Path planning trajectory based on particle swarm optimization (PSO)

• Main Author: Say, Marc Francis Q.
• Format: text
• Language: English
• Published: Animo Repository 2021
• Subjects: Trajectories (Mechanics); Drone aircraft; Electrical and Electronics
• Online Access: https://animorepository.dlsu.edu.ph/etdm_ece/1; https://animorepository.dlsu.edu.ph/context/etdm_ece/article/1001/viewcontent/Say2.pdf
• Institution: De La Salle University

The research develops a path planning trajectory using the particle swarm optimization (PSO) for unmanned aerial vehicle (UAV) application. In order to create a practical trajectory, a cost function containing the environmental constraints and trajectory characteristics are used. The main characteristics being studied are the surveillance area importance (SAI), energy consumption (EC), and flight risk (FR). A trajectory having a high SAI value, low EC and FR are desirable for an autonomous UAV to use. Using PSO, trajectories for three UAVs are being generated to be used to reach a target location. For post disaster applications, it can be useful to generate a path planning trajectory for a drone pilot to use instead of manual flight. In this study, assuming a mountain environment with a landslide scenario, the PSO algorithm computes for the best path the UAVs can take to maximize the area of interest (SAI), minimize the battery consumption (EC) and the risk of flight (FR). In order to compare the performance of the PSO generated trajectories, a genetic algorithm (GA) based trajectory was also created. The results presented that the PSO generated paths has the better trajectory characteristics as compared to the GA.