Machine Learning Algorithm of Immune Neuro-Fuzzy Anticollision Embedded System for Autonomous Unmanned Aerial Vehicles Team

2nd International Conference on Applications of Intelligent Systems (APPIS 2019): Proceedings. ACM International Conference Proceedings Series 2019
Mihails Gorobecs, Leonīds Ribickis, Anatolijs Ļevčenkovs, Anna Beinaroviča

This study is dedicated to solve a collision prevention task for autonomous unmanned aerial vehicles’ (UAVs) team by Immune Neuro-Fuzzy Network (INFN) application. It is a part of the project aimed at the development of intelligent safety and optimal control systems of autonomous electric vehicles and transport in general. The goal of the current research is to develop the machine learning algorithm for autonomous UAV, that will give possibility for UAVs to train themselves without a teacher to avoid the collisions in the most effective way by changing the UAVs trajectory of the flight and without human intervention, i.e. the system should be self-organized. For this purpose, authors have improved previously developed immune neuro-fuzzy logic method to minimize collision probability. The experiments prove the workability and advantages of the developed algorithm.

Atslēgas vārdi
Machine learning, neural network, fuzzy logic, immune network, anti-collision system, autonomous vehicle, unmanned aerial vehicle, vehicle’s team
DOI
10.1145/3309772.3309797
Hipersaite
https://dl.acm.org/citation.cfm?doid=3309772.3309797

Gorobecs, M., Ribickis, L., Ļevčenkovs, A., Beinaroviča, A. Machine Learning Algorithm of Immune Neuro-Fuzzy Anticollision Embedded System for Autonomous Unmanned Aerial Vehicles Team. No: 2nd International Conference on Applications of Intelligent Systems (APPIS 2019): Proceedings. ACM International Conference Proceedings Series, Spānija, Las Palmas de Gran Canaria, 7.-9. janvāris, 2019. New York: ACM, 2019, Article number 25. ISBN 978-1-4503-6085-2. Pieejams: doi:10.1145/3309772.3309797

Publikācijas valoda
English (en)