Skip to content

Latest commit

 

History

History
75 lines (48 loc) · 4.73 KB

README.md

File metadata and controls

75 lines (48 loc) · 4.73 KB
Raw

JORB-SLAM

JORB-SLAM is a multi-agent SLAM system based on ORB-SLAM2. Authors are Martin D Deegan, Yuanxin Zhong, Purva Kulkarni, Christine Searle, Kaustav Chakraborty

ORB-SLAM2

Authors: Raul Mur-Artal, Juan D. Tardos, J. M. M. Montiel and Dorian Galvez-Lopez (DBoW2)

ORB-SLAM2 is a real-time SLAM library for Monocular, Stereo and RGB-D cameras that computes the camera trajectory and a sparse 3D reconstruction (in the stereo and RGB-D case with true scale). It is able to detect loops and relocalize the camera in real time. Please check here for the Original Implementation

Related Publications:

Raúl Mur-Artal and Juan D. Tardós. ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras. IEEE Transactions on Robotics, vol. 33, no. 5, pp. 1255-1262, 2017. PDF.

Wang, John, and Edwin Olson. AprilTag 2: Efficient and robust fiducial detection. 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2016. PDF.

Build

Clone the repository:

git clone https://github.com/um-mobrob-t12-w19/ORB_SLAM2 ORB_SLAM2

We provide a script build.sh to build the Thirdparty libraries and ORB-SLAM2. Before running the build script, please turn to install_apriltags.sh and install_pangolin.sh for Pangolin and AprilTags installation, which are two dependencies of this program:

cd ORB_SLAM2

./install_apriltags.sh
./install_pangolin.sh

chmod +x build.sh
./build.sh

If you don't run install_apriltags.sh, you need to install libyaml-cpp with

sudo apt install libyaml-cpp-dev

This will create libORB_SLAM2.so at lib folder and the executables rgbd_tum and stereo_kitti in Examples folder.

Running Examples

Note: Follow the instructions in the command line when running.

(Stereo) KITTI Dataset

  1. Download the dataset (grayscale images) from http://www.cvlibs.net/datasets/kitti/eval_odometry.php

  2. Execute the following command. Change KITTIX.yamlto KITTI00-02.yaml, KITTI03.yaml or KITTI04-12.yaml for sequence 0 to 2, 3, and 4 to 12 respectively. Change PATH_TO_DATASET_FOLDER to the uncompressed dataset folder. Change SEQUENCE_NUMBER to 00, 01, 02,.., 11.

./Examples/Stereo/stereo_kitti Vocabulary/ORBvoc.txt Examples/Stereo/KITTIX.yaml PATH_TO_DATASET_FOLDER/dataset/sequences/SEQUENCE_NUMBER

Sample data sequence

We included a sample data sequence out of box under folder datasets, this sample is cut out from sequence 00 of KITTI odometry dataset. Unzip file kitti_odometry_00.zip into dataset/kitti_odometry_00 and execute run_kitti.sh, then you can find the system running on the sample sequence.

(RGBD) EECS Building Dataset

  1. Download a sequence from our EECS Building Dataset collected using RGB-D cameras and uncompress it. Or you can use the sample data sequence provided at the folder datasets.

  2. Unzip file EECSBlue$.zip into dataset/EECSBlue, EECSOrange$.zip into dataset/EECSOrange ($ can be a number or nothing).

  3. Execute run_eecs.sh then you can find the system running on the sample sequence. The final cloud view shows convisibility constraints in green lines and April tags constraint in red lines.

Processing your own sequences

You will need to create a settings file with the calibration of your camera. See the settings file provided for the TUM and KITTI datasets for monocular, stereo and RGB-D cameras. We use the calibration model of OpenCV, and you can do calibration with ROS calibration package or MATLAB Calibration App. RGB-D input must be synchronized and depth registered. Specifically for Intel Realsense Camera data bag, we provide Python scripts to convert the data bag to TUM dataset format.

SLAM and Localization Modes

You can change between the SLAM and Localization mode using the GUI of the map viewer.

SLAM Mode

This is the default mode. The system runs in parallal three threads: Tracking, Local Mapping and Loop Closing. The system localizes the camera, builds new map and tries to close loops.

Localization Mode

This mode can be used when you have a good map of your working area. In this mode the Local Mapping and Loop Closing are deactivated. The system localizes the camera in the map (which is no longer updated), using relocalization if needed.