This repository contains the Pytorch implementation of the radio mapping and path planning algorithm proposed in [1].
In the first step of the algorithm. the UAVs collaborate to build a model of the outage probability in the sky. For instance, in what follows, I presented the true location of the Base Stations and the outage model learned by the UAV agents. For this purpose, we use Federated Learning.
Structure of Federated Learning to estimate the outage probability:
A sample trajectory of the UAV based on RRT-star path planning algorithm:
If using this code for research purposes, please cite:
[1] B. Khamidehi and E. S. Sousa. "Federated learning for cellular-connected UAVs: Radio mapping and path planning." GLOBECOM 2020-2020 IEEE Global Communications Conference. IEEE, 2020.
@inproceedings{khamidehi2020federated,
title={Federated learning for cellular-connected UAVs: Radio mapping and path planning},
author={Khamidehi, Behzad and Sousa, Elvino S},
booktitle={GLOBECOM 2020-2020 IEEE Global Communications Conference},
pages={1--6},
year={2020},
organization={IEEE}
}