(For USM Staff/Student Only)

EngLib USM > Ω School of Aerospace Engineering >

The development of a robust algorithm for uav path planning in 3d environment / Kok Kai Yit

The development of a robust algorithm for uav path planning in 3d environment_Kok Kai Yit _A2_2016_MYMY
Penyelidikan menyeluruh telah dijalankan berkaitan perancangan laluan Pesawat Udara Tanpa Pemandu (UAV) dengan menggunakan algoritma evolusi seperti pengoptimuman kerumunan zarah (PSO), algoritma genetik (GA), evolusi kebezaan (DE), dan pengoptimuman berasaskan biogeografik (BBO). Bagaimanapun, prestasi kebanyakan algoritma ini akan menurun dari segi kos fungsi dan pengiraan apabila digunakan dalam sistem yang teguh. Oleh itu, algoritma baru yang dikenali sebagai evolusi jangkitan (IE) telah dibina dalam kajian ini. IE memudahkan pengiraan dan memaksimumkan kecekapan menjana perancangan laluan yang lebih baik dalam persekitaran 3D. 9 peta telah digunakan sebagai kajian kes, dan 100 simulasi telah dijalankan dalam setiap kes untuk mendapat purata prestasi algoritma. Semua simulasi telah dijalankan melalui MATLAB dengan pembayangan perancangan laluan UAV. Prestasi algoritma IE telah dibandingkan dengan PSO, GA, DE dan BBO pada tetapan optimum algoritma masing-masing. IE berjaya merancang laluan UAV yang lebih pendek dengan kadar kebarangkalian 92 peratus dalam 100 kajian kes. Selain itu, IE mencapai kelajuan pemprosesan yang lebih cepat berbanding dengan algoritma lain dengan kadar kebarangkalian 97 peratus. Oleh itu, algoritma IE menunjukkan potensi yang besar dalam perancangan laluan UAV. _______________________________________________________________________ Significant research has been conducted on Unmanned Aerial Vehicle (UAV) path planning using evolutionary algorithms, such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Differential Evolution (DE), and Biogeographic-Based Optimization (BBO). However, the performance of most of these algorithms tend to decline in terms of function and computational cost when dealing with robust systems. Thus, a new algorithm known as infection evolution (IE) was developed in this study. IE simplifies calculation and maximizes the efficiency of generating an improved path plan in a 3D environment. Nine terrain maps were used as case studies, and 100 simulations were carried out for each case to determine the average performance of the proposed algorithm. All simulations were performed using MATLAB with visualization of UAV path planning. The performance of the IE algorithm was compared with that of PSO, GA, DE, and BBO at their respective optimized settings. IE attained a 92% probability rate of achieving a short path length in 100 case studies. With regard to computational cost, IE attained a 97% probability rate of achieving a faster processing speed in comparison with tested algorithms. Therefore, the IE algorithm exhibits significant potential for UAV path planning optimization.
Contributor(s):
Kai Yit Kok - Author
Primary Item Type:
Thesis
Identifiers:
Accession Number : 875008852
Language:
English
Subject Keywords:
Significant; Biogeographic; visualization
Sponsor - Description:
Pusat Pengajian Kejuruteraan Aeroangkasa -
First presented to the public:
3/1/2016
Original Publication Date:
8/21/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Aerospace Engineering
Citation:
Extents:
Number of Pages - 164
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-08-21 08:34:17.935
Submitter:
Mohamed Yunus Yusof

All Versions

Thumbnail Name Version Created Date
The development of a robust algorithm for uav path planning in 3d environment / Kok Kai Yit1 2020-08-21 08:34:17.935