| Summary: | Bird damage to fruit crops causes significant monetary losses to farmers annually. The application of traditional bird repelling methods such as bird cannons and tree netting became inefficient in the long run, keeping high maintenance and reduced mobility. Due to their versatility, Unmanned Aerial Vehicles (UAVs) can be beneficial to solve this problem. However, due to their low battery capacity that equals low flight duration, it is necessary to evolve path planning optimization. A path planning optimization algorithm of UAVs based on Particle Swarm Optimization (PSO) is presented in this dissertation. This technique was used due to the need for an easy implementation optimization algorithm to start the initial tests. The PSO algorithm is simple and has few control parameters while maintaining a good performance. This path planning optimization algorithm aims to manage the drone's distance and flight time, applying optimization and randomness techniques to overcome the disadvantages of the traditional systems. The proposed algorithm's performance was tested in three study cases: two of them in simulation to test the variation of each parameter and one in the field to test the influence on battery management and height influence. All cases were tested in the three possible situations: same incidence rate, different rates, and different rates with no bird damage to fruit crops. The proposed algorithm presents promising results with an outstanding reduced average error in the total distance for the path planning obtained and low execution time. However, it is necessary to point out that the path planning optimization algorithm may have difficulty finding a suitable solution if there is a bad ratio between the total distance for path planning and points of interest. The field tests were also essential to understand the algorithm's behavior of the path planning algorithm in the UAV, showing that there is less energy discharged with fewer points of interest, but that do not correlates with the flight time. Also, there is no association between the maximum horizontal speed and the flight time, which means that the function to calculate the total distance for path planning needs to be adjusted.
|
|---|