This aim of this project is to develop a library for multiple object tracking. This a work in progress and it currently integrates an implementation of the Gaussian Mixture Probability Hypothesis Density filter used to estimate states of the target and some dynamic and measurement models.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
This library itself depends on Eigen 3 and was tested only on Ubuntu 18.04 and 20.04. You can download Eigen using apt:
apt-get install libeigen3-dev
The library is stored in folder motlib_cppas a standalone CMake project. To install it first clone the project and run
make
git clone https://github.com/rdesarz/motlib.git
cd motlib/motlib_cpp
mkdir build
cd build
cmake -DBUILD_TESTS=OFF -DCMAKE_BUILD_TYPE=Release ..
sudo make install
A ROS package is also provided with examples. It currently provides a simple 2D state estimation example with random single point targets. You will need ROS Noetic to run it. Please check http://wiki.ros.org/noetic/Installation for the installation instruction.
Then from the root folder, move to motlib_ros folder, run catkin_make to build the package
cd motlib_ros
catkin_make
An ROS launch file provides a tracking example of multiple single points target in a squared field of view. You can run it using the following command in motlib_ros folder:
source devel/setup.bash
roslaunch motlib_ros target_tracking_example.launch
- Romain Desarzens - Initial work - rdesarz
This project is licensed under the MIT License - see the LICENSE.md file for details