(2023) Damiani, Enrico Francesco; Abreu, Leonardo Duarte Malta de; Carrete, Luis Filipe Sanchez; Castanares, Manuel
The aim of this project is to develop a reinforcement learning algorithm with its
purpose being to find shipwrecked people using a swarm of drones. A simulated environment
was also developed to train and visualize the outcome of the trained algorithm. This project
does not discuss image recognition of shipwrecked people, since the true focus of this project
is to optimize the search routine of a drone to find the target in the quickest way possible. The
implemented Reinforce algorithm takes into account a dynamic map of probabilities,
representing the chances of a person being found, as well as the position of other agents.
Keywords: Multi-agent; Reinforcement learning, Shipwrecked people; Drone swarms.