A Bioinspired Neural Network-Based Approach for Cooperative Coverage Planning of UAVs

This paper describes a bioinspired neural-network-based approach to solve coverage planning problem for fleet of unmanned aerial vehicles exploring critical areas. The main goal is fully cover the map, maintaining uniform distribution on and avoiding collisions between other obstacles. specific task suitable surveillance applications, where in map permits them reach any position as fast possible emergency scenarios. To this problem, neural network structure adopted. Specifically, consists grid neurons, each neuron has local cost connection only with neighbor neurons. influences its neighbors, generating an attractive contribution unvisited We introduce several controls precautions minimize risk optimize planning. Then, preliminary simulations are performed different scenarios by testing algorithm four maps fleets consisting 3 10 vehicles. Results confirm ability proposed manage coordinate providing full every tested scenario, vehicles, uniformly distributing map.