The objective of the package is to find goal assignments for 2D formation of aerial vehicles that guarantee collision free trajectories while minimizing the flight time. The package uses the "Concurrent Goal Assignment and Collision-Free Trajectory Generation for Multiple Aerial Robots" paper by Benjamin Gravell and Tyler Summers to develop python scripts that assign N aerial robots to N desired locations based on a one-to-one mapping that ensures minimal time-in-flight.
The package is composed on Python scripts and modules. Aside from the standard Python libraries, the package requires Scipy. To get scipy, follow the instruction in https://www.scipy.org/install.html.
The package was tested in a Linux environment. The sample usage in cga_cfa_alt.py might not work because it uses / for paths, but the class itself should still be supported.
Either download the zip file and extract at a desired location or clone the package using git clone https://github.com/The-SS/cga_cfa.git in a terminal window. Nothing else is required.
A sample usage is provided through main in cga_cfa_alt.py. You can execute this script and it should print out the output of solving the assignment problem.
You can also import the CGA_CFA_ALT class in a Python script and use the provided class API to update the parameters, solve the assignment problem, and retrieve the relevant results. The API and its usage are explained in cga_cfa_alt.py.