Path planning based on unmanned aerial vehicle performance with segmented cellular genetic algorithm

Unmanned Aerial Vehicles (UAV) have a wide range of use on industrial, military and commercial areas. Comprehensive and error-free sub-systems are needed to provide planning, management and coordination of UAVs which are designed for variable purposes with different capabilities and sizes. An important part of UAV technological development consists of improvements in the scope of path planning. Different choices can be made in path planning according to operational priorities, it may be preferred to reach the destination as fast as possible or to increase the airtime by compromising speed. Fuel data of cruise, climb and descent phases are used in the path planning algorithm for every speed and altitude that the UAV can fly. Thus, economical and airtime-maximizing paths could be produced on the basis of performance characteristics compatible with the kinematic constraints customized for the UAV. In this thesis, Cellular (cGA) and Segmented Cellular Genetic Algorithm (scGA) are proposed. The novel overprotective algorithm which has a fixed initial population and segmented chromosome structure achieves a high convergence speed to optimal solution and can generate paths which have 5.2 times higher fitness value on average compared with a conventional Genetic Algorithm (GA). It has been observed that scGA improves the initial population in terms of the best solutions 1.9 times and the general population 5.8 times better compared with GA.