Title: Drone Route Planning for Military Image Acquisition Using Parallel Simulated Annealing

Issue Number: Vol. 7, No. 3
Year of Publication: Sep - 2017
Page Numbers: 77-88
Authors: Soha S. Zaghloul, Eman Alsafi
Journal Name: International Journal of New Computer Architectures and their Applications (IJNCAA)
- Hong Kong
DOI:  http://dx.doi.org/10.17781/P002316

Abstract:


Recently, drones became one of the most interesting technologies used in diverse applications. One of the major challenges associated with them is to set a short route plan in the minimum period of time. This is an important inquiry in many applications, especially in military ones. On the other hand, the simulated annealing algorithm is one of the metaheuristics selected to generate a feasible solution to solve such problem. However, the algorithm requires long execution time. Therefore, the simulated annealing algorithm is parallelized in order to find a suitable solution of the problem in less time. Parallel programming divides the problem into smaller independent tasks, and then executes the subtasks simultaneously. Three performance metrics are used to measure the effectiveness of the parallel program; namely, the gained speedup, the efficiency and the scalability. Experiments are conducted to compare the performance of the sequential and parallel version of simulated annealing algorithm. In this paper, the algorithm is implemented on a sharedmemory processor environment. The results reported in this research show that the performance of the parallel simulated annealing on shared memory processor is higher than its sequential counterpart in almost all aspects. In addition, the given final distance is less than the sequential simulated annealing.