ORB_SLAM2-Python-Jetbot

Part I: ORB-SLAM for Python 3.6 Setup

Clone the repositories into same folder.

Pangulin

https://github.com/stevenlovegrove/Pangolin

• Dependency

sudo apt install libgl1-mesa-dev
sudo apt install libglew-dev
sudo apt install cmake
sudo apt-get install libxkbcommon-x11-dev

• Build

git clone https://github.com/stevenlovegrove/Pangolin.git
cd Pangolin
mkdir build
cd build
cmake ..
cmake --build .

ORB-SLAM2

sudo apt-get install libeigen3-dev sudo apt install libwayland-dev https://github.com/raulmur/ORB_SLAM2

1. add #include <unistd.h> into Include/System.h.
2. Patch the repository (see Python Binding).
3. Build project

chmod +x build.sh
./build.sh

6. Python binding requires sudo make install.

Python Binding

https://github.com/jskinn/ORB_SLAM2-PythonBindings

1. Patch the orb-slam repository.

git apply ../ORB_SLAM2-PythonBindings/orbslam-changes.diff

2. Build ORB-SLAM2 (see ORB-SLAM2).
3. Modify the python version from 3.5 to 3.6 in "CMakeLists.txt"

31 find_package(PythonLibs 3.5 REQUIRED) -> 3.6
33 find_package(Boost 1.45.0 REQUIRED COMPONENTS python-py35) -> 36
72 install(TARGETS ${TARGET_MODULE_NAME} DESTINATION lib/python3.5/dist-packages) -> 3.6

4. Some Dependency.

sudo apt-get install libboost-all-dev

5. Build project.

mkdir build
cd build
cmake ..
make
make install

6. Test the python import

python3
>>> import orbslam2 

Test on TUM Dataset

TUM Datasets

https://vision.in.tum.de/data/datasets/rgbd-dataset/download

Recommand Sequence 'freiburg3_long_office_household' https://vision.in.tum.de/rgbd/dataset/freiburg3/rgbd_dataset_freiburg3_long_office_household.tgz

Run Code

1. Copy ORB_SLAM2-PythonBindings/txamples/orbslam_mono_tum.py to ORB-SLAM2/
2. Make folder Datasets/ and extract the dataset into it.
3. Run orb-slam

python3 orbslam_mono_tum.py Vocabulary/ORBvoc.txt Examples/Monocular/TUM3.yaml Datasets/rgbd_dataset_freiburg3_long_office_household

Part II: Camera Calibration & Jetbot SLAM Live Demo

Collect Data

Search Chessboard

Calibration

Extract Camera Config

Record Dataset

Some Problem

• restart camera

sudo systemctl restart nvargus-daemon

• Failed to load module “canberra-gtk-module”

sudo apt-get install libcanberra-gtk-module

Opencv Calibration Tutorial

https://opencv-python-tutroals.readthedocs.io/en/latest/py_tutorials/py_calib3d/py_calibration/py_calibration.html

Matplotlib

• FreeType http://www.linuxfromscratch.org/blfs/view/svn/general/freetype2.html ./configure sudo make install
• LibPng sudo apt-get install -y libpng-dev

Plotly

1. Install python plotly library

pip3 install plotly

2. Install jupyter extension and rebuild jupyter.

sudo jupyter labextension install @jupyterlab/plotly-extension
jupyter lab build

3. Restart your jetbot.