An Improve Crested Porcupine Algorithm for UAV Delivery Path Planning in Challenging Environments

Abstract With the rapid advancement of drone technology, the need for effective path planning in complex environments has become increasingly critical. Traditional algorithms struggle with complex terrain, obstacles, and weather changes, often falling into local optima. This study introduces an Improved Crown Porcupine Optimizer (ICPO) for drone path planning. Inspired by porcupines' defense mechanisms, a visuo-auditory synergy perspective is adopted, improving early convergence by balancing visual and auditory defenses. The study also employs a good point set population initialization strategy to enhance diversity and eliminates the traditional population reduction mechanism. To avoid local optima in later stages, a novel periodic retreat strategy inspired by porcupines' precise defenses is introduced for better position updates. Analysis on the IEEE CEC2022 test set shows that ICPO achieves near-optimal values, demonstrating robustness and stability. In complex terrains, including mountainous, urban, and composite environments, ICPO outperformed other advanced algorithms, proving its effectiveness and reliability in drone delivery path planning.