Research on UAV path planning method based on improved HPO algorithm in multi-task environment

Linan Zu,Zhipeng Wang,Cong Liu,Shuzhi Sam Ge
DOI: https://doi.org/10.1109/jsen.2023.3297666
IF: 4.3
2023-01-01
IEEE Sensors Journal
Abstract:When unmanned aerial vehicles (UAVs) are carrying out missions, they are faced with complex task environments. To enhance the adaptability of UAV applications, it is required that they possess fast path-planning capabilities. This article takes the execution of complex tasks by multiple UAVs in a 3-D environment as background and transforms the path-planning problem into a multiconstraint optimization problem. Innovatively combining the hunter–prey optimizer (HPO) algorithm and task allocation mechanism, this study achieves collaborative path planning for multiple UAVs for complex tasks. In order to improve the optimization speed of the HPO algorithm, the following improvements have been made: first, introducing the chaotic mapping model to improve the performance of population initialization; second, adopting the golden sine strategy to change the population update strategy and accelerate the convergence speed of the algorithm. Then, the single-peak and multipeak benchmark functions are used to test the average, standard deviation, and optimal values of the algorithm. Finally, path planning experiments are carried out in a 3-D map with multiple task points and obstacles, and the results show that the adapt golden sine HOP (AGSHPO) algorithm improves the robustness and real-time performance of the UAVs when executing complex tasks to a certain extent.
engineering, electrical & electronic,instruments & instrumentation,physics, applied
What problem does this paper attempt to address?