Yu Liu

Absolute visual localization based on onboard cameras and georeferenced data is the dominant solution for unmanned aerial vehicles (UAVs) when the global navigation satellite system is not available. However, current image-based matching methods are easily affected by significant feature differences between UAV images and georeferenced images, and map-based matching methods are affected by image rotation and resolution differences. To address the above-mentioned challenges, this article proposes a rotation and scale invariance map matching method (RSIM) to realize UAV visual localization in urban scenarios. Specifically, RSIM first extracts building contour information from UAV images. Then, based on the shape and spatial relationship features of the buildings, a scene matching method is designed to match the UAV image and the vector e-map. Finally, the centers of the matched building individuals are used as control points for UAV position solving. Extensive experiments conducted on the LoFTR and DKM datasets demonstrate that the proposed RSIM algorithm achieves superior localization performance compared to existing deep learning-based image matching methods. Moreover, our RSIM effectively enables UAV localization in urban scenarios even when the orientation and resolution of UAV images are unknown, achieving performance comparable to the SSRM map matching method, which relies on prior knowledge of these parameters.

In urban scenes, buildings are usually dense and exhibit similar shapes. Thus, existing autonomous unmanned aerial vehicle (UAV) localization schemes based on map matching, especially the semantic shape matching (SSM) method, cannot capture the uniqueness of buildings and may result in matching failure. To solve this problem, we propose a new method to locate UAVs via shape and spatial relationship matching (SSRM) of buildings in urban scenes as an alternative to UAV localization via image matching. SSRM first extracts individual buildings from UAV images using the SOLOv2 instance segmentation algorithm. Then, these individual buildings are subsequently matched with vector e-map data (stored in .shp format) based on their shape and spatial relationship to determine their actual latitude and longitude. Control points are generated according to the matched buildings, and finally, the UAV position is determined. SSRM can efficiently realize high-precision UAV localization in urban scenes. Under the verification of actual data, SSRM achieves localization errors of 7.38 m and 11.92 m in downtown and suburb areas, respectively, with better localization performance than the radiation-variation insensitive feature transform (RIFT), channel features of the oriented gradient (CFOG), and SSM algorithms. Moreover, the SSRM algorithm exhibits a smaller localization error in areas with higher building density.

Cholesteric liquid crystal microcapsules (CLCMs) are used to improve the stability of liquid crystals while ensuring their stimulus response performance and versatility, with representative applications such as sensing, anticounterfeiting, and smart fabrics. However, the reflectivity and angular anisotropy decrease because of the anchoring effect of the polymer shell matrix, and the influence of particle size on this has not been thoroughly studied. In this study, the effect of synthesis technology on microcapsule particle size was investigated using a complex coalescence method, and the effect of particle size on the reflectivity and angular anisotropy of CLCMs was investigated in detail. A particle size of approximately 66 µm with polyvinyl alcohol (PVA, 1:1) exhibited a relative reflectivity of 16.6% and a bandwidth of 20 nm, as well as a narrow particle size distribution of 22 µm. The thermosetting of microcapsules coated with PVA was adjusted and systematically investigated by controlling the mass ratio. The optimized mass ratio of microcapsules (66 µm) to PVA was 2:1, increasing the relative reflectivity from 16.6% (1:1) to 32.0% (2:1) because of both the higher CLCM content and the matching between the birefringence of the gelatin–arabic shell system and PVA. Furthermore, color based on Bragg reflections was observed in the CLCM-coated ortho-axis and blue-shifted off-axis, and this change was correlated with the CLCM particle size. Such materials are promising for anticounterfeiting and color-based applications with bright colors and angular anisotropy in reflection.

Localization is the most important basic information for unmanned aerial vehicles (UAV) during their missions. Currently, most UAVs use GNSS to calculate their own position. However, when faced with complex electromagnetic interference situations or multipath effects within cities, GNSS signals can be interfered with, resulting in reduced positioning accuracy or even complete unavailability. To avoid this situation, this paper proposes an autonomous UAV localization method for low-altitude urban scenarios based on POI and store signage text matching (LPS) in UAV images. The text information of the store signage is first extracted from the UAV images and then matched with the name of the POI data. Finally, the scene location of the UAV images is determined using multiple POIs jointly. Multiple corner points of the store signage in a single image are used as control points to the UAV position. As verified by real flight data, our method can achieve stable UAV autonomous localization with a positioning error of around 13 m without knowing the exact initial position of the UAV at take-off. The positioning effect is better than that of ORB-SLAM2 in long-distance flight, and the positioning error is not affected by text recognition accuracy and does not accumulate with flight time and distance. Combined with an inertial navigation system, it may be able to maintain high-accuracy positioning for UAVs for a long time and can be used as an alternative to GNSS in ultra-low-altitude urban environments.

In the traditional power equipment positioning system, to achieve tag positioning, the system needs servers, base stations, and tags. To ensure more stability of the system, it needs to range with four base stations during tag positioning. However, the system will have excessive power consumption, and the time to initiate ranging between multiple tags is random, which will cause serious data distortion. Therefore, to adjust the system with complex power equipment management, this paper proposes a low power label time-sharing location method. First, establishing an ultra-wideband UWB between the BTSs to synchronize the time of each BTS. Then, the tag just ranges with one of the BTSs, and the other BTSs get a receiving time. Finally, the time difference method is used to convert the distance between the tag and all other base stations, to reduce the power consumption of the system. The feasibility of the proposed method is verified by an example. This method can effectively solve the problem of multi-label location, which makes the power consumption of label location low and the location will not conflict, and has good engineering practical value.

Road network extraction plays a critical role in data updating, urban development, and decision support. To improve the efficiency of labeling road datasets and addressing the problems of traditional methods of manually extracting road networks from high-resolution images, such as their slow speed and heavy workload, this paper proposes a semi-automatic method of road network extraction from high-resolution remote-sensing images. The proposed method needs only a few points to extract a single road in the image. After the roads are extracted one by one, the road network is generated according to the width of each road and the spatial relationships among the roads. For this purpose, we use regional growth, morphology, vector tracking, vector simplification, endpoint modification, road connections, and intersection connections to generate road networks. Experiments on four images with different terrains and different resolutions show that this method has high extraction accuracy under different image conditions. The comparisons with the semi-automatic GVF-snake method based on regional growth also showed its advantages and potentiality. The proposed method is a novel form of semi-automatic road network extraction, and it significantly increases the efficiency of road network extraction.

Many systems are required to perform a series of missions with finite breaks or limited maintenance resources between successive missions. In such cases, one of the most widely used maintenance policies is a selective maintenance in which a subset of feasible maintenance actions is chosen to satisfy the subsequent mission requirements. Traditional selective maintenance optimization problems reported in the literature focus on simple binary state systems. Most of systems in industries have more than two states in the process of deterioration, however. In this paper, we consider a selective maintenance problem of multi-state systems (MSS). In order to develop a selective maintenance model, we review Kijima's imperfect maintenance model and suggest age reduction factor as a function of cost. The extended universal generation function (UGF) is used to evaluate the probability of completing a single mission successfully. The non-linear 0-1 programming in binary selective maintenance optimization is extended to non-linear continuous programming which is difficult to solve through previously reported enumeration methods. We adapt the genetic algorithm (GA) method to solve the problem. A case study of a power station coal transportation system is presented to illustrate the proposed methodology. Finally, comparisons between the strategies with imperfect and without imperfect maintenance are given, and it is shown that incorporating imperfect maintenance quality into selective maintenance provides better solutions.