Tightly Coupled Lidar-Inertial-GPS Environment Detection and Landing Area Selection Based on Powered Parafoil UAV

Enhancing the success rate of airdrops and minimizing resource wastage is crucial during disaster relief or wartime operations. However, it can be challenging to comprehend ground information in complex or inaccessible terrains and select suitable landing areas. To address this issue, this article introduces a novel integration of multiple sensors with a powered parafoil uncrewed aerial vehicle (UAV). Based on the characteristics of the parafoil, this study proposes a high-altitude release and low-altitude collection method along with a novel data processing method. By utilizing a tightly coupled Lidar simultaneous localization and mapping (SLAM), a point cloud map of the target area is constructed, and the drift of the map is constrained by leveraging the parafoil’s substantial inertia and data slicing. Additionally, suitable landing areas will be selected autonomously by map filter, data dimensionality reduction, and image processing in challenging environments. Finally, we designed an evaluation function to judge the quality of the airdrop landing area. Through simulation and real-flight experiments, it was verified that the system-generated point cloud map accurately represents the surrounding environment. Landing area with higher scores successfully avoids obstacles, with radius estimation matching the actual value within a 7% margin of error. This approach aligns with the requirements of airdrop landings and application scenarios involving emergency relief and battlefield situations.