DocumentCode
1782857
Title
Polygamy based Genetic Algorithm for Unmanned Aerial Vehicle (UAV) power optimization: A proposal
Author
Aibinu, A.M. ; Bello Salau, H. ; Akachukwu, C.M. ; Nwohu, M.N.
Author_Institution
Dept. of Mechatron. Eng., Fed. Univ. of Technol., Minna, Nigeria
fYear
2014
fDate
Sept. 29 2014-Oct. 1 2014
Firstpage
1
Lastpage
5
Abstract
One of the challenges in the operation of Unmanned Aerial Vehicle (UAV) is power optimization under different operation mode. In solving the aforementioned problem, polygamy based selection Genetic Algorithm technique has been proposed in this work. The proposed technique involves parameter initialization, problem coding and optimization. The power requirement is coded as a bit of strings subject to power limit constraint. The initial solutions are evaluated based on the UAV power system objective function. The evaluated solutions are clustered into two different classes using K-Means algorithm. Chromosomes in the cluster with minimal centroid are then made to undergo polygamy mating subject to population control mechanism. Resulting solution were then mutated and the whole process continue till number of generation is reached or other termination criteria met. Application of the proposed technique shows that the average efficiency of UAV can be improved using the proposed algorithm.
Keywords
aircraft power systems; autonomous aerial vehicles; genetic algorithms; K-means algorithm; UAV power optimization; UAV power system objective function; parameter initialization; polygamy based genetic algorithm; population control mechanism; power limit constraint; problem coding; unmanned aerial vehicle; Algorithm design and analysis; Biological cells; Clustering algorithms; Linear programming; Robot kinematics; Robot sensing systems; Genetic Algorithm; K-Means Algorithm; Power Optimization; Unmanned Aerial Vehicle;
fLanguage
English
Publisher
ieee
Conference_Titel
Electronics, Computer and Computation (ICECCO), 2014 11th International Conference on
Conference_Location
Abuja
Print_ISBN
978-1-4799-4108-7
Type
conf
DOI
10.1109/ICECCO.2014.6997555
Filename
6997555
Link To Document