A software system aimed at the modeling of self-organizing swarm of drones equipped with collective intelligence for area monitoring and tasks performance is planned to promote.
For the purpose, our methods and algorithms for constructing optimal and fault-tolerant schemes (gossip/broadcast models) will all be adapted thus creating premises for developing self-organizing and decentralized swarm of logically linked drones. During a quasi-random walk (rotor-router model), drones will perform dynamic snapshotting of the studied area; image coding (sandpile model); reliable and full exchange of captured images along with transferring the results to the data processing center.
Based on the integration of received data, the dynamic image of the area will be reproduced, the collective investigation of which will be carried out on a multi-user software platform. As a result, targeted tasks will be assigned, for instance, task to send a new swarm of drones to newly determined coordinates. During the entire process of snapshotting, the swarm of drones will not use any global positioning system and external management tool.