Accurate Landing of Unmanned Aerial Vehicles Using Ground Pattern Recognition

The capability of drones to perform autonomous missions has led retail companies to use them for deliveries, saving time and human resources. In these services, the delivery depends on the Global Positioning System (GPS) to define an approximate landing point. However, the landscape can interfere with the satellite signal (e.g., tall buildings), reducing the accuracy of this approach. Changes in the environment can also invalidate the security of a previously defined landing site (e.g., irregular terrain, swimming pool). Therefore, the main goal of this work is to improve the process of goods delivery using drones, focusing on the detection of the potential receiver. We developed a solution that has been improved along its iterative assessment composed of five test scenarios. The built prototype complements the GPS through Computer Vision (CV) algorithms, based on Convolutional Neural Networks (CNN), running in a Raspberry Pi 3 with a Pi NoIR Camera (i.e., No InfraRed—without infrared filter). The experiments were performed with the models Single Shot Detector (SSD) MobileNet-V2, and SSDLite-MobileNet-V2. The best results were obtained in the afternoon, with the SSDLite architecture, for distances and heights between 2.5–10 m, with recalls from 59%–76%. The results confirm that a low computing power and cost-effective system can perform aerial human detection, estimating the landing position without an additional visual marker.

“…Wubben et al proposed a solution that can land a UAV, equipped with a low-cost camera, on an ArUco marker [37]. In the first phase, the UAV makes an initial approach to the landing site.…”

Section: Related Workmentioning

confidence: 99%

UAV Landing Using Computer Vision Techniques for Human Detection

Safadinho

Ramos

Ribeiro

et al. 2020

“…A huge number of articles are dedicated to the task of UAV landing on moving platforms, which has been in particular accelerated by the Mohamed Bin Zayed International Robotics Challenge (MBZIRC) [ 2 , 18 , 19 , 20 , 21 ]. Most of the works on landing are dedicated to control [ 22 , 23 , 24 , 25 ] or localization and computer vision for UAV positioning in the landing station [ 26 , 27 , 28 , 29 , 30 , 31 ].…”

Section: Introductionmentioning

confidence: 99%

UAV Positioning Mechanisms in Landing Stations: Classification and Engineering Design Review

Galimov

Fedorenko

Klimchik

2020

Landing platforms’ automation is aimed at servicing vertical take-off and landing UAVs between flights and maintaining their airworthiness. Over the last few years, different designs for the landing platforms have been proposed. This shows a strong development and establishment of automatic landing platforms with UAV positioning devices on the landing site. Positioning and safe fixation of the UAV are some of the main features of the landing platform, especially if it is mounted on a movable vehicle. This article focuses exclusively on the landing platform and its elements that provide the positioning of the UAV by affecting it during and after the landing. Both active devices and mechanisms and passive elements used for positioning are considered. This article, based on the review of recent patents and publications, gives the classification of positioning approaches used in landing stations with the analysis of the required landing precision, as well as the pros and cons of each type of approach.

“…Object detection is a common topic and has attracted the most interest in recent studies. In object detection, traditional handcrafted feature-based methods showed limited performance [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 ]. A competitive approach is to apply deep-learning-based methods, which have gained popularity in recent years [ 13 , 14 , 15 , 16 ].…”

Section: Introductionmentioning

confidence: 99%

SlimDeblurGAN-Based Motion Deblurring and Marker Detection for Autonomous Drone Landing

Truong

Lee

Owais

et al. 2020

Deep learning-based marker detection for autonomous drone landing is widely studied, due to its superior detection performance. However, no study was reported to address non-uniform motion-blurred input images, and most of the previous handcrafted and deep learning-based methods failed to operate with these challenging inputs. To solve this problem, we propose a deep learning-based marker detection method for autonomous drone landing, by (1) introducing a two-phase framework of deblurring and object detection, by adopting a slimmed version of deblur generative adversarial network (DeblurGAN) model and a You only look once version 2 (YOLOv2) detector, respectively, and (2) considering the balance between the processing time and accuracy of the system. To this end, we propose a channel-pruning framework for slimming the DeblurGAN model called SlimDeblurGAN, without significant accuracy degradation. The experimental results on the two datasets showed that our proposed method exhibited higher performance and greater robustness than the previous methods, in both deburring and marker detection.