The future of aircrafts is in unmanned aerial vehicles (UAVs), and any improvement in UAVs will play an important role, especially when it comes to intelligence and capabilities for air combat manoeuvring. The ultimate goal in such work is to bring computers to the level of a pilot’s intelligence capability in air combat. In order to achieve this goal, operations research is required. The present study is based on the fight or flight situation in air combat manoeuvring and aims to improve unmanned aircrafts and better understand the difficulties of modelling intelligence. Since the project’s focus is on the problem of path planning for moving targets and enemy situations, particle swarm optimization and genetic algorithms are modelled and tested against each other in a dog fight scenario. Also, multiple targets and enemies’ scenarios are developed to compare them against each other. Moreover, imperfect information affect and dynamic environment are evaluated in this research and required actions and options are analysed. Overall, this research aims to show the importance of artificial intelligence, articulate the role of the operations research and assess the implementation of intelligence through certain heuristics.
AIR COMBAT WITH PARTICLE SWARM OPTIMIZATION AND GENETIC ALGORITHM
2014
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0