A Review of Precise Positioning for Logistics Drones in Rainy Night Conditions

Abstract

With fast development of the low-altitude economy and ongoing improvement of intelligent logistics infrastructure, drone delivery has become an important new way to solve the "last-mile" delivery problem. Even so, complicated weather conditions, like flying at night and when it rains, create big technical problems for getting accurate location and safe landing of unmanned aerial vehicles (UAVs). This is a key bottleneck that stops logistics drones from being used widely for business purposes. This paper gives a full look at existing research on technologies for UAV precise positioning and landing, and it sums up what current domestic and international studies have achieved in four main topic areas: guidance methods that use vision, GNSS improvement techniques, ways to combine multiple sensors, and new passive microwave radiometric imaging technologies. Based on all this information, our analysis finds several key problems that are common in current work: single sensors have limits in what they can detect, multi-sensor fusion strategies don't go deep enough into how different modes interact, there are not enough systematic studies that specifically look at the combined problem of night and rainy weather, it's really hard to keep a good balance between real-time working and how much computing power you use, and there are no standard benchmark and evaluation frameworks that everyone can use. In the end, based on all the review we did above, this paper points out some possible future research directions that could guide new developments in this area.