A UAV Autonomous Exploration Method Based on High-Quality Viewpoints and Infilled Guidance
Xiaotao Liu, Min Cao, Guangnian Lu, Yangyang Xue, Jing Liu · IEEE/ASME Transactions on Mechatronics · 2025
In the field of autonomous exploration, although mainstream methods have achieved notable progress in this area, challenges, such as incomplete detection and insufficient guidance, continue to hinder exploration efficiency. To overcome these challenges, we introduce a novel autonomous exploration method that utilizes high-quality viewpoints for infilled-guided exploration. First, we present a mechanism for calculating and selecting quality viewpoints that allows the uncrewed aerial vehicle to observe frontier regions comprehensively while efficiently exploring unknown areas. This approach also identifies often-overlooked regions, minimizing unnecessary back-and-forth movements. Secondly, we propose an infilled-guided global navigation strategy which further enhances the overall performance throught minimizing the likelihood of frontiers in the outer circle. Extensive simulations and real-world experiments validate the accuracy and efficiency of our method. Compared with state-of-the-art approaches, our method reduces exploration time by at least 9% and shortens travel distance by over 5%.