parallel-genetic-algorithm-UAV-path-planning

UAV (Unmanned Aerial Vehicle) path planning using the Genetic Algorithm parallelized with CUDA. Search for the shortest path for a UAV to fly pass all selected cities in a real region of US. Implementation is based on this paper.

How to use

  1. Make sure Numba is installed. We use version 0.42.0.
  2. Download the United States Cities Database and put it in folder "data/".
  3. Run main.py to get the results shown below.

Results

Problem Setting: 6 UAVs should fly pass all US cities with population more than 100k.

The K-means clustering result:

The planned paths for all UAVs: