UAVs for Spatial Modelling and Urban Informatics 3031541138, 9783031541131

Citation preview

Tony H. Grubesic · Jake R. Nelson · Ran Wei

UAVs for Spatial Modelling and Urban Informatics

UAVs for Spatial Modelling and Urban Informatics

Tony H. Grubesic • Jake R. Nelson • Ran Wei

UAVs for Spatial Modelling and Urban Informatics

Tony H. Grubesic Center for Geospatial Sciences, School of Public Policy University of California at Riverside Riverside, CA, USA

Jake R. Nelson Department of Geosciences Auburn University Auburn, AL, USA

Ran Wei Center for Geospatial Sciences, School of Public Policy University of California at Riverside Riverside, CA, USA

ISBN 978-3-031-54113-1    ISBN 978-3-031-54114-8 (eBook) https://doi.org/10.1007/978-3-031-54114-8 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

Preface

Well, this is our second run at a book about drones. We welcome Ran Wei (UC Riverside) to the team in this effort. Her insights on spatial analysis were invaluable to our journey. In our last book's preface, we discussed some common misconceptions about why authors write books. Again, we are not looking to become rich and famous, tour the world on a lecture circuit, or appear on late-night talk shows bloviating about this book and why it matters. Our motivations remain simple. We think UAVs are fantastic. We love flying them, we value the data they produce, we love making maps, and solving tricky human-environmental, urban planning, and policy problems with spatial analysis. Once again, if you like these things too, this might be the book you are looking for. Similar to our last effort, sUAS technology is moving so quickly that it is almost impossible to keep up. By the time Springer publishes this book, there will be substantial advances to UAV platform capabilities and many new ways to consume and derive value from UAV imagery. These advances are undoubtedly exciting but remain frustrating for us as authors as we struggle to keep up with the latest developments. As always, we encourage authors to dig deeper and use this book as a foundation for thinking about creative ways to solve complex urban problems with UAV and GIS technology. Much like our last book, this book should appeal to a wide range of scholars, including geographers, planners, sociologists, criminologists, environmental scientists, and any other discipline that may, at least from time to time, require sUAS as part of their research portfolio. We also believe this book would be ideal for instructors developing and teaching an applications course using UAVs for urban analysis. Our work is ongoing. We hope readers enjoy this book as much as we liked writing it. Let's get flying! Riverside, CA, USA Auburn, AL, USA Riverside, CA, USA December 2023

Tony H. Grubesic Jake R. Nelson Ran Wei

v

Acknowledgments

We want to express our thanks to several people. First, thanks to Eddie Helderop (UC Riverside) for his fantastic data wrangling and analysis skills. We could not have completed this book without him. Second, thanks go to Lu Liang (UC Berkeley) and John South (University of North Texas) for helping organize and support our fieldwork in Texas. Third, we thank Karissa Bremer at RDO Equipment in Riverside, Ca, for providing us with the necessary equipment to complete our work. Finally, thanks to all our students who humor early morning (but not cool temperatures) flights. A special shout-out to all the fire ants that attacked our ankles in Texas, the Southlake porcupine, and a quick hello to the SAME dude who asked us to violate federal statute 18 USC Section 1702 in 2018. We saw this guy riding his dirt bike again, but he said something about the “children crying” this time. Whatever bro…

vii

Contents

1

An Overview of UAVs for Spatial Modeling and Urban Informatics������������������������������������������������������������������������������������������������    1 1.1 Introduction��������������������������������������������������������������������������������������    1 1.2 UAVs as Mobile Sensors������������������������������������������������������������������    2 1.3 UAVs and the City����������������������������������������������������������������������������    3 1.4 Methods, Tools, Data, and Process ��������������������������������������������������    5 1.5 Results����������������������������������������������������������������������������������������������    6 1.5.1 Keyword Citation Clusters����������������������������������������������������    6 1.5.2 Citation Bursts, Degree, Centrality, and Sigma��������������������    8 1.6 Discussion and Conclusion ��������������������������������������������������������������   11 1.7 Book Organization����������������������������������������������������������������������������   13 References��������������������������������������������������������������������������������������������������   14

2

UAV Operating Environments����������������������������������������������������������������   17 2.1 Introduction��������������������������������������������������������������������������������������   17 2.2 UAV Operators����������������������������������������������������������������������������������   18 2.2.1 Recreational Operators���������������������������������������������������������   19 2.2.2 Commercial (Part 107) Operators����������������������������������������   21 2.2.3 Government Operators����������������������������������������������������������   21 2.3 UAV Technology������������������������������������������������������������������������������   22 2.3.1 Sensors����������������������������������������������������������������������������������   23 2.3.2 UAV Platform Changes��������������������������������������������������������   25 2.3.3 Remote ID (RID)������������������������������������������������������������������   26 2.4 UAV Flight Operations ��������������������������������������������������������������������   28 2.4.1 Operations Over People��������������������������������������������������������   28 2.4.2 Flying at Night����������������������������������������������������������������������   30 2.5 Conclusion����������������������������������������������������������������������������������������   31 References��������������������������������������������������������������������������������������������������   31

3

 UAVs for Monitoring Property Code Violations ����������������������������������   33 3.1 Introduction��������������������������������������������������������������������������������������   33 3.2 Systems for Urban Monitoring and Service Requests����������������������   34 ix

x

Contents

3.3 Study Area and Data ������������������������������������������������������������������������   36 3.3.1 UAV Flight����������������������������������������������������������������������������   37 3.3.2 Processing 311 Data��������������������������������������������������������������   37 3.4 Results����������������������������������������������������������������������������������������������   39 3.4.1 Parcel Level Analysis������������������������������������������������������������   39 3.4.2 Neighborhood-Level Analysis����������������������������������������������   43 3.5 Conclusion����������������������������������������������������������������������������������������   49 References��������������������������������������������������������������������������������������������������   50 4

 Unmasking Invisible Infrastructure Systems with UAVs ��������������������   53 4.1 Introduction��������������������������������������������������������������������������������������   53 4.2 Locator Language: A Primer������������������������������������������������������������   55 4.2.1 Color Codes for Utilities������������������������������������������������������   57 4.2.2 Locator Marking Quality������������������������������������������������������   58 4.2.3 Locator Language and Symbols�������������������������������������������   59 4.3 Study Area, Data, and Methods��������������������������������������������������������   61 4.3.1 Study Area����������������������������������������������������������������������������   61 4.3.2 Data ��������������������������������������������������������������������������������������   61 4.3.3 Methods��������������������������������������������������������������������������������   61 4.4 Results����������������������������������������������������������������������������������������������   66 4.5 Discussion ����������������������������������������������������������������������������������������   68 4.6 Conclusion����������������������������������������������������������������������������������������   71 References��������������������������������������������������������������������������������������������������   71

5

Maximizing the Thermal Comfort of Pedestrians with UAV Imagery and Multiobjective Spatial Optimization ������������������������������   73 5.1 Introduction��������������������������������������������������������������������������������������   73 5.2 Background ��������������������������������������������������������������������������������������   74 5.3 Study Area and Data ������������������������������������������������������������������������   76 5.4 Methods��������������������������������������������������������������������������������������������   77 5.4.1 Street Shading ����������������������������������������������������������������������   77 5.4.2 Path Identification ����������������������������������������������������������������   80 5.5 Results����������������������������������������������������������������������������������������������   82 5.6 Discussion and Conclusion ��������������������������������������������������������������   83 References��������������������������������������������������������������������������������������������������   86

6

 Evaluating Rooftop Solar Energy Inequity with UAVs������������������������   89 6.1 Introduction��������������������������������������������������������������������������������������   89 6.2 Background ��������������������������������������������������������������������������������������   91 6.2.1 Not All Rooftops Are Created Equal������������������������������������   91 6.2.2 Energy Equity ����������������������������������������������������������������������   92 6.3 Study Area, Data, and Methods��������������������������������������������������������   93 6.3.1 UAV Mission Details������������������������������������������������������������   95 6.3.2 Data Preparation�������������������������������������������������������������������   96 6.3.3 Solar Analysis ����������������������������������������������������������������������   97

Contents

xi

6.4 Results����������������������������������������������������������������������������������������������  101 6.5 Discussion and Conclusion ��������������������������������������������������������������  104 References��������������������������������������������������������������������������������������������������  105 7

 UAVs for Rapid Storm Damage Assessment������������������������������������������  107 7.1 Introduction��������������������������������������������������������������������������������������  107 7.2 Background ��������������������������������������������������������������������������������������  109 7.3 Study Area, Data, and Methods��������������������������������������������������������  111 7.3.1 Study Area����������������������������������������������������������������������������  111 7.3.2 Data ��������������������������������������������������������������������������������������  111 7.3.3 Methods��������������������������������������������������������������������������������  113 7.4 Results����������������������������������������������������������������������������������������������  113 7.4.1 Area 1������������������������������������������������������������������������������������  113 7.4.2 Area 2������������������������������������������������������������������������������������  114 7.4.3 Areas 3 and 4������������������������������������������������������������������������  117 7.4.4 Limitations����������������������������������������������������������������������������  117 7.5 Discussion and Conclusion ��������������������������������������������������������������  118 References��������������������������������������������������������������������������������������������������  121

8

Identifying Critical Micro-infrastructures��������������������������������������������  123 8.1 Introduction��������������������������������������������������������������������������������������  123 8.2 Critical Infrastructure and Key Resources (CIKR)��������������������������  125 8.2.1 CIKR Vulnerability ��������������������������������������������������������������  126 8.2.2 Interdependencies ����������������������������������������������������������������  127 8.2.3 Network Resilience��������������������������������������������������������������  128 8.2.4 Tracking and Cataloging Infrastructure Elements����������������  129 8.3 Study Area, Data, and Methods��������������������������������������������������������  130 8.3.1 Study Area����������������������������������������������������������������������������  130 8.3.2 Data ��������������������������������������������������������������������������������������  131 8.3.3 Image Analysis and Deep Learning Model��������������������������  132 8.4 Results����������������������������������������������������������������������������������������������  136 8.4.1 Catalogued Micro-infrastructure������������������������������������������  136 8.4.2 Uncatalogued Micro-infrastructure��������������������������������������  138 8.5 Discussion, Limitations, and Conclusions����������������������������������������  141 8.5.1 Public Policy for Community Resilience������������������������������  143 8.5.2 Conclusion����������������������������������������������������������������������������  144 References��������������������������������������������������������������������������������������������������  144

9

Drones and Their Future Applications��������������������������������������������������  149 9.1 Introduction��������������������������������������������������������������������������������������  149 9.2 Predictions����������������������������������������������������������������������������������������  150 9.2.1 Commerce and Logistics������������������������������������������������������  150 9.2.2 Geospatial Intelligence���������������������������������������������������������  152 9.2.3 Disaster Response ����������������������������������������������������������������  154

xii

Contents

9.2.4 Precision Agriculture������������������������������������������������������������  156 9.2.5 Medical Services������������������������������������������������������������������  157 9.3 UAV Policy and Personal Privacy����������������������������������������������������  159 9.3.1 UAV Policy and BVLOS������������������������������������������������������  159 9.3.2 Personal Privacy and Facial Recognition������������������������������  160 References��������������������������������������������������������������������������������������������������  161 Index������������������������������������������������������������������������������������������������������������������  169

Chapter 1

An Overview of UAVs for Spatial Modeling and Urban Informatics

1.1 Introduction Spatial modeling is a relatively generic term that refers to using various tools for conducting geospatial analysis, often in conjunction with a geographic information system (GIS). These tools include basic statistical and geocomputational analysis but often involve more advanced modeling frameworks such as agent-based modeling (ABM), artificial intelligence (AI), spatial optimization, and multicriteria decision-­making analysis (MCDA). In short, one can consider spatial models as “formal languages” to design, implement, and execute geospatial analysis workflows (Qiang, 2021). Moreover, spatial models help analysts represent a range of urban, environmental, cultural, social, and economic processes for evaluating current conditions or estimating future trends. Urban informatics is an emergent, interdisciplinary science that seeks to develop theories and methods for improving the design, management, and functionality of cities (Shi et al., 2021). Urban informatics draws upon many domains, including urban science, geography, geomatics, engineering, and a wide range of natural, socio-economic, and planning sciences (e.g., sociology, economics, environmental science, and forestry) to deepen our understanding of urban spaces. Standing alone, both spatial modeling and urban informatics can provide substantial insight into the way cities work. However, the fusion of spatial modeling and urban informatics is particularly potent for identifying patterns and their underlying processes in urban spaces. First, sensors are now ubiquitous in cities. For example, sensors enable local agencies to monitor air quality (Liu et al., 2020) and wastewater effluence for pollen or pathogens (Kadadou et al., 2022), track pedestrian flows (Huang et al., 2021), and maintain free-flowing automobile and truck traffic during rush hour (Fu et al., 2020). Second, the massive data streams that sensors generate mean new tools and models are necessary to ingest, analyze, visualize, and extract actionable information and intelligence from these data flows. Finally, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. H. Grubesic et al., UAVs for Spatial Modelling and Urban Informatics, https://doi.org/10.1007/978-3-031-54114-8_1

1

2

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

because much of these data flow in real-time, novel spatial modeling approaches can leverage them for evaluating individual mobility, the spatiotemporal dynamics of urban pollution, and the spread of pathogens through time and space. In short, stitching all of these data, tools, and outputs together is central to urban informatics and critical for deepening our understanding of urban metabolisms (Shi et  al., 2021) – but the addition of spatial modeling radically improves how we think about and reason within the city.

1.2 UAVs as Mobile Sensors Unmanned aerial vehicles (UAVs) represent an incredible technological platform for blending the best of urban informatics and spatial modeling for generating geospatial intelligence data pipelines. Currently, most urban sensing equipment exists in a fixed location. Consider, for example, traffic cameras, weather stations, and air quality monitoring equipment. Most of this sensing infrastructure is mounted on supporting structures. For example, many traffic cameras are mounted on signal lighting poles, and air quality sensors often mount on utility poles. These fixed locations provide several benefits for collecting data. First, fixed locations allow stakeholders to track and monitor any changes in a geographical context that might impact sensor performance (e.g., new construction, expanding tree canopy, etc.). Second, fixed locations allow stakeholders to create time-series databases where the variable of interest (e.g., traffic flow at a single intersection) is measured at different points in time (e.g., weekly). Third, fixed locations allow optimizing sensor locations and maximizing spatial coverages while adhering to a hardware budget (Sangwan & Singh, 2015). Finally, recent empirical work also suggests that fixed sensors outperform mobile sensors for specific applications. For example, Yang and Bou-Zeid (2019) suggest that evenly distributed fixed (EDF) sensor networks are more accurate for estimating periods of extreme temperature events than mobile sensors. However, fixed sensor networks do exhibit limitations. Most notably, these systems often underrepresent spatial information. Specifically, sensors can be expensive, and blanketing an entire city with them is inefficient and uneconomical. Further, it can be challenging to obtain siting permissions as the size of a fixed network grows. As a result, there are often numerous spatial gaps in fixed sensor networks. Mobile sensors, including those connected to unmanned aerial systems (UAS), can overcome these limitations by adding observations, measurements, and contextual spatial information for places lacking fixed sensors. For example, Yang and Bou-Zeid (2019) note that the addition of mobile sensors added the necessary spatial information to improve the accuracy of temperature measurements in multiple urban areas. Thus, when combining their fixed network with a mobile network of sensors  – the hybrid system outperformed both wholly fixed or completely mobile sensor systems.

1.3  UAVs and the City

3

Ultimately, most urban informatics and spatial modeling applications benefit from reductions in data uncertainty, more complete spatial coverages, and the fusion of fixed and mobile sensing data. This outcome is one of the reasons that the use of unmanned aerial systems continues to grow in urban areas – they excel in collecting data in places that are difficult to reach or impossible to allocate fixed sensors for monitoring.

1.3 UAVs and the City Academic research also reflects the growth of UAV use in urban areas. For example, a Clarivate Web of Science search of the term “UAVs or drones and urban” revealed that 841 articles were published in 2015. Five years later, in 2020, there were 2996 articles published with these keywords, and in 2022 there were 3816.1 Figure 1.1 highlights the treemap chart of the research domains contributing most to drone research (not inclusive) between 2013 and 2023. While many of these substantive areas seem obvious (e.g., electrical engineering, remote sensing, environmental science), the prevalence of drone research in fields such as telecommunications may surprise some readers. Aside from the need to use telecommunications to control drones in flight, scholars increasingly view UAVs as core components of the Internet of Things (IoT) (Alsamhi et al., 2019). In this context, drones and the IoT support a range of smart city applications.

Fig. 1.1  Treemap of research domains contributing to urban research with drones: 2013–2023

 We recognize that “drone” may also refer to male honeybees or ants, but this is for illustrative purposes only. 1

4

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

such as communication, transportation, agriculture, safety, security, service delivery, weather monitoring, and healthcare – leveraging their capabilities as sensors and haulers. Embedded within the IoT is the Internet of Drones (IoD), which refers to the infrastructure designed to provide control and access between users and drones over telecommunications systems (e.g., the Internet), but also allows for coupling multiple UAVs for a single mission and scalable data offloading to cloud storage systems (Abdelmaboud, 2021). These varied connections to telecommunications and other fields indicate the role of UAVs as a general-purpose technology (GPT). Specifically, Lipsey et  al. (2005) define general-purpose technologies as (1) single, generic technologies, recognizable as such over its whole lifetime, (2) initially having much scope for improvement and eventually coming to be widely used in the economy, (3) having many uses, and (4) generating many spillover effects. Grubesic and Nelson (2020) note that drones closely adhere to this GPT typology and, as a result, support a wide range of urban applications. In particular, the advancing smart cities paradigm has many potential connections to drones and the development of geospatial intelligence portfolios. By definition, a city is smart when “investments in human and social capital and traditional (transport) and modern (ICT) communication infrastructure fuel sustainable economic growth and high quality of life, with a wide management of natural resources, through participatory governance” (Caragliu et al., 2011, p. 70). To date, there are many literature reviews and syntheses that cover drones and their connections to smart cities (Alsamhi et  al., 2019), general applications of drones (Ayamga et  al., 2021), drone futures (Elmeseiry et  al., 2021), and drone typologies (Hassanalian & Abdelkefi, 2017). However, rather than continue with the usual approach, this chapter aims to provide a more holistic (but brief) evaluation of drones and their urban applications as a knowledge domain. This departure is essential for several reasons. First, it provides a foundation for the book by shedding light on the complexities of drone use in urban areas – highlighting the many ways in which the subdomains highlighted in Fig.  1.1 are interconnected. Second, it will help identify the important scientific work that forms the connective tissue of this research domain. Finally, by identifying critical contributions to the literature, it will be possible to extrapolate and identify potential future trends for UAVs in spatial modeling and urban informatics. We will draw upon scientometrics to accomplish these tasks  – which analyze science, innovation, technology, and their production (Abramo, 2018). Specifically, we will utilize scientometric techniques (Nelson & Grubesic, 2018; Wei et al., 2015) to map the research related to the urban applications of drones. This analysis includes co-citation and co-occurrence networks to provide an overview of this evolving research domain between 2013 and 2023.

1.4  Methods, Tools, Data, and Process

5

1.4 Methods, Tools, Data, and Process As mentioned earlier, scientometrics is a field that seeks to deepen our understanding of knowledge production and its interconnections, both within and across scientific domains. We aim to create a functional roadmap of how urban informatics and unmanned aerial systems are helping fuel important research in cities. As highlighted in Fig. 1.1, these contributions likely originate in a wide range of scientific disciplines, so we will leverage the Web of Science (WOS) to disentangle the essential contributions in these domains, focusing on the last decade, 2013–2023. We also use the power of CiteSpace, a desktop application written in Java, to visualize and analyze the bibliographic linkages between authors, keywords, and references using the citation information drawn from the WOS (Chen, 2006). Specifically, we will focus on four key metrics for this analysis. First, we will identify key contributions using betweenness centrality (BC). BC captures the boundary-­ spanning potential of papers that can lead to shifts in research or theories (Chen et  al., 2009). Higher betweenness centrality scores suggest that a given paper is better at connecting varied research domains. Second, we will measure the burst levels for each paper to determine whether or not a manuscript received an above-­ average number of citations for a given year. For example, a paper receiving two or three citations during the first 3 years after publication would have a low burst level. However, a paper that receives ten or more citations each year after publication would have a much higher burst level. Sustained bursts typically indicate that a paper is (and remains) essential to the research domain. We will also rely on the sigma measure, which measures the combined strength of structural and temporal properties of a node – combining betweenness centrality and citation bursts (Chen et  al., 2009). Lastly, we will focus on using article keywords and the clustering capabilities within CiteSpace to identify essential and emergent research themes for urban applications of UAVs. We built the urban/UAV knowledge domain network using the Web of Science and CiteSpace. The WOS query was “UAV and urban” from January 2013 to April 2023. We retrieved 1076 articles using this search and mined all the bibliometric information from this corpus of papers for analysis. After pre-processing and sensitivity testing, we included 12,594 references (98.12% of those retrieved) and generated a network of 426 nodes with 1257 links.2 It is essential to acknowledge that while the WOS is a reasonably comprehensive and complete resource for building bibliometric databases, it fails to include all articles for all journals and all conferences. Empirical analysis also suggests that the WOS exhibits a positive bias toward scientific literature published in the United States (Mongeon & Paul-Hus, 2016). While this suggests some bias in our data,

 The 241 references were deemed invalid due to missing author information, lack of a DOI, or related problems. 2

6

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

CiteSpace is agnostic regarding citation information. Thus, even when an article references a paper outside the WOS index, CiteSpace will still include that publication in the citation network.

1.5 Results 1.5.1 Keyword Citation Clusters Figure 1.2 highlights the generated citation network, organized by keyword citation clusters for 2013–2023. For example, the manuscripts belonging to the largest cluster (#0) share “drone” as a keyword, while the second largest cluster (#1) shares the keyword “solar-powered UAVs.” There are three key metrics for helping disambiguate each of the clusters. The first is the term frequency*inverse document frequency (TF*IDF) metric, which searches for common terms in all of the manuscripts and ranks the importance of these terms based on their frequency of use (Ramos, 2003). A second way to differentiate clusters is through latent semantic indexing (LSI), which explores the higher-order semantic structure of documents and decomposes them into weighted combinations of terms and their importance (Deerwester et al., 1990). Finally, one can use log-likelihood ratio (LLR) tests (Dunning, 1993) to identify unique word combinations and thematic coverages within the clusters (Lee et al., 2016). When one reexamines Fig.  1.2 with the context that Table  1.1 provides, it becomes easier to identify the major specialties in UAV research between 2013 and 2023. We label each cluster in Fig. 1.2 with its top LLR term, but as detailed in Table 1.1, there are multiple key labels for each cluster – all of which one can consider to be threads of the primary specialties within this research domain. This

Fig. 1.2  Keyword citation clusters for “UAV and Urban”: 2013–2023

1.5 Results

7

Table 1.1  Urban/UAV research domains and clusters with keywords Mean Cluster Size Silhouette (Year) LSI 0 57 0.789 2019 Unmanned aerial vehicles; resource management; resource optimization; uplink throughput; altitude decision 1 42 0.858 2018 Unmanned aerial vehicles; autonomous systems; solar-­ powered UAVs; wireless communication; rapidly exploring random tree navigation 2 38 0.912 2016 Path planning; urban areas; task analysis; cooperative systems; unmanned aerial vehicles 3

34

0.876

2017

Unmanned aerial vehicle; reshuffling strategy; seamless coverage; cellular networks; path planning

5

21

0.988

2018

6

19

0.847

2016

Sensor networks; aerial vehicle; atmospheric modeling; pollution measurement; location-awareness Collision avoidance; unmanned aerial vehicles; fleet path planning; trajectory planning; jitter problem

7

19

0.901

2019

Unmanned aerial vehicle; urban environments; coverage path planning; sparse waypoint; traveling salesman problem

8

18

0.886

2020

Vehicular ad; unmanned aerial vehicles; task analysis; energy efficiency; crowd emotion prediction;

9

14

0.997

2014

Cover; flood risk; areas; remote sensing images;

11

10

0.889

2020

Collision avoidance; unmanned aerial vehicles; reinforcement learning; vehicular ad; multiaccess-edge computing

LLR Drone; unsupervised learning; coverage probability; stochastic geometry Solar-powered UAVs; unmanned aerial vehicles (UAVs); autonomous systems; urban forest; tree height Target tracking; urban areas; path planning; task analysis; cooperative control Unmanned aerial vehicles; received signal strength (RSS); wireless channels; radio-wave propagation; ray tracing Air quality; unmanned aerial vehicle; pollution; land surface temperature; UAV platform Collision avoidance; enhanced potential field; fleet path planning; indoor positioning; collision prediction Coverage path planning; urban logistics; visual; delivery of courier and postal items; particulate matter Structural health monitoring; delay; discrete-space consensus algorithm; UAV-based inspection; traffic monitoring Areas; flood risk; classification; cover; vegetation mapping Machine learning; reinforcement learning; collision avoidance; artificial intelligence; adversarial control

8

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

interpretation is similar for the LSI terms. Finally, higher silhouette values in Table 1.1 suggest that the cluster is distinct from its peers (Kaufman & Rousseeuw, 1990), and the Mean Year column represents the average year of publication for the papers in the cluster. While all of the silhouette values are quite high, the values for Clusters 5 (0.988) and 9 (0.997) are the highest, suggesting that the articles in these clusters most closely relate to our search term “UAVs and urban.” For example, Cluster 5 (n = 21) focuses on sensor networks, evaluating air quality and land surface temperature – all substantial concerns in urban environments. Similarly, Cluster 9 (n = 14) primarily focuses on using UAVs for evaluating flood risk, land cover classification, and vegetation mapping. Cluster 0 (n = 57) is the largest grouping and focuses on UAV platforms and their ability to interface with command and control communication systems (e.g., cellular networks) and their potential civil applications (e.g., search and rescue, infrastructure inspection), as well as insights on addressing challenges such as collision avoidance and network security.

1.5.2 Citation Bursts, Degree, Centrality, and Sigma Aside from the keyword clusters, there are many important individual papers worth highlighting because they represent essential contributions to the literature on urban applications for UAVs. Identifying these papers necessitates a second citation analysis, focusing on authors and references rather than keywords. Figure 1.3 highlights

Fig. 1.3  Subject category clusters and key author contributions

1.5 Results

9

Table 1.2  Citation bursts for urban/UAV research papers Bursts References 4.43 Mozaffari, M., Saad, W., Bennis, M., Nam, Y. H., & Debbah, M. (2019). A tutorial on UAVs for wireless networks: Applications, challenges, and open problems. IEEE Communications Surveys & Tutorials, 21(3), 2334–2360. 3.49 Al-Hourani, A., Kandeepan, S., & Jamalipour, A. (2014, December). Modeling air-to-ground path loss for low altitude platforms in urban environments. In 2014 IEEE Global Communications Conference (pp. 2898–2904). IEEE. 3.49 Al-Hourani, A., Kandeepan, S., & Lardner, S. (2014). Optimal LAP altitude for maximum coverage. IEEE Wireless Communications Letters, 3(6), 569–572. 3.07 Colomina, I., & Molina, P. (2014). Unmanned aerial systems for photogrammetry and remote sensing: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 92, 79–97. 2.70 Zeng, Y., Zhang, R., & Lim, T. J. (2016). Wireless communications with unmanned aerial vehicles: Opportunities and challenges. IEEE Communications Magazine, 54(5), 36–42. 2.67 Primatesta, S., Guglieri, G., & Rizzo, A. (2019). A risk-aware path planning strategy for UAVs in urban environments. Journal of Intelligent & Robotic Systems, 95, 629–643. 2.58 Shakhatreh, H., Sawalmeh, A. H., Al-Fuqaha, A., Dou, Z., Almaita, E., Khalil, I., ... & Guizani, M. (2019). Unmanned aerial vehicles (UAVs): A survey on civil applications and key research challenges. IEEE Access, 7, 48572–48634.

Cluster Begin/End 2 2021/2023

0

2017/2019

0

2017/2019

9

2015/2018

2

2018/2019

0

2020/2021

6

2021/2023

this alternative perspective with subject categories (rather than keywords) as the cluster labels. Node sizes correspond to citation bursts. For example, Table 1.2 displays these key papers’ seven highest burst scores and their begin/end dates. Mozaffari et al. (2019a, b) ranks the highest with a score of 4.43 and, as of May 2023, has 1860 citations, according to Google Scholar. This paper focused on using UAVs with wireless communication systems and their potential to serve as flying base stations for enabling real-time video streaming and related tasks via cellular networks. Similarly, the Al-Hourani, et al. (2014a, b) contributions, each with a burst score of 3.49, also focused on the potential of UAVs to function as air-to-ground wireless service hubs, especially to assist in the recovery of damaged terrestrial wireless infrastructure after an extreme event (e.g., earthquake). The early contribution from Colomina and Molina (2014) provided a review on UAVs and focused on their uses for photogrammetry and remote sensing. With a burst score of 3.07, its burst phase began one year after publication and lasted four years, from 2015 to 2018. As of May 2023, it has an accumulated citation count of more than 2800 and continues to provide a balanced perspective on UAVs and their use in remote sensing. Zeng et al. (2016) provided an early snapshot of the potential for UAVs and wireless communication systems, focusing on network architecture, channel characteristics, and the trade-offs associated with using energy-constrained UAVs for reliable system service. Finally, both the Primatesta et  al. (2019) and

10

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

Shakhatreh et al. (2019) papers directly address the use of UAVs in urban environments, with the former focusing on an approach for path planning to improve the safety of operations over people and the latter providing a holistic review of potential civil applications for UAVs, from their use in communications systems and remote sensing to infrastructure inspection, precision agriculture, and the delivery of goods. The nodal degree (Table 1.3) and centrality (Table 1.4) scores reveal a different array of key papers. Of particular interest is the Menouar et al. (2017) paper, which has a nodal degree of 35 and a centrality score of 0.10. This work explores the potential applications of UAVs for intelligent transportation systems (ITS) in smart cities, addressing deployment optimization, security, and privacy challenges. In many ways, one can view Aggarwal and Kumar (2020), which has a nodal degree of 33 and a centrality score of 0.10, as a companion piece focusing on UAV path planning techniques. In this context, path planning aims to find the optimal and shortest path – simultaneously guaranteeing a collision-free environment. Both papers collectively lay a foundation for addressing many smart city challenges, including urban observation, infrastructure evaluation, transport management, goods delivery, and many others. Further, both of these papers serve as critical connective tissue to much of the allied work concerning urban applications for UAVs. Finally, Table 1.5 highlights the sigma measure, which captures the combined strength of structural and temporal properties of a node – combining betweenness centrality and citation bursts (Chen et  al., 2009). Mozaffari et  al. (2019a, b) and Table 1.3  Structural importance of papers focusing on urban/UAV research Degree References 35 Menouar, H., Guvenc, I., Akkaya, K., Uluagac, A. S., Kadri, A., & Tuncer, A. (2017). UAV-enabled intelligent transportation systems for the smart city: Applications and challenges. IEEE Communications Magazine, 55(3), 22–28. 33 Aggarwal, S., & Kumar, N. (2020). Path planning techniques for unmanned aerial vehicles: A review, solutions, and challenges. Computer Communications, 149, 270–299. 30 Aanæs, H., Jensen, R. R., Vogiatzis, G., Tola, E., & Dahl, A. B. (2016). Large-scale data for multiple-view stereopsis. International Journal of Computer Vision, 120, 153–168. 30 Fan, X., Zhang, L., Brown, B., & Rusinkiewicz, S. (2016). Automated view and path planning for scalable multi-object 3D scanning. ACM Transactions on Graphics (TOG), 35(6), 1–13. 30 Battaglia, P. W., Hamrick, J. B., & Tenenbaum, J. B. (2013). Simulation as an engine of physical scene understanding. Proceedings of the National Academy of Sciences, 110(45), 18327–18332. 30 Dosovitskiy, A., Ros, G., Codevilla, F., Lopez, A., & Koltun, V. (2017, October). CARLA: An open urban driving simulator. In Conference on Robot Learning (pp. 1–16). PMLR. 30 Alsadik, B., Gerke, M., & Vosselman, G. (2013). Automated camera network design for 3D modeling of cultural heritage objects. Journal of Cultural Heritage, 14(6), 515–526.

Cluster 5

8

1

1

1

1

1

1.6  Discussion and Conclusion

11

Table 1.4 Betweenness centrality: papers functioning as connective tissue between research domains Centrality References 0.10 Menouar, H., Guvenc, I., Akkaya, K., Uluagac, A. S., Kadri, A., & Tuncer, A. (2017). UAV-enabled intelligent transportation systems for the smart city: Applications and challenges. IEEE Communications Magazine, 55(3), 22–28. 0.10 Aggarwal, S., & Kumar, N. (2020). Path planning techniques for unmanned aerial vehicles: A review, solutions, and challenges. Computer Communications, 149, 270–299. 0.05 Fotouhi, A., Qiang, H., Ding, M., Hassan, M., Giordano, L. G., Garcia-­ Rodriguez, A., & Yuan, J. (2019). Survey on UAV cellular communications: Practical aspects, standardization advancements, regulation, and security challenges. IEEE Communications Surveys & Tutorials, 21(4), 3417–3442. 0.05 Shakhatreh, H., Sawalmeh, A. H., Al-Fuqaha, A., Dou, Z., Almaita, E., Khalil, I., ... & Guizani, M. (2019). Unmanned aerial vehicles (UAVs): A survey on civil applications and key research challenges. IEEE Access, 7, 48572–48634. 0.04 Dorling, K., Heinrichs, J., Messier, G. G., & Magierowski, S. (2016). Vehicle routing problems for drone delivery. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 47(1), 70–85. 0.04 Alsamhi, S. H., Ma, O., Ansari, M. S., & Almalki, F. A. (2019). Survey on collaborative smart drones and internet of things for improving smartness of smart cities. IEEE Access, 7, 128125–128152. 0.04 Zeng, Y., Lyu, J., & Zhang, R. (2018). Cellular-connected UAV: Potential, challenges, and promising technologies. IEEE Wireless Communications, 26(1), 120–127.

Cluster 5

8

0

6

5

4

0

Shakhatreh et al. (2019) reemerge as structurally important nodes in the networks with a growing number of citations (i.e., burst). It is essential to note that both papers are relatively recent contributions, and their relative importance to the literature is still growing. Again, both bursts began in 2021 (Table 1.2), but we cannot know (yet) when their bursts will end. Currently, both are carrying into 2023, but more data is required (e.g., 2024, 2025, etc.) to determine how long their respective bursts might last.

1.6 Discussion and Conclusion The citation analysis for urban applications of UAVs is revealing. While there is undoubtedly a large corpus of research on the subject, many of the most important papers are review-oriented or speculative pieces highlighting the potential applications of UAVs in smart cities for intelligent transportation systems, infrastructure inspection, package delivery, and many other domains. Much of the other work in this domain focuses on the technical and engineering elements of UAV command/ control and communication systems – critically important factors for safely flying these platforms throughout populated regions.

12

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

Table 1.5  Sigma: a measure of structural and temporal importance for the urban/UAV literature Sigma References 1.15 Mozaffari, M., Saad, W., Bennis, M., Nam, Y. H., & Debbah, M. (2019). A tutorial on UAVs for wireless networks: Applications, challenges, and open problems. IEEE Communications Surveys & Tutorials, 21(3), 2334–2360. 1.13 Shakhatreh, H., Sawalmeh, A. H., Al-Fuqaha, A., Dou, Z., Almaita, E., Khalil, I., ... & Guizani, M. (2019). Unmanned aerial vehicles (UAVs): A survey on civil applications and key research challenges. IEEE Access, 7, 48572–48634. 1.07 Al-Hourani, A., Kandeepan, S., & Jamalipour, A. (2014, December). Modeling air-to-ground path loss for low altitude platforms in urban environments. In 2014 IEEE Global Communications Conference (pp. 2898–2904). IEEE. 1.02 Zeng, Y., Zhang, R., & Lim, T. J. (2016). Wireless communications with unmanned aerial vehicles: Opportunities and challenges. IEEE Communications Magazine, 54(5), 36–42. 1.01 Primatesta, S., Guglieri, G., & Rizzo, A. (2019). A risk-aware path planning strategy for UAVs in urban environments. Journal of Intelligent & Robotic Systems, 95, 629–643. 1.01 Al-Hourani, A., Kandeepan, S., & Lardner, S. (2014). Optimal LAP altitude for maximum coverage. IEEE Wireless Communications Letters, 3(6), 569–572. 1.00 Menouar, H., Guvenc, I., Akkaya, K., Uluagac, A. S., Kadri, A., & Tuncer, A. (2017). UAV-enabled intelligent transportation systems for the smart city: Applications and challenges. IEEE Communications Magazine, 55(3), 22–28.

Cluster 2

6

0

2

0

0 5

Unfortunately, largely missing from this literature are real-world, urban applications of unmanned aerial systems for geographic observation, spatial modeling, and urban informatics. Grubesic and Nelson (2020) highlighted the many profound challenges and constraints of conducting fieldwork with UAVs in urban areas. However, with (1) rapid increases in drone autonomy, which helps reduce human error, (2) lighter platforms, which help reduce the potential physical damage from potential human/drone interactions and impacts, and (3) a growing need to collect and analyze data at a scale granular enough to capture the nuances of local urban phenomena, there has never been a more exciting time for pursuing UAV research in urban areas. In fact, UAVs offer a promising mechanism to fill gaps in our understanding of urban areas, serving as a measurement platform that can rapidly and inexpensively collect data and monitor changes in cities. Again, their use is fraught with social, operational, regulatory, and technical challenges for successful deployments – but these constraints are not so onerous that stakeholders should avoid pursuing innovative work and the construction of geospatial intelligence portfolios with UAVs. Quite the opposite is true. Applied urban analyses with UAVs can potentially improve cities, the quality of life for their residents, and city operations in meaningful and measurable ways. This book provides a resource for urbanists (e.g., planners, geographers, sociologists, epidemiologists, engineers), educators, and students who work with geographic information and seek to enhance their work with information from unmanned aerial vehicles. At the same time, we provide operational and methodological frameworks for carrying out these advanced analyses in a

1.7  Book Organization

13

manner that considers the challenges of incorporating UAVs in research within the urban environment. Finally, we provide seven unique applications of UAVs for urban analysis, detailing relevant policy and empirical questions, mission parameters, data collection, spatial modeling, and the associated empirical results. Further, we discuss how best to integrate these results into actionable geospatial intelligence and policy development to improve city infrastructure systems, sustainability, the environment, and neighborhood quality.

1.7 Book Organization In the next chapter, we provide an overview and update on the current state of UASs in the United States and abroad. This update includes an overview of market conditions and projected growth. In addition, this chapter provides updates on recent developments in Federal Aviation requirements for remote identification and the importance of UAS facility maps for determining where the FAA may authorize part 107 UAS operations without additional safety analysis. Chapter 3 highlights the use of UAVs for monitoring 311 complaints and violations. The 311 system provides telephone access to non-emergency municipal services. Many cities also provide a web-based system for accessing 311 services, where residents can report abandoned vehicles, code violations, rubbish dumping, graffiti, landscaping issues, or problems with municipal infrastructure, such as loose manholes or malfunctioning streetlights. UAVs provide a platform for monitoring compliance efforts throughout the community, at scale, with superior efficiency – including violations in locations that may be difficult to see from the street (e.g., semi-public spaces). Chapter 4 explores a unique approach for tracking urban infrastructure projects without accessing municipal permit systems or interfacing with private infrastructure providers (e.g., cable) or infrastructure subcontractors that may (or may not) be engaging with public-facing permits. Specifically, we detail the nuances of underground utility markings and the ability of UAVs to both sense and monitor potentially disruptive digging projects at the neighborhood level. In Chap. 5, we develop a spatial optimization model using hyper-local UAS-­ derived data for reducing pedestrian heat exposure and stress. The “cool walk” model leverages tree-canopy information from a UAS-derived point cloud and digital orthophotography combined with origin and destination data to identify the k-coolest thermal paths for pedestrians. We highlight the potential of the cool walk model for school children in Lewisville, Texas. In Chap. 6, we explore the impact of structural limitations across neighborhoods or varying socio-economic status on rooftop solar energy output by leveraging high-­ resolution UAV data. We demonstrate how structural differences across neighborhoods limit the economic viability of rooftop solar when explicitly considering the cost of electricity, solar energy output, and solar power systems.

14

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

In Chap. 7, we explore the utility of using UAVs to rapidly assess storm damage at the neighborhood level. Specifically, we use hyper-local UAV imagery and digital surface models (DSMs) to evaluate storm damage after a monsoon event in Phoenix, Arizona. While the results are compelling, assessment challenges remain due to image processing complications and computational burdens. Chapter 8 combines the capability of GeoAI with UAV imagery for extracting and cataloging micro-infrastructure across the urban landscape. Specifically, micro-­ infrastructure refers to the smallest elements of the urban infrastructure fabric. It includes coaxial cable drops, electric transformer boxes, water meters, fiber huts, and many other visible (yet physically small) elements that urban residents depend upon for daily activities. However, because many of these infrastructure elements are poorly cataloged, any extreme event (e.g., hurricane or earthquake) that disrupts local systems often requires a prolonged and dangerous locational rediscovery process for these micro-infrastructure elements. We highlight the utility of UAV imagery and GeoAI for easing these efforts. Finally, Chap. 9 provides concluding thoughts and discusses urban futures with UAVs and other airborne sensing platforms. We explore several applications for UAVs and their policy implications – considering privacy issues and an eye toward future challenges.

References Abdelmaboud, A. (2021). The internet of drones: Requirements, taxonomy, recent advances, and challenges of research trends. Sensors, 21(17), 5718. https://doi.org/10.3390/s21175718 Abramo, G. (2018). Revisiting the scientometric conceptualization of impact and its measurement. Journal of Informetrics, 12(3), 590–597. https://doi.org/10.1016/j.joi.2018.05.001 Aggarwal, S., & Kumar, N. (2020). Path planning techniques for unmanned aerial vehicles: A review, solutions, and challenges. Computer Communications, 149, 270–299. https://doi. org/10.1016/j.comcom.2019.10.014 Al-Hourani, A., Kandeepan, S., & Jamalipour, A. (2014a). Modeling air-to-ground path loss for low altitude platforms in urban environments. In 2014 IEEE global communications conference (pp. 2898–2904). https://doi.org/10.1109/GLOCOM.2014.7037248 Al-Hourani, A., Kandeepan, S., & Lardner, S. (2014b). Optimal LAP altitude for maximum coverage. IEEE Wireless Communications Letters, 3(6), 569–572. https://doi.org/10.1109/ LWC.2014.2342736 Alsamhi, S. H., Ma, O., Ansari, M. S., & Almalki, F. A. (2019). Survey on collaborative smart drones and internet of things for improving smartness of smart cities. IEEE Access, 7, 128125–128152. https://doi.org/10.1109/ACCESS.2019.2934998 Ayamga, M., Akaba, S., & Nyaaba, A. A. (2021). Multifaceted applicability of drones: A review. Technological Forecasting and Social Change, 167, 120677. https://doi.org/10.1016/j. techfore.2021.120677 Caragliu, A., Del Bo, C., & Nijkamp, P. (2011). Smart cities in Europe. Journal of Urban Technology, 18(2), 65–82. https://doi.org/10.1080/10630732.2011.601117 Chen, C. (2006). CiteSpace II: Detecting and visualizing emerging trends and transient patterns in scientific literature. Journal of the American Society for Information Science and Technology, 57(3), 359–377. https://doi.org/10.1002/asi.20317

References

15

Chen, C., Chen, Y., Horowitz, M., Hou, H., Liu, Z., & Pellegrino, D. (2009). Towards an explanatory and computational theory of scientific discovery. Journal of Informetrics, 3(3), 191–209. https://doi.org/10.1016/j.joi.2009.03.004 Colomina, I., & Molina, P. (2014). Unmanned aerial systems for photogrammetry and remote sensing: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 92, 79–97. https:// doi.org/10.1016/j.isprsjprs.2014.02.013 Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T. K., & Harshman, R. (1990). Indexing by latent semantic analysis. Journal of the American Society for Information Science, 41(6), 391–407. https://doi.org/10.1002/(SICI)1097-­4571(199009)41:63.0.CO;2-­9 Dunning, T. (1993). Accurate methods for the statistics of surprise and coincidence. Computational Linguistics, 19(1), 61–74. https://dl.acm.org/doi/10.5555/972450.972454 Elmeseiry, N., Alshaer, N., & Ismail, T. (2021). A detailed survey and future directions of unmanned aerial vehicles (UAVs) with potential applications. Aerospace, 8(12), 363. https:// doi.org/10.3390/aerospace8120363 Fu, H., Wang, Y., Tang, X., Zheng, N., & Geroliminis, N. (2020). Empirical analysis of large-­ scale multimodal traffic with multi-sensor data. Transportation Research Part C: Emerging Technologies, 118, 102725. https://doi.org/10.1016/j.trc.2020.102725 Grubesic, T.  H., & Nelson, J.  R. (2020). UAVs and urban spatial analysis: An introduction. Springer. https://doi.org/10.1007/978-­3-­030-­35865-­5 Hassanalian, M., & Abdelkefi, A. (2017). Classifications, applications, and design challenges of drones: A review. Progress in Aerospace Sciences, 91, 99–131. https://doi.org/10.1016/j. paerosci.2017.04.003 Huang, B., Mao, G., Qin, Y., & Wei, Y. (2021). Pedestrian flow estimation through passive WiFi sensing. IEEE Transactions on Mobile Computing, 20(4), 1529–1542. https://doi.org/10.1109/ TMC.2019.2959610 Kadadou, D., Tizani, L., Wadi, V. S., Banat, F., Alsafar, H., Yousef, A. F., Barceló, D., & Hasan, S. W. (2022). Recent advances in the biosensors application for the detection of bacteria and viruses in wastewater. Journal of Environmental Chemical Engineering, 10(1), 107070. https:// doi.org/10.1016/j.jece.2021.107070 Kaufman, L., & Rousseeuw, P.  J. (Eds.). (1990). Finding groups in data. Wiley. https://doi. org/10.1002/9780470316801 Lee, Y.-C., Chen, C., & Tsai, X.-T. (2016). Visualizing the knowledge domain of nanoparticle drug delivery technologies: A scientometric review. Applied Sciences, 6(1), 11. https://doi. org/10.3390/app6010011 Lipsey, R. G., Carlaw, K., & Bekar, C. (2005). Economic transformations: General purpose technologies and long-term economic growth. Oxford University Press. Liu, X., Jayaratne, R., Thai, P., Kuhn, T., Zing, I., Christensen, B., Lamont, R., Dunbabin, M., Zhu, S., Gao, J., Wainwright, D., Neale, D., Kan, R., Kirkwood, J., & Morawska, L. (2020). Low-­ cost sensors as an alternative for long-term air quality monitoring. Environmental Research, 185, 109438. https://doi.org/10.1016/j.envres.2020.109438 Menouar, H., Guvenc, I., Akkaya, K., Uluagac, A.  S., Kadri, A., & Tuncer, A. (2017). UAV-­ enabled intelligent transportation systems for the smart city: Applications and challenges. IEEE Communications Magazine, 55(3), 3. https://doi.org/10.1109/MCOM.2017.1600238CM Mongeon, P., & Paul-Hus, A. (2016). The journal coverage of web of science and scopus: A comparative analysis. Scientometrics, 106(1), 213–228. https://doi.org/10.1007/s11192-­015-­1765-­5 Mozaffari, M., Saad, W., Bennis, M., Nam, Y.-H., & Debbah, M. (2019a). A tutorial on UAVs for wireless networks: Applications, challenges, and open problems. IEEE Communications Surveys & Tutorials, 21(3), 2334–2360. https://doi.org/10.1109/COMST.2019.2902862 Mozaffari, M., Saad, W., Bennis, M., Nam, Y.-H., & Debbah, M. (2019b). A tutorial on UAVs for wireless networks: Applications, challenges, and open problems. IEEE Communications Surveys & Tutorials, 21(3), 3. https://doi.org/10.1109/COMST.2019.2902862 Nelson, J., & Grubesic, T. (2018). Environmental justice: A panoptic overview using scientometrics. Sustainability, 10(4), 1022. https://doi.org/10.3390/su10041022

16

1  An Overview of UAVs for Spatial Modeling and Urban Informatics

Primatesta, S., Guglieri, G., & Rizzo, A. (2019). A risk-aware path planning strategy for UAVs in urban environments. Journal of Intelligent & Robotic Systems, 95(2), 629–643. https://doi. org/10.1007/s10846-­018-­0924-­3 Qiang, Y. (2021). Geospatial analysis and model building. Geographic Information Science & Technology Body of Knowledge, 2021(Q1). https://doi.org/10.22224/gistbok/2021.1.12 Ramos, J. (2003). Using tf-idf to determine word relevance in document queries. In Proceedings of the first instructional conference on machine learning (pp. 133–142). Sangwan, A., & Singh, R.  P. (2015). Survey on coverage problems in wireless sensor networks. Wireless Personal Communications, 80(4), 1475–1500. https://doi.org/10.1007/ s11277-­014-­2094-­3 Shakhatreh, H., Sawalmeh, A.  H., Al-Fuqaha, A., Dou, Z., Almaita, E., Khalil, I., Othman, N. S., Khreishah, A., & Guizani, M. (2019). Unmanned aerial vehicles (UAVs): A survey on civil applications and key research challenges. IEEE Access, 7, 48572–48634. https://doi. org/10.1109/ACCESS.2019.2909530 Shi, W., Goodchild, M. F., Batty, M., Kwan, M.-P., & Zhang, A. (Eds.). (2021). Urban informatics. Springer. https://doi.org/10.1007/978-­981-­15-­8983-­6 Wei, F., Grubesic, T. H., & Bishop, B. W. (2015). Exploring the GIS knowledge domain using CiteSpace. The Professional Geographer, 67(3), 374–384. https://doi.org/10.1080/0033012 4.2014.983588 Yang, J., & Bou-Zeid, E. (2019). Designing sensor networks to resolve spatio-temporal urban temperature variations: Fixed, mobile or hybrid? Environmental Research Letters, 14(7), 074022. https://doi.org/10.1088/1748-­9326/ab25f8 Zeng, Y., Zhang, R., & Lim, T. J. (2016). Wireless communications with unmanned aerial vehicles: Opportunities and challenges. IEEE Communications Magazine, 54(5), 36–42. https:// doi.org/10.1109/MCOM.2016.7470933

Chapter 2

UAV Operating Environments

2.1 Introduction The UAV operating environment comprises UAV operators, UAV platforms (i.e., the drone), and the airspace in which UAVs operate. Growth or change due to technological advancements and regulations generally results in changes across all aspects of UAV operating environments. The expansion of low-cost UAV systems and advances in UAV technology that support their integration into enterprise-level business supply chains (e.g., Amazon delivery services) and military applications is driving much of the current growth in the operational environment. Recent market reports estimate that the global UAV industry will reach between $27 billion and $100 billion by 2030 (Astute Analytica, 2022). In the United States, both recreational and commercial UAVs continue to increase. For example, the Federal Aviation Administration’s official UAV registration statistics indicate more than 850,000 registrations in November 2023 – although this may be an underestimate (Kestelo, 2022; NCSL, 2023). In response, local, state, and federal policies have been adopted or adapted in recognition of both the technological advances and the number of registered UAVs. At the local level, cities are adopting ordinances that reflect FAA guidelines that provide city officials with the ability to enforce no-fly zones, safe operation, and registration requirements without the possibility of FAA preemption (Zickuhr et al., 2016). At the state level, nearly 44 states have enacted laws since 2013 that address issues ranging from strictly defining UAVs to how public officials and the general public can use UAVs (NCSL, 2023). Moreover, at the federal level, various changes involving where and when UAVs are permitted to operate in US airspace and new requirements for the operator and the technology have either already been adopted or are expected to be implemented very soon (FAA, 2023b). Although the evolution of policies surrounding the three facets of the UAV operational environment has slowed since the introduction of Part 107 rules in April of © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 T. H. Grubesic et al., UAVs for Spatial Modelling and Urban Informatics, https://doi.org/10.1007/978-3-031-54114-8_2

17

18

2  UAV Operating Environments

2016 (FAA, 2016), the FAA and other governing bodies recognize the need for policy updates that reflect ongoing innovation. That said, many of the same rules that govern the operation of UAVs under Part 107 still apply. For example, the maximum flight altitude and ground speed remain 400 feet and 100 mph, respectively. Operations in Class B, C, D, and E airspace are still not permitted without prior Air Traffic Control (ATC) authorization, and UAVs must remain within the visual line of sight (VLOS) of the operator and observer (although exceptions for operations beyond VLOS are becoming more common through FAA waivers). One of the larger areas of UAV research when it comes to operating in an urban environment is their social acceptance – whether that be related to UAVs delivering take-out, monitoring traffic patterns, or collecting images of commercial and residential properties (Wang et al., 2023). Regardless of the UAV system, these conversations generally revolve around a reasonable expectation of privacy, accountability for errant operators, and mechanisms to ensure that operators are abiding by UAV flight rules (Wang et al., 2023). We see solutions to some of these concerns reflected in new regulations for both commercial and recreational UAV operators. This chapter explores the current UAV operational environment, the likely changes, and how those changes may impact who, where, and how UAVs operate.

2.2 UAV Operators The decreased cost of owning a UAV drives the growing commercial and residential UAV market. Leading manufacturers like DJI and Parrot offer well-equipped entry-­ level UAVs for a few hundred dollars,1 with flight times upwards of 40 minutes and communication and control ranges of up to a mile or more. Due in part to the lower cost of entry, the number of registered UAVs and operators has steadily increased (FAA, 2023c, Fig. 2.1). Official FAA numbers put the total number of UAVs registered at the end of 2022 at around 650,000. As of the end of 2023, this number is more than 850,000 (FAA, 2023a). It is worth noting that there were reports of more than 1 million UAV registrations in the United States (USDOT, 2018), corresponding to numbers pulled from the official FAA website (FAA, 2023c). However, there may have been an error in record-keeping (Fig. 2.1). Interestingly, the number of UAV sightings by other aircraft has remained relatively constant since 2017 (FAA, 2023c). Generally speaking, if another aircraft can spot a UAV in the vicinity, the operator is likely violating at least one Part 107 rule – whether that be an altitude violation, unauthorized airspace travel, or not yielding to other aircraft. That said, it does appear that since 2021, when several regulatory changes went into effect, the number of sightings has dropped (Fig. 2.2). It is impossible to determine whether this decrease has anything to do with the policies or whether the decline will continue.

 The DJI Mavic Mini costs $450.00 as of November 2023.

1

2.2  UAV Operators

19

Fig. 2.1  The number of registered UAVs with the FAA over time. Data was taken directly from the FAA dataset and reported as is (FAA, 2023c)

The recreational market has and continues to make up the bulk of registered UAVs in the United States. Of the 850,000+ registrations, over half are registered for recreational purposes (n = 506,635), while the remainder are for commercial use under Part 107. For years after the introduction of Part 107 certifications, the FAA did not require any flight certifications for recreational operators. For many years, this was a major source of unease regarding the public acceptance of UAVs (Nelson & Gorichanaz, 2019). Perhaps as an acknowledgment of this regulatory gap, one of the most considerable changes occurring within the UAV operational environment was the requirement of a knowledge test for recreational operators. This change went into effect in late 2021. Simultaneously, several other changes to the certification process (recreational and commercial) also went into effect.

2.2.1 Recreational Operators With recreational operators making up the bulk of registered UAVs in the US, it was only a matter of time before enacting certification requirements for operation. In July 2021, the FAA adopted a new rule establishing a recreational UAV operations certification process. The Recreational UAS Safety Test (TRUST) is a requirement for all UAV recreational operators. Like the Part 107 certification for commercial operators, it covers aeronautical and safety knowledge but to a lesser degree. Testing

20

2  UAV Operating Environments

Fig. 2.2  The number of UAV sightings by other aircraft. These numbers in the figure are aggregated over states and generated based on quarterly reports by the FAA (2023c)

happens at an approved testing center (all online), and successful completion will result in a proof of passage certificate that must be carried when operating their UAV. The FAA requires operators to provide the certificate should any law enforcement official request it. On its face, the TRUST certificate may appear to be bureaucracy for bureaucracy’s sake. The test itself is free and relatively simple. There are only 23 questions, and one can take the test as many times as required to achieve a passing score. This exam is a very low bar, despite what one might read from the recreational UAV community on message boards and forums. When conducting previous research on the topic of UAV ordinance adoption at the city level, we noticed that one of the biggest complaints by cities was the lack of enforcement ability (Nelson & Gorichanaz, 2019; Nelson, 2019) when it came to requiring operators to have at least some form of certification. To that extent, the TRUST certificate can serve as a mechanism that allows local and state law enforcement to require certification. Specifically, for many years, local and state governments have not been allowed to require recreational operators to hold any certification that goes above and beyond what the FAA mandated (FAA regulations in court would ultimately supersede those laws), which was nothing. However, now that the FAA requires the TRUST certificate, local and state laws can adopt an ordinance that requires a UAV pilot to obtain and carry a TRUST certificate when flying. At a minimum, it ensures that all UAV operators have at least some baseline understanding of safe UAV operation. It also provides a pathway from which local law enforcement can enforce the TRUST

2.2  UAV Operators

21

ordinance without the possibility of FAA regulatory preemption. While it may seem to some like more overreach by the federal government, we believe the intentions remain good  – focusing on safety and education  – especially given the steadily increasing number of recreational operators in the United States.

2.2.2 Commercial (Part 107) Operators On the commercial UAV operator side, no major changes concerning the Part 107 certification process have occurred since the Part 107 rules came out in April 2016. There were, however, several more minor and essential changes made. Some of them had to do with the actual operation of the UAV (covered in the following sections), while others had to do with the certification process. To the delight of all Part 107 holders, one of the changes did away with the recurrent testing requirement. Before the change, there were requirements for a Part 107 holder to take an in-person knowledge test refresher every 24 months, which had to occur at one of the FAA’s 700 testing centers across the country. The new rule changed that requirement. Now, operators only need to take an online test (FAA, 2023d). The online test is free and not nearly as extensive as the written exam. All other requirements for Part 107 certification holders remain the same, aside from a new mandatory proof of certification task. Specifically, Part 107 certification holders must: • Have a remote pilot certificate with a small UAS rating, plus corresponding identification (such as a driver’s license) in their physical possession and readily accessible when exercising the privileges of that remote pilot certificate. • Be able to present the remote pilot certificate and identification upon a request from the FAA, NTSB, TSA, or any Federal, state, or local law enforcement officer. • Be prepared to make available, upon request, to the FAA any document, record, or report required under FAA regulations. • Be willing, upon request, to allow the FAA to test or inspect the drone, the remote pilot in command, the actual drone pilot, and, if applicable, the visual observer to determine compliance with the rule.

2.2.3 Government Operators Policies and regulations surrounding government use of UAVs (local, state, or federal bodies) have seen a fair amount of activity in recent years, although this has primarily occurred at the state level. Much of the regulatory activity concerns how and when law enforcement agencies may use UAVs for monitoring the public domain. Some state policies have expanded the use of UAVs by law enforcement, while others have moved to restrict it. For example, in 2020, Idaho and Minnesota

22

2  UAV Operating Environments

passed laws that outlined the circumstances for law enforcement to use UAVs. In the same year, Vermont established laws that prohibited law enforcement from using UAVs for facial recognition with a few exceptions (see Chap. 9 for a deeper discussion on future facial recognition using UAVs). The following year, Florida, Tennessee, and Texas expanded law enforcement use of UAVs under certain circumstances. At the same time, California, Maryland, and Washington placed certain restrictions on when law enforcement agencies could use UAVs for programs operated by the federal government (mainly associated with the use of “weaponized” UAVs or requiring local permission for federal entities to use UAVs within the state). The state-level policy arena is changing rapidly. The policy environment has likely changed since the writing of this book. There are a few websites that keep track of state-level changes. We recommend interested readers consult websites such as FindLaw (2021) or the National Conference of State Legislatures (NCSL, 2023) to find the most up-to-date information on state-level regulations. Many concerns surrounding the government’s use of UAVs for monitoring persons or property involve privacy, reasonable search and seizure, and what constitutes a reasonable expectation of privacy (Finn & Wright, 2016). The Fourth Amendment provides some protections, although using aerial imagery for criminal and civil prosecutions has not always been protected under the Fourth Amendment. Previous cases concerning the use of aerial photography by law enforcement have determined that manned operations (via aircraft or helicopter) did not violate an individual’s Fourth Amendment rights (McNeal, 2014). This decision exhibited complexities because the observations in the case were made by an aircraft operating in the public domain, and the observer (in the airplane) was using their naked eyes for observation. Thus, while UAV flights share many of the same qualities as an airplane, observations are not made by the naked eye – instead, such observations utilize a high-resolution digital camera. The use of this camera complicates Fourth Amendment protections. However, as UAVs become more ubiquitous, any argument that the use of a UAV by a government official is an advantage that is not commonly available to the public becomes less persuasive. To clarify, the authors of this book are not lawyers (thankfully) and are not qualified to fully comment on the legality of UAV use for law or code enforcement. However, ongoing litigation in the case of LONG LAKE TOWNSHIP v MAXON (2021) will be informative (see Chap. 3 for further details on the case) and may open up new opportunities for UAV monitoring of the public domain. The recognition of the benefits that UAVs provide for surveillance is likely the reason why we see most of the regulatory activity at the state level focused on law enforcement.

2.3 UAV Technology UAV technology continues to evolve in the sensors they carry, the platforms, and the algorithms used to analyze the data they collect. Where the latter is concerned, advances in Artificial Intelligence (AI) and Machine Learning (ML) are primary

2.3  UAV Technology

23

drivers of growth within the UAV marketplace (Fortune, 2020). For example, e-commerce and last-mile delivery services provided by UAVs will require extensive AI and ML programming to operate safely and effectively in urban environments (Carroll, 2019). These platforms will include automated obstacle avoidance and GPS-free navigation for return-to-home functionality, correct package placement, locating charging stations, and emergency procedures. Research in the area of AI for UAVs has seen a tremendous amount of growth in both academic and commercial settings (Xue & Gonsalves, 2021). With the current and expected growth in the UAV market within the United States and globally (Astute Analytica, 2022), policymakers and UAV manufacturers are taking steps to enhance the ability of UAVs to operate safely and effectively. This effort includes advances in UAV payloads (sensor technology), flight duration, and changes to UAV technical requirements that affect the overall UAV operational environment. See Chap. 9 of this book for more details.

2.3.1 Sensors UAV sensors used for imagery collection and object detection are one of the primary areas of technological advancement. This area is changing just as rapidly as others. For example, when we acquired our first UAV platform in mid-2015, LiDAR was only available for a select few models due in part to the cost and size of the available sensors. Now, several companies make LiDAR sensors for UAVs that are both small enough for something like the DJI Matrice to carry and affordable enough for a larger portion of the commercial UAV market to afford. When we wrote this book, Velodyne offered their smallest UAV LiDAR puck (Puck LITE) for $5500,2 the least expensive LiDAR sensor (that we are aware of) on the market. Moreover, while many UAV LiDAR solutions involved only copter-based platforms, this is changing. For example, the Fixar 007 fixed-wing platform we use now has multiple LiDAR options, including the YellowScan Mapper+OEM, the Velodyne Ultra, and the Avia sensor.3 In addition, LiDAR sensors already play an essential role in UAV navigation. The WingtraOne Gen II and Fixar 007 are equipped with small LiDAR rangefinders to aid in landing procedures. However, as fully autonomous AI-enabled UAVs enter the marketplace and LiDAR becomes smaller and more cost-effective, we will likely see it integrated as a means for near-real-time object recognition and navigation (Liao et al., 2016). However, until then, passive sensors remain the most widely utilized source of both imagery collection and object recognition for UAVs (Gyagenda et  al., 2022). These sensors include high-resolution cameras for

 https://velodynelidar.com/products/puck-lite/  https://fixar.pro/lidar-scanner-drone/

2 3

24

2  UAV Operating Environments

orthoimagery and photogrammetry (RGB), multispectral, and (more recently) hyperspectral sensors that all contribute to the UAV operational environment. RGB and multispectral sensors are the mainstays for UAV imagery and photogrammetry. Many RGB camera options (and their associated UAV platforms) vary in technical specifications. For example, one can equip the WingtraOne with the powerful Sony A7R Mark IV, RGB61 mapping camera. At 61 megapixels and a wide field of view (FOV), this sensor can capture imagery at a Ground Sampling Distance (GSD) of .7 cm/pixel. More impressively, a single image taken at 120 m altitude will cover 310 hectares, equating to fewer images per flight for more spatial coverage in less flight time (Wingtra, 2023). For operations focused more on monitoring and less on mapping, the P3 by PhaseOne is a copter-based platform that assists public utility and infrastructure operations. The extended focal length (150 mm) combined with a 100-megapixel lens provides operators with a large field of view and the ability to identify small details (0.11 cm/pixel) at a distance.4 Multispectral sensors are those that contain more than three spectral bands. The most common configuration is a five-band sensor using Red, Green, Blue, Near Infrared (NIR), and Red Edge. The inclusion of the NIR and Red Edge bands makes these sensors advantageous for agricultural applications that utilize band combinations, such as the Normalized Difference Vegetation Index (NDVI), to highlight vegetation abundance or health (Fernandez-Figueroa et al., 2022). The two leading distributors of UAV multispectral sensors are Micasense (RedEdge)5 and Parrot (Sequoia),6 both of which provide the same multispectral band configurations (RGB, NIR, Red Edge). One of the downsides of multispectral imagery, however, is the generally larger GSD (roughly 8 cm/pixel GSD at 120 m altitude). However, one exciting development has recently taken place to improve the GSD.  Specifically, Micasense has begun to include an additional Panchromatic band with its multispectral sensor that allows users to pan-sharpen the other five bands to achieve GSDs closer to 3 cm/pixel (Amro et al., 2011). Hyperspectral sensors are one of the next technical frontiers regarding UAV sensor innovation. Like multispectral sensors, hyperspectral sensors capture data across multiple spectral bands. However, unlike multispectral sensors, hyperspectral sensors are generally capturing 100 or more individual spectral bands. The review by Adão et al. (2017) offers an excellent overview of the technology behind hyperspectral remote sensing for UAV applications and lists many hyperspectral sensors available for UAV platforms. We encourage interested readers to explore that work. Like LiDAR sensors 5 years ago, we expect to see innovation in this area through the miniaturization of the sensors. Current hyperspectral solutions are relatively heavy, requiring the deployment of special gimbals and copter-based UAV platforms. The sensors are also extremely costly, running anywhere from $10,000 to over $50,000,

 https://geospatial.phaseone.com/drone-payload/p3-payload-for-drones/  https://support.micasense.com/hc/en-us 6  https://www.parrot.com/en 4 5

2.3  UAV Technology

25

depending on the configuration and design.7 As we saw with LiDAR 5 years ago, hyperspectral sensors may become smaller for easier integration into more UAV platforms. However, might take some time before they come down in price.

2.3.2 UAV Platform Changes A combination of advances in power supply and design innovation has dramatically enhanced the endurance of UAVs. A growing number of models can fly for an hour or more at extended ranges. For example, our two fixed-wing platforms (Fixar 007 and WingtraOne) have impressive telemetry ranges. The WingtraOne Gen II has a standard range of 6 miles, while the Fixar 007 has a reported range of 18 miles, given a direct line of sight (i.e., no obstacles). We have also noticed that companies now specialize in this type of Beyond Visual Line of Sight (BVLOS) platform. The Sentaero 5 from Censys Technologies8 is one platform explicitly designed for BVLOS with a telemetry range of up to 55 miles. The DeltaQuad Pro also has an impressive range of nearly 32 miles. An even more exciting innovation we see as a potential industry standard is integrating a 4G/LTE network or 5G systems for UAV communication and control (Chen et al., 2018; Li et al., 2019). The DeltaQuad Pro already offers this option. UAVs equipped with 4G/LTE would only be limited in range by their power supply and, in addition to operational control, could transfer data across massive ranges in near real-time. It is important to note that operating a UAV BVLOS is still not permitted under a Part 107 certificate. However, with all the advances in UAV endurance, more and more operators are seeking waivers for these operations (specifically for Part 107.31). The FAA keeps a published list of all waivers granted to UAV operators, including those for Part 107.31. This waiver is one of the more popular options, alongside night flying (no longer an issue – see the next section). At the time of this writing, 218 BVLOS waivers have been issued (FAA, 2023e) to a surprisingly wide range of operators across many industries. Notable applicants include utility companies (American Electric Power), UAV manufacturers (Censys Technologies), government agencies (Texas Department of Transportation), universities (Fullerton College), and e-commerce companies (Fed-Ex). Even Tesla has received a waiver for BVLOS UAV operations. It seems clear that BVLOS UAV operations are increasingly important for the operational environment, whether for package delivery across a community or monitoring tens of miles of energy infrastructure in rural areas. Regardless of the application, quickly identifying operators (in case of an emergency) is crucial. Indeed, one of the reasons Part 107 requires UAVs to remain within VLOS is to 1) ensure that the operator can respond to other aircraft flying in the vicinity and 2) to ensure

 The Cubert Ultris 20 hyperspectral sensor cost $53,500 in November 2023.  https://censystech.com/sentaero-5/

7 8

2  UAV Operating Environments

26

the UAV operator can be identified should anything go awry with the operation. One of the mechanisms the FAA uses to maintain the safe operation of UAVs within US airspace (either VLOS or BVLOS) is the newly proposed remote identification (RID) requirement.

2.3.3 Remote ID (RID) RID is one of the most significant changes to the technical requirements of UAVs operating in US airspace. Part 89 of Title 14 in the Code of Federal Regulations (CFR) (Part 89 for short) establishes the requirements of RID for unmanned aircraft systems (14 CFR Part 89, 2023). The RID rule thus applies to any UAV requiring registration in the United States. It is important to note that Part 89 establishes the RID as an operating requirement, but it is not an operating rule. In other words, flight is still possible if one does not have an RID-enabled UAV, but such flights will have limits. After the Part 89 compliance date, the UAV operation can continue to occur under three scenarios based on the RID capability (Table 2.1). In addition, Part 89 requirements are in addition to all Part 107 rules – having an RID does not give the UAV operator any additional leeway regarding the rules outlined in Part 107. The RID capability includes two operational bins: • Standard RID: broadcasts RID messages directly from the UAV radio frequency and includes the UAV serial number, the UAV position, and speed, as well as the control station’s (i.e., operator’s) position and emergency status. • RID broadcast module: an RID device is attached to the UAV (external device) or built into the aircraft (does not broadcast from radio frequency) and transmits the serial number, positional information of the UAV, and the position of the take-off and landing point for the flight. Regardless of which RID capability one is using, the registration for the UAV must be updated to include the serial number of the broadcast module. Table 2.1 clarifies that the FAA allows some flexibility in the new rule. First, UAVs produced after the compliance date should have RID capabilities built into Table 2.1  Part 89 Remote identification operational requirements based on the RID capabilities RID capability Standard RID RID broadcast module No RID

Description Any UAV produced after the compliance date (18 months from the rule effective date) Existing UAV without standard RID technology (produced before compliance date). RID is an add-on device incorporated by software upgrade UAVs without a standard RID or external module

Part 89 operational requirement Limited to VLOS Limited to VLOS

Operation in FAA-recognized identification areas (FRIAs) and limited the VLOS

2.3  UAV Technology

27

the system. The WingtraOne Gen II, for example, already has this RID capability and is Part 89 compliant right out of the box. Many of the new DJI models are also compliant with built-in transmitters. However, hundreds of thousands of UAVs in operation were produced well before the compliance date. The Fixar 007 model was designed and produced several years ago and does not have the standard RID capability. As a result, for us to be Part 89 compliant and fly “as usual,” we have installed a small RID broadcast device. We opted to use the DroneTag Mini,9 which is Part 89 compliant, by broadcasting the required information over a range of about 2 miles while the UAV is in operation. UAVs without a standard RID or a broadcast module can fly only in FAA-­ recognized identification areas (FRIAs). There are 206 FRIAs scattered throughout the US, and they are admittedly relatively sparse (Fig. 2.2). For example, there are only two FRIAs across all of Montana. Although it is an extra investment that operators must make, we encourage readers to consider it. Without RID, the overarching restrictions radically limit where one can operate (Fig. 2.3). We have one final tip for readers regarding RID equipment. Before operators purchase and install RID modules for their unequipped UAVs, they must check two critical pieces of information. First, in the case of the Fixar 007, it has been designed to distribute the weight of all the components evenly to establish the platform’s center of gravity. The autonomous control of the UAV and all the associated flight algorithms depend on the constant center of gravity. Adding an RID module, no matter how small, may alter the center of gravity, ultimately affecting the UAV flight and (worse) voiding any warranties should something happen. Of course, this will likely be more of an issue with fixed-wing platforms. However, we still highly recommend checking with the manufacturer about RID installation procedures before buying and installing an RID tag. Second, for UAVs to be Part 89 compliant, the RID tag (or the UAV itself) must meet the FAA requirements. It is up to the manufacturer to submit their technology to the FAA for review. The FAA will issue the technology a Declaration of Compliance (DOC) if it passes review. The best way to know if a UAV or an RID tag meets the Part 89 requirements is to check for a registered DOC. The FAA keeps a running list of the UAV models and external RID tags that have received a DOC.10 Interested readers will be able to find both the WingtraOne Gen II and the DroneTag Mini listed on the DOC page for RID compliance. Initially, we expected Part 89 to go into effect on September 16, 2023. However, the FAA extended a grace period of 6 months to March 16, 2024, to allow UAV operators more time to identify solutions to ensure their UAVs are compliant.

 https://dronetag.cz/products/mini/  https://uasdoc.faa.gov/listDocs/

9

10

28

2  UAV Operating Environments

Fig. 2.3  Locations where UAV operations are permitted without the use of a standard RID or broadcast device are indicated by blue pins. National UAV restricted areas are indicated in red

2.4 UAV Flight Operations There have been a few significant changes, or rather clarifications, to the original Part 107 rules regarding where and when UAVs can fly within US airspace. The FAA still restricts UAVs from flying in controlled airspace without prior authorization from the Air Traffic Control (ATC) towers. We refer readers to our previous book (Grubesic & Nelson, 2020) for a primer on airspace designations and aeronautical charts. However, we document several modest changes in Part 107 below. These changes include flights over people and operating UAVs at night.

2.4.1 Operations Over People The FAA recently made clarifications to Part 107.39 regarding operations over people (OOP). Specifically, Section 107.39 prohibits the operation of a UAV over another person unless that person is (a) directly involved in the operation or (b) within a safe cover, such as inside a stationary vehicle or under a protective structure. The addition to the rule is a new clarification (subpart “c” to Part 107.39) that

2.4  UAV Flight Operations

29

provides categorical exemptions that allow UAV operation over people without a waiver, provided the UAV falls into one of four operational categories. These categories represent different safety requirements that a UAV must meet before operations over people (Table  2.2). Categories 1–3 are the most relevant to Part 107 certificate holders, so we focus on those here. Category 4 involves obtaining an FAA airworthiness certificate, which only applies to UAVs over 55 lbs. (not covered under a Part 107 certificate). The kinetic energy limits in Table 2.2 are one of the most limiting factors for categorical exclusions and are about the amount of force the UAV exerts upon impact with a rigid object. In other words, how bad would it hurt if a UAV were to crash into a person? While the weight requirements are easy to accommodate, the operational requirement to meet the kinetic energy limit may not be possible (i.e., slower speed and lower altitude), nor would it be feasible for many operations. One of the only options we see for UAVs to meet the kinetic energy requirement would be a parachute deployed in emergencies or upon landing, and we are not aware of any UAV manufacturer currently exploring this option. In short, even with the new categorical waiver exemptions, most UAVs that weigh over 0.55lbs (i.e., not in Category 1) are not permitted to fly directly over people. Moreover, it is up to the UAV manufacturer to request the FAA category approval. Until there is enough demand from consumers for a model that meets these requirements, it is difficult for us to imagine manufacturers investing the time and resources into developing a compliant platform. That said, at the time of this writing, the eBee X series UAV manufactured by AgEagle (senseFly) has received a DOC (Category 3) from the FAA for OOP. The Ebee remains the only model that can be flown directly over people without obtaining a waiver. However, this comes with an enormous caveat. As a Category 3 eligible platform, additional flight requirements exist to fly over people. Specifically, the FAA restricts Category 3 operations to situations where: • The operation is conducted over a closed- or restricted-access site, and all persons within the site are made aware that the UAV will be operating, or. • The operation is not within a closed- or restricted-access site, and the UAV does not sustain flight over any person who is not directly involved in the operation of the UAV. Table 2.2  UAVs must meet safety requirements for operations over people to occur without an FAA waiver Weight limit Eligibility requirements FAA category approval Label requirement

Category 1 Category 2 Category 3