Super eagle optimization algorithm based three-dimensional ball security corridor planning method for fixed-wing UAVs

TL;DR: This paper proposes a novel super eagle optimization algorithm (SEOA) for planning feasible paths for fixed-wing UAVs, achieving higher accuracy and efficiency in security corridor planning with smaller average rank and higher solving successful rate compared to other algorithms.

Abstract: In modern warfare, unmanned aerial vehicles (UAVs) have become a vital military force. Planning feasible paths for UAVs has become an essential part of forming autonomous attack systems. Aiming at the task requirements of fix-wing UAVs, this paper proposes a ball security corridor planning method by combining the ball Said-Ball curve and a novel super eagle optimization algorithm (SEOA). Firstly, the ball security corridor planning model is established by taking the center curve and radius function of the ball Said-Ball (BSB) curve as the flight path and security boundary. It not only considers the smoothness of the flight path, but also reserves enough security range to avoid cumulative errors. Then, a novel super eagle optimization algorithm is proposed to solve the established model with higher accuracy and efficiency. It simulates the hunting process of virtual super eagles with intelligent thinking. For super eagles, two patterns are designed for determining prey at different stages to avoid premature convergence. An information sharing strategy is also introduced to balance the exploitation and exploration abilities. For prey, they can choose orderly and emergency strategies based on emotional function to escape capture to speed up the convergence of SEOA. According to the results of solving the test function in CEC2017, the SEOA has smaller average rank compared to other flying animal-based algorithms and recent popular algorithms. For the solving successful rate, the SEOA reaches 82.7586% and 89.6552%, respectively. Additionally, the results of Wilcoxon rank-sum test statistically prove that the SEOA is a newly developed intelligent optimization method with significantly higher accuracy. Finally, the SEOA is used to solve the established model by optimizing the positions and radii of control balls of BSB curves. Experimental results in three complex simulation scenes show that the flight corridors obtained by SEOA outperform those generated by other algorithms in overall fitness values. This indicates a more rational optimization in flight altitude, distance, smoothness, and security range. Its faster convergence speed can ensure higher efficiency in planning a security corridor satisfying constraints and requirements.