Skip to content

wlin70/OpenVSLAM_Dev

BranchesTags
 
 

Repository files navigation

OpenVSLAM_Dev

This repository has the process to setup OpenVSLAM and how to use it.

System:

The below code is tested on:

  • Ubuntu 18.05 LTS
  • NVIDIA GTX 950M

Initial Installation

Do initial installation as per this.

For Linux, the commands are all laid out here:

# Clone the official VSLAM repo
git clone https://github.com/xdspacelab/openvslam

# grant yourself root permissions. 
sudo -s

apt update -y
apt upgrade -y --no-install-recommends

# basic dependencies
apt install -y build-essential pkg-config cmake git wget curl unzip

# g2o dependencies
apt install -y libatlas-base-dev libsuitesparse-dev

# OpenCV dependencies
apt install -y libgtk-3-dev
apt install -y ffmpeg
apt install -y libavcodec-dev libavformat-dev libavutil-dev libswscale-dev libavresample-dev

# eigen dependencies
apt install -y gfortran

# other dependencies
apt install -y libyaml-cpp-dev libgoogle-glog-dev libgflags-dev

# Download and install Eigen from source.
cd ~/openvslam/
wget -q https://gitlab.com/libeigen/eigen/-/archive/3.3.7/eigen-3.3.7.tar.bz2
tar xf eigen-3.3.7.tar.bz2
rm -rf eigen-3.3.7.tar.bz2
cd eigen-3.3.7
mkdir -p build && cd build
cmake \
    -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=/usr/local \
    ..
make -j4
make install

# Download, build and install OpenCV from source.

cd ~/openvslam/
wget -q https://github.com/opencv/opencv/archive/3.4.0.zip
unzip -q 3.4.0.zip
rm -rf 3.4.0.zip
cd opencv-3.4.0
mkdir -p build && cd build
cmake \
    -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=/usr/local \
    -DENABLE_CXX11=ON \
    -DBUILD_DOCS=OFF \
    -DBUILD_EXAMPLES=OFF \
    -DBUILD_JASPER=OFF \
    -DBUILD_OPENEXR=OFF \
    -DBUILD_PERF_TESTS=OFF \
    -DBUILD_TESTS=OFF \
    -DWITH_EIGEN=ON \
    -DWITH_FFMPEG=ON \
    -DWITH_OPENMP=ON \
    ..
make -j4
make install


# Common Installation Instructions
# Download, build and install the custom DBoW2 from source.

cd ~/openvslam/
git clone https://github.com/shinsumicco/DBoW2.git
cd DBoW2
mkdir build && cd build
cmake \
    -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=/usr/local \
    ..
make -j4
make install

# Download, build and install g2o.

cd ~/openvslam/
git clone https://github.com/RainerKuemmerle/g2o.git
cd g2o
git checkout 9b41a4ea5ade8e1250b9c1b279f3a9c098811b5a
mkdir build && cd build
cmake \
    -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=/usr/local \
    -DCMAKE_CXX_FLAGS=-std=c++11 \
    -DBUILD_SHARED_LIBS=ON \
    -DBUILD_UNITTESTS=OFF \
    -DBUILD_WITH_MARCH_NATIVE=ON \
    -DG2O_USE_CHOLMOD=OFF \
    -DG2O_USE_CSPARSE=ON \
    -DG2O_USE_OPENGL=OFF \
    -DG2O_USE_OPENMP=ON \
    ..
make -j4
make install

# (if you plan on using PangolinViewer)
# Download, build and install Pangolin from source.

cd ~/openvslam/
git clone https://github.com/stevenlovegrove/Pangolin.git
cd Pangolin
git checkout ad8b5f83222291c51b4800d5a5873b0e90a0cf81
mkdir build && cd build
cmake \
    -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=/usr/local \
    ..
make -j4
make install

# (if you plan on using SocketViewer)
# Download, build and install socket.io-client-cpp from source.

cd ~/openvslam/
git clone https://github.com/shinsumicco/socket.io-client-cpp.git
cd socket.io-client-cpp
git submodule init
git submodule update
mkdir build && cd build
cmake \
    -DCMAKE_BUILD_TYPE=Release \
    -DCMAKE_INSTALL_PREFIX=/usr/local \
    -DBUILD_UNIT_TESTS=OFF \
    ..
make -j4
make install

# (if you plan on using SocketViewer)
# Install Protobuf.

# for Ubuntu 18.04 (or later)
apt install -y libprotobuf-dev protobuf-compiler

Build Instructions

When building with support for PangolinViewer, please specify the following cmake options: -DUSE_PANGOLIN_VIEWER=ON and -DUSE_SOCKET_PUBLISHER=OFF.

cd ~/openvslam/
mkdir build && cd build
cmake \
    -DBUILD_WITH_MARCH_NATIVE=ON \
    -DUSE_PANGOLIN_VIEWER=ON \
    -DUSE_SOCKET_PUBLISHER=OFF \
    -DUSE_STACK_TRACE_LOGGER=ON \
    -DBOW_FRAMEWORK=DBoW2 \
    -DBUILD_TESTS=ON \
    ..
make -j4

When building with support for SocketViewer, please specify the following cmake options: -DUSE_PANGOLIN_VIEWER=OFF and -DUSE_SOCKET_PUBLISHER=ON.

cd ~/openvslam/
mkdir build && cd build
cmake \
    -DBUILD_WITH_MARCH_NATIVE=ON \
    -DUSE_PANGOLIN_VIEWER=OFF \
    -DUSE_SOCKET_PUBLISHER=ON \
    -DUSE_STACK_TRACE_LOGGER=ON \
    -DBOW_FRAMEWORK=DBoW2 \
    -DBUILD_TESTS=ON \
    ..
make -j4

After building, check to see if it was successfully built by executing

./run_kitti_slam -h

The output should be like this:

Allowed options:
-h, --help             produce help message
-v, --vocab arg        vocabulary file path
-d, --data-dir arg     directory path which contains dataset
-c, --config arg       config file path
--frame-skip arg (=1)  interval of frame skip
--no-sleep             not wait for next frame in real time
--auto-term            automatically terminate the viewer
--debug                debug mode
--eval-log             store trajectory and tracking times for evaluation
-p, --map-db arg       store a map database at this path after SLAM

Here onwards scroll down and check the official documentation to run in SocketViewer.

Launching socketviewer

$ cd ~/openvslam/viewer
$ node app.js

The output should be as follows:

WebSocket: listening on *:3000
HTTP server: listening on *:3001

Open a new browser tab and type in :

http://localhost:3001/

You should now be able to see the Socket Viewer. Continue with Simple Tutorials here.

Running on any dataset:

There are many sample datasets as seen here. To extract any one specific folder in openvslam/build, do the following

  • Right click on the scenario and copy link
  • Copy the FILE_ID part of the link.

For example, while trying to extract for aist_factory_A.zip, the link is

https://drive.google.com/file/d/10JZjmZYntdpiDvIxe8sxLlF4xZ0exyw0/view?usp=sharing

Here the FILE_ID is

10JZjmZYntdpiDvIxe8sxLlF4xZ0exyw0

Now, to download and unzip anything from this directory, execute the following commands:

$ FILE_ID="10JZjmZYntdpiDvIxe8sxLlF4xZ0exyw0"

$ curl -sc /tmp/cookie "https://drive.google.com/uc?export=download&id=${FILE_ID}" > /dev/null

$ CODE="$(awk '/_warning_/ {print $NF}' /tmp/cookie)"

$ curl -sLb /tmp/cookie "https://drive.google.com/uc?export=download&confirm=${CODE}&id=${FILE_ID}" -o aist_factory_A.zip

$ unzip aist_factory_A.zip

You can see the extracted folder in the build directory now.

Running on a standard dataset - Kitti Dataset

The dataset can be downloaded from here. Download the greyscale one.

Paste the folders in the right directory. I pasted the kitti dataset in the openvslam/example/kitti/ directory. You can see the sequences of images in them. To launch, start the socket viewer as shown previously. Then, enter the following commands:

$ cd ~/openvslam/build

Launching socketviewer

All commands are followed as per this

$ cd ~/openvslam/viewer
$ node app.js

The output should be as follows:

WebSocket: listening on *:3000
HTTP server: listening on *:3001

Open a new browser tab and type in :

http://localhost:3001/

Open a new terminal

$ sudo -s

$ ./run_kitti_slam -v ../build/orb_vocab/orb_vocab.dbow2 -d ../example/kitti/dataset/sequences/00/ -c ../example/kitti/KITTI_mono_00-02.yaml -p kitti_map.msg

Press Terminate on socketviewer to stop and save the map msg file.

Running on ROS

Launching socketviewer

All commands are followed as per this

$ cd ~/openvslam/viewer
$ node app.js

The output should be as follows:

WebSocket: listening on *:3000
HTTP server: listening on *:3001

Open a new browser tab and type in :

http://localhost:3001/

Open a new terminal, and Start roscore

$ roscore

Open a new terminal

$ source ~/openvslam/ros/devel/setup.bash
$ rosrun publisher video -m /path/to/video.mp4

Open a new terminal, Republish the ROS topic to /camera/image_raw

$ source ~/openvslam/ros/devel/setup.bash
$ rosrun image_transport republish raw in:=/video/image_raw raw out:=/camera/image_raw

Open a new terminal

$ source ~/openvslam/ros/devel/setup.bash
$ rosrun openvslam run_slam -v build/orb_vocab/orb_vocab.dbow2 -c /path/to/config.yaml

Open a new terminal

$ source ~/openvslam/ros/devel/setup.bash
$ rosrun openvslam run_localization -v build/orb_vocab/orb_vocab.dbow2 -c /path/to/config.yaml --map-db build/map.msg

Visualization and using point cloud data on a custom video

Paste the custom test_video.mp4 in ~/openvslam/example directory and run the following command while SocketViewer is running:

$ ./run_video_slam -v ../build/orb_vocab/orb_vocab.dbow2 -m ../example/test_video.mp4 -c ../build/aist_factory_A_1/config.yaml --map-db ~/openvslam/build/test_video_map.msg

After running OpenVSLAM, complete the slam map generation by clicking the terminate button in SocketViewer. Following this, find the generated test_video_map.msg file. Run the following command in the terminal from the /example folder:

$ python3 msg2csv.py ~/openvslam/build/test_video_map.msg output_file.csv

This will save the msg into a csv file format. Now, this can be visualized using open3d. Run the following command in the terminal from the /example folder:

$ python3 csv2ply.py

The visualization can be seen now.

General commands to run each program

Visualizing using open3d and .msg file

Run from the ~/openvslam/example folder

$ python3 visualize_openvslam_map.py ~/openvslam/build/map.msg

Converting from .msg file to .csv

Run from the ~/openvslam/example folder

$ python3 msg2csv.py ~/openvslam/build/map.msg output_file.csv

Converting from .msg file to .pcd

Run from the ~/openvslam/example folder

$ python3 msg2pcd.py ~/openvslam/build/map.msg output_file.pcd

Converting from .csv file to .ply file

Run from the ~/openvslam/example folder

$ python3 csv2ply.py

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

0 stars

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%