multi-map SLAM with monocular, stereo and RGB-D cameras, using pin-hole and fisheye lens models. The first main novelty is a feature-based tightly-integrated visual-inertial SLAM system that fully relies on Maximum-a-Posteriori (MAP) estimation, even during the IMU initialization phase. The second main novelty is a multiple map system that relies on a new place recognition method with improved recall. code link. paper link.

Image description — Fig. 1. Workflow diagram of ORB-SLAM3

2 ORB-SLAM3 Compilation Process

2.1 Requirement

(1)This package uses use the new thread and chrono functionalities of C++11.
(2)This package uses Pangolin for visualization and user interface .
(3)This package uses OpenCV to manipulate images and features. Required at leat 3.0.
(4)Eigen3 required by g2o. Required at least 3.1.0.
(5)This package uses modified versions of the DBoW2 library to perform place recognition and g2o library to perform non-linear optimizations.
(6)ROS Melodic

2.2 Build ORB-SLAM3

Clone the repository:


            git clone https://github.com/UZ-SLAMLab/ORB_SLAM3.git ORB_SLAM3

This package provides a script build.sh to build the Thirdparty libraries and ORB-SLAM3. Please make sure you have installed all required dependencies. Execute:


            cd ORB_SLAM3

            chmod +x build.sh

            ./build.sh

This will create libORB_SLAM3.so at lib folder and the executables in Examples folder.

Evaluation

Platforms	Sequences	Length(m)	ATE
A Handheld Device	Escalator	77.460	1.852
	MCR normal dark	76.499	X
	MCR aggressive 6dof light	100.871	2.854
	Parkway loop night	461.049	32.194
	Forest	130.937	1.324
A UGV	Elevator	39.336	X
	Indoor loop	270.674	1.359
	MCR hdr	193.918	8.527
	Street day	2064.475	12.276
	Parkway loop night	461.051	X
A QR	Underground	98.312	1.227
	MCR hdr	85.08	0.107
	Forest	108.037	0.643
A UAV	MCR loop light	104.989	1.134
A Car	Urban night loop	1807.884	17.645