A centralized control of movements using a collision avoidance algorithm for a swarm of autonomous agents

Abstract

This work proposes a method for moving a swarm of autonomous Unmanned Aerial Vehicles to accomplish an specific task. The approach uses a centralized strategy which considers a trajectory calculation and collision avoidance. The solution was implemented in a simulated scenario as well as in a real controlled environment using a swarm of nano drones, together with a setup supported by a motion capture system. The solution was tested while planting virtual seeds in a field composed by a grid of points that represent the places to be sown. Experiments were performed for measuring completion times and attempts to prevent impacts in order to test the effectiveness, scalability and stability of the solution as well as the robustness of the collision avoidance algorithm while increasing the number of agents to perform the task.