期刊
DRONES
卷 7, 期 5, 页码 -出版社
MDPI
DOI: 10.3390/drones7050300
关键词
delivery; drone; marker; YOLOv5; balcony
This research paper proposes a drone delivery system for multi-story apartment buildings with balconies, utilizing Vertical Grid Screening (VGS) methods to identify the specific balcony for delivery. The drone is equipped with a 295 mm frame size, a stereo camera, and a ranging sensor. It uses machine learning for image recognition and achieves a solution for last-mile delivery in urban areas.
Delivery drones typically perform delivery by suspending the parcel vertically or landing the drone to drop off the package. However, because of the constrained landing area and the requirement for precise navigation, delivering items to customers who reside in multi-story apartment complexes poses a unique challenge. This research paper proposes a novel drone delivery system for multi-story apartment buildings with balconies that employ two methods for Vertical Grid Screening (VGS), i.e., Grid Screening (GS) and Square Screening (SS), to detect unique markers to identify the precise balcony that needs to receive the product. The developed drone has a frame size of 295 mm and is equipped with a stereo camera and a ranging sensor. The research paper also explores the scanning and trajectory methods required for autonomous flight to accurately approach the marker location. The proposed machine learning system is trained on a YOLOv5 model for image recognition of the marker, and four different models and batch sizes are compared. The 32-batch size with a 960 x 1280 resolution model provides an average of 0.97 confidence for an extended range. This system is tested outdoors and shows an accuracy of 95% for a planned trajectory with 398 ms detection time as a solution for last-mile delivery in urban areas.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据