Search

Unmanned and aerial systems as interactors among different system components for communications, have opened up great opportunities for truth data discovery in Mobile Crowd Sensing (MCS) which has not been properly solved in the literature. In this paper, an Unmanned Aerial Vehicles-supported Intelligent Truth Discovery (UAV-ITD) scheme is proposed to obtain truth data at low-cost communications for MCS. The main innovations of the UAV-ITD scheme are as follows: (1) UAV-ITD scheme takes the first step in employing UAV joint Deep Matrix Factorization (DMF) to discover truth data based on the trust mechanism for an Information Elicitation Without Verification (IEWV) problem in MCS. (2) This paper introduces a truth data discovery scheme for the first time that only needs to collect a part of n data samples to infer the data of the entire network with high accuracy, which saves more communication costs than most previous data collection schemes, where they collect n or kn data samples. Finally, we conducted extensive experiments to evaluate the UAV-ITD scheme. The results show that compared with previous schemes, our scheme can reduce estimated truth error by 52.25%–96.09%, increase the accuracy of workers’ trust evaluation by 0.68–61.82 times, and save recruitment costs by 24.08%–54.15% in truth data discovery.

Abstract Manned and unmanned systems are prevalent in a wide range of aerial searching applications. For aircraft whose trajectory is not or cannot be planned on-the-fly, optimal deterministic search pattern generation is a critical area of research. Lissajous curves have recently caught attention as excellent candidates for all kinds of aerial search applications, but little fundamental research has been done to understand how best to design Lissajous pattern (LP)s for this use. This paper examines the optimization of these search patterns from analytical, numerical, and data-driven perspectives to establish the state of the field in Lissajous curves for aerial search. From an analytical perspective, it was found that the average expected distance between a Lissajous searcher and a random target on a unit square approaches 0.586 as search time increases. Furthermore, an analytical approximation for the average searcher speed was found to guarantee error of no more than 22.1%. Important outcomes from the numerical optimization of Lissajous search patterns include the development of an intuitive evaluation criterion and the conclusion that irrational frequency ratios near 0.8 typically yield highest performance. Finally, while a robust predictive model for fast pattern optimization is yet out of reach, initial results indicate that such an approach shows promise.

Summary: Drones, unmanned aerial systems, are of growing interest to organizations involved in disaster risk reduction, particularly in post-disaster emergency response. Despite the potential benefits of drones, their use is not well established and practical challenges need to be understood. This study examines the role of drones in disaster management by analyzing various applications of drones in response to the Noto Peninsula earthquake in January 2024. Drones were used on the ground in a variety of new ways, including transport of emergency supplies, restore of cellphone communications, and inspect on damaged facilities. Several issues were identified, including the need to incorporate drone capabilities into disaster management plans, develop appropriate laws and regulations, establish public-private coordination mechanisms, address technological limitations due to advances in technology, and implement training programs specifically for drone operators. Collaboration among government agencies, private organizations, and industry associations in disaster response highlighted the importance of fostering partnerships and mobilizing collective expertise in disaster management. The study concludes by highlighting the important role that drones can play in enhancing emergency response efforts and mitigating the impact of future disasters.

In this paper we advocate for the establishment of an aviation “innovative ecosystem” in Ukraine, particularly regarding the technology of unmanned aerial vehicles (UAVs). This ambitious strategy necessitates the introduction of new business models, taking advantage of new opportunities for achieving sustainable development goals in the increasingly resource-limited world. Moreover, this initiative faces a notable challenge: navigating the “double transition” of adopting both green and digital transformations across multiple domains. We believe the solution to creating an innovative ecosystem lies in harnessing the intellectual and infrastructural capabilities of top-tier research institutions – which in Ukraine include the National Aviation University (NAU) and the National Technical University “KPI.” By combining the efforts of Ukrainian higher education institutions and actively enhancing international cooperation, in particular with the Łukasiewicz Research Network – Institute of Aviation in Warsaw – we argue – Ukraine can expedite the development of “factories” of ideas and innovations by supporting the training, incubation and growth of startups, encourage the formation of micro and small high-tech enterprises and bringing their products to global and regional markets, improving the mechanisms of interaction of the startup teams with venture capital funds and investment companies.

Building upon the existing literature on terror innovation, this research article introduces a new analytical framework that highlights four key elements of emerging technologies which facilitate their adoption by terrorist groups. We posit that technologies that are inherently ‘disruptive,’ ‘dual-use,’ ‘democratized,’ and ‘diffused’ are particularly susceptible to misuse by terrorists, and that it is the interplay between these factors that helps to drive future directions of tech-enabled terror. We draw on terror and extremist interest in and use of two key technology areas – unmanned aerial systems and additive manufacturing, specifically 3‑D‑printed firearms – as examples to highlight how the factors in the framework interact and how they help to drive innovative terror behavior. We conclude with a series of concise recommendations to mitigate harmful terrorist use of these and other disruptive technologies and limit the impact of future tech-enabled terrorist innovations.

One of the most important and widespread corn/maize virus diseases is maize dwarf mosaic (MDM), which can be induced by sugarcane mosaic virus (SCMV). This study explores a machine learning analysis of five-band multispectral imagery collected via an unmanned aerial system (UAS) during the 2021 and 2022 seasons for SCMV disease detection in corn fields. The three primary objectives are to (i) determine the spectral bands and vegetation indices that are most important or correlated with SCMV infection in corn, (ii) compare spectral signatures of mock-inoculated and SCMV-inoculated plants, and (iii) compare the performance of four machine learning algorithms, including ridge regression, support vector machine (SVM), random forest, and XGBoost, in predicting SCMV during early and late stages in corn. On average, SCMV-inoculated plants had higher reflectance values for blue, green, red, and red-edge bands and lower reflectance for near-infrared as compared to mock-inoculated samples. Across both years, the XGBoost regression model performed best for predicting disease incidence percentage (R2 = 0.29, RMSE = 29.26), and SVM classification performed best for the binary prediction of SCMV-inoculated vs. mock-inoculated samples (72.9% accuracy). Generally, model performances appeared to increase as the season progressed into August and September. According to Shapley additive explanations (SHAP analysis) of the top performing models, the simplified canopy chlorophyll content index (SCCCI) and saturation index (SI) were the vegetation indices that consistently had the strongest impacts on model behavior for SCMV disease regression and classification prediction. The findings of this study demonstrate the potential for the development of UAS image-based tools for farmers, aiming to facilitate the precise identification and mapping of SCMV infection in corn.

Abstract Crane populations are declining worldwide, with anthropogenically exacerbated habitat loss emerging as the primary causal threat. The endangered Grey Crowned Crane (Balearica regulorum) is the least studied of the three crane species that reside in southern Africa. This data paucity hinders essential conservation planning and is primarily due to ineffective monitoring methods and this species' use of inaccessible habitats. In this study, we compared the behavioural responses of different Grey Crowned Crane social groupings to traditional on‐foot monitoring methods and the pioneering use of drones. Grey Crowned Cranes demonstrated a lower tolerance for on‐foot monitoring approaches, allowing closer monitoring proximity with drones (22.72 (95% confidence intervals ‐ 13.75, 37.52) m) than on‐foot methods (97.59 (86.13, 110.59) m) before displaying evasive behaviours. The behavioural response of flocks was minimal at flight heights above 50 m, whilst larger flocks were more likely to display evasive behaviours in response to monitoring by either method. Families displayed the least evasive behaviours to lower flights, whereas nesting birds were sensitive to the angles of drone approaches. Altogether, our findings confirm the usefulness of drones for monitoring wetland‐nesting species and provide valuable species‐specific guidelines for monitoring Grey Crowned Cranes. However, we caution future studies on wetland breeding birds to develop species‐specific protocols before implementing drone methodologies.

The rapid advancement of unmanned aerial systems in various civilian roles necessitates improved safety measures during their operation. A key aspect of enhancing safety is effective collision avoidance, which is based on conflict detection and is greatly aided by accurate trajectory prediction. This paper represents a novel data-driven trajectory prediction methodology based on applying the Long Short-Term Memory (LSTM) prediction algorithm to the UAS-S4 Ehécatl. An LSTM model was designed as the baseline and then developed into a Staked LSTM to better capture complex and hierarchical temporal trajectory patterns. Next, the Bidirectional LSTM was developed for a better understanding of the contextual trajectories from both its past and future data points, and to provide a more comprehensive temporal perspective that could enhance its accuracy. LSTM-based models were evaluated in terms of mean absolute percentage errors. The results reveal the superiority of the Bidirectional LSTM, as it could predict UAS-S4 trajectories more accurately than the Stacked LSTM. Moreover, the developed Bidirectional LSTM was compared with other state-of-the-art deep neural networks aimed at aircraft trajectory prediction. Promising results confirmed that Bidirectional LSTM exhibits the most stable MAPE across all prediction horizons.

The short-rotation coppice (SRC) culture of trees provides a sustainable form of renewable biomass energy, while simultaneously sequestering carbon and contributing to the regional carbon feedstock balance. To understand the role of SRC in carbon feedstock balances, field inventories with selective destructive tree sampling are commonly used to estimate aboveground biomass (AGB) and canopy structure dynamics. However, these methods are resource intensive and spatially limited. To address these constraints, we examined the utility of publicly available airborne Light Detection and Ranging (LiDAR) data and easily accessible imagery from Unmanned Aerial Systems (UASs) to estimate the AGB and canopy structure of an American sycamore SRC in the piedmont region of North Carolina, USA. We compared LiDAR-derived AGB estimates to field estimates from 2015, and UAS-derived AGB estimates to field estimates from 2022 across four planting densities (10,000, 5000, 2500, and 1250 trees per hectare (tph)). The results showed significant effects of planting density treatments on LIDAR- and UAS-derived canopy metrics and significant relationships between these canopy metrics and AGB. In the 10,000 tph, the field-estimated AGB in 2015 (7.00 ± 1.56 Mg ha−1) and LiDAR-derived AGB (7.19 ± 0.13 Mg ha−1) were comparable. On the other hand, the UAS-derived AGB was overestimated in the 10,000 tph planting density and underestimated in the 1250 tph compared to the 2022 field-estimated AGB. This study demonstrates that the remote sensing-derived estimates are within an acceptable level of error for biomass estimation when compared to precise field estimates, thereby showing the potential for increasing the use of accessible remote-sensing technology to estimate AGB of SRC plantations.

In recent times, the objective of reducing the environmental impact of the agricultural industry has led to the mechanization of the sector. One of the consequences of this is the everyday increasing use of Unmanned Aerial Systems (UAS) for different tasks in agriculture, such as spraying operations, mapping, or diagnostics, among others. Aerial spraying presents an inherent problem associated with the drift of small droplets caused by their entrainment in vortical structures such as tip vortices produced at the tip of rotors and wings. This problem is aggravated by other dynamic physical phenomena associated with the actual spray operation, such as liquid sloshing in the tank, GPS inaccuracies, wind gusts, and autopilot corrections, among others. This work focuses on analyzing the impact of nozzle position and liquid sloshing on droplet deposition through numerical modeling. To achieve this, the paper presents a novel six degrees of freedom numerical model of a DJI Matrice 600 equipped with a spray system. The spray is modeled using Lagrangian particles and the liquid sloshing is modeled with an interface-capturing method known as Volume of Fluid (VOF) approach. The model is tested in a spraying operation at a constant velocity of 2 m/s in a virtual vineyard. The maneuver is achieved using a PID controller that drives the angular rates of the rotors. This spraying mission simulator was used to obtain insights into optimal nozzle selection and positioning by quantifying the amount of droplet deposition.

More than a century has passed since the third dimension of space (airspace) has become an important place where combat and non-combat activities take place. The ability of the Air Force to carry out a wide range of content of combat operations and other activities from the airspace ensured the development of new qualitative characteristics of modern armed conflicts. The absence of physical limitations, high maneuverability and dynamism, short reaction time, as well as the high intensity of actions carried out by the Air Force, have enabled these forces to have strategic importance in armed conflicts since the Second World War. In parallel with the development of civilization and the emergence of qualitatively new technologies, rapidly improved technological solutions are being developed, which are changing the way of life of people around the world, improving it literally day by day. Each technological generation brings a new level of civilizational development, leading to the symbiosis of man as a human being and modern technical and technological means. Undoubtedly, this is also the case with the development of modern weapon systems and equipment used in armed conflicts. This is also the case with the revolutionary development of new weapon systems in air forces around the world. The development of modern weapons in the air force is best seen through the development of unmanned aerial platforms (unmanned aerial vehicles). The development of these assets is a consequence of the industrial revolution of the fourth generation and the importance of these platforms in the realization of tasks in modern armed conflicts is almost immeasurable. Namely, every conflict brings exponential development of these assets, both in technical and tactical sense. The development of these means in modern armed conflicts is a continuous process that is realized on the basis of previous experiences from the tactical use of unmanned aerial vehicles in concrete combat actions. There is no doubt that the technological progress of these means is something that in the future will enable the almost unlimited use of these aircraft in the execution of a wide variety of tasks without minimal danger to human life. This paper, whose main goal is to describe the unmanned aerial vehicle as a tool for the execution of a wide range of tasks in an unlimited war format, consists of three chapters in addition to the introduction and conclusion. In the first chapter, the theoretical determination of the unmanned aerial vehicle as a weapon-equipment system was made. In the second chapter, the use of unmanned aerial vehicles is analyzed on the examples of armed conflicts in Syria and Nagorno-Karabakh. In the third chapter, an analysis of potential aviation patterns of unmanned aerial vehicles and loitering munition (also known as a suicide drone or kamikaze drone) in combat operations was carried out.

Landslides represent great dangers that can cause fatalities and huge property damage. To prevent or reduce all possible consequences that landslides cause, it is necessary to know the kinematics of the surface and undersurface sliding masses. Geodetic surveying techniques can be used for landslide monitoring and creating a kinematic model of the landslide. One of the most used surveying techniques for landslide monitoring is the photogrammetric survey by Unmanned Aerial System. The results of the photogrammetric survey are dense point clouds, digital terrain models, and digital orthomosaic maps, where landslide displacements can be determined by comparing these results in two measurement epochs. This paper presents a new data processing method with a novel approach for calculating landslide displacements based on Unmanned Aerial System photogrammetric survey data. The main advantage of the new method is that it does not require the production of dense point clouds, digital terrain models, or digital orthomosaic maps to determine displacements. The applicability and accuracy of the new method were tested in a test field with simulated displacements of known values within the range of 20-40 cm in various directions. The new method successfully determined these displacements with a 3D accuracy of ±1.3 cm.

On-board state estimation is a persistent challenge to fielding unmanned aerial systems (UAS), particularly when global positioning system (GPS) measurements are not available. The dominant estimation tool remains the extended Kalman filter (EKF) applied to an inertial measurement unit (IMU). The growing availability of low-cost commercial-grade IMUs raises questions about how to best improve sensor readings into an estimate when measurements are available from multiple IMUs. This paper evaluates four different approaches to attitude estimation from multiple IMU measurements and their application in high dynamic motion. The four approaches are fusion of measurements (virtual IMU), fusion of state estimates (Federated KF), feedback fusion state estimate (Feedback Federated KF), and an EKF design incorporating the additional measurements (Augmented KF); these correspond to fusion before, within, or after state estimation. The performance of the approaches are quantified for varying IMU number theoretically and experimentally. The experiments use on-board autopilot hardware implementations of the estimators during maneuvers in a motion capture volume and flights on-board an Unmanned Aerial Systems (UAS) implementation, with the peak and root-mean-square (RMS) errors used to quantify accuracy. The RMS error results indicate that the feedback federated Kalman Filter using five IMUs returns 38% compared to general federated Kalman Filter using 37% accuracy improvement over a single IMU. This improvement compares to a 19% improvement for virtual IMU and 9% improvement for the Augmented KF respectively. These results indicate that both the Federated KF approach achieves the lowest RMS error relative to the virtual IMU and augmented KF approaches and inform the design of multi-IMU UAS pose estimators.

Abstract Long‐term monitoring programs to evaluate climate‐driven changes to tidewater glaciers, an important habitat for harbor seals (Phoca vitulina) in Alaska, are primarily carried out by costly, weather‐dependent aerial surveys from fixed‐winged aircraft. Unmanned aerial systems (UAS) can be an alternative cost‐effective solution for gathering image data to quantify, monitor, and manage these habitats. However, there is a paucity of information related to the accuracy of using imagery collected by UAS for purposes of measuring floating icebergs. We evaluated the accuracy of using a UAS with a built‐in 20‐megapixel (MP) camera as well as a consumer‐grade digital 16‐MP camera to capture images of floating and stationary icebergs for the purpose of collecting vertical height measurements. Images (n = 869) were captured of simulated icebergs (cuboidal foam boxes, Cb) (n = 5) and real icebergs (n = 5) that were either grounded or floating. The mean error ratios (Ers) obtained were less than 10% and derived by comparing the mean calculated measurements of heights of Cb obtained from images captured by UAS with the physically measured heights of these Cb. The mean Er for height measurements of grounded icebergs (n = 4) and one floating iceberg was also less than 10%. Within an object–image distance range of 6–25 m, the cameras captured images that were suitable to accurately calculate the heights of floating and grounded objects, and drift or uncontrolled movement of the UAS caused by wind or temporary loss of GPS did not have any effect on measurement error. Our study provides substantial evidence of the accuracy associated with using images captured by UAS for measuring dimensions of structures positioned on either water or land surfaces. Ultimately, accurate surveys of glacial ice used by harbor seals will improve our understanding of the role of decreasing habitat in explaining population variability between different tidewater glaciers.

This paper presents the realization of a robotic workcell capable of autonomous, flexible, fabrication of complete, full-scale, functional mechatronic systems. Using a 6-DOF robotic arm with a hybrid Additive Manufacturing (AM)/Pick and Place (PnP) end effector, the workcell can rapidly print enclosures, insert and wire functional electronics, and embed them into the printed structure. The workcell's scalable control system, digital workflow for creating assembly instructions, hybrid AM/PnP toolhead design, and autonomous part tending mechanism are detailed. In addition, a modular electronics infrastructure is presented to enable facile, error-free robotic wiring of large-gauge copper wire. The workcell's ability to manufacture is verified in a case study wherein it is used to autonomously fabricate a variety of Unmanned Aerial Systems (UASs). In the study, the workcell produces a functional quadrotor drone, releases the drone from the workcell, and allows the drone to record the workcell constructing a second, different drone - all without human involvement. The resulting workcell, which combines digitally-integrated manufacturing, robotics and automation, and hybrid process manufacturing, enables autonomous manufacture of complete, functional mechatronic systems that are tailored to custom design needs in a just-in-time manner.

Abstract Coastal wetland biomass is an important indicator of wetland productivity, carbon storage, health, and vulnerability to climate change. The ability to estimate aboveground biomass (AGB) in wetlands at ecologically relevant scales is complicated by the spatial variability inherent to patterns in wetland vegetation and the biogeomorphic processes that help create them. Remote sensing provides an approach for mapping wetland biomass, but the spatial resolutions of satellite and airborne imagery often constrain the types of ecological patterns and processes that can be detected. Unmanned Aerial Vehicles (UAVs) have previously been used to capture fine‐scale (≤1 m) variability in AGB in coastal wetland settings. However, it remains unclear if a UAV approach to estimating wetland biomass is transferrable across diverse wetland sites or how it compares to commonly used satellite‐based approaches. Here, we test the capabilities of UAVs in remotely quantifying AGB and compare biomass estimation using UAV and Landsat satellite imagery (30 m resolution) in several wetland sites in Southern California. Field surveys highlight significant spatial variability in wetland plant community AGB and height that influence remote biomass estimation. Relationships between UAV vegetation indices and AGB were site‐specific and influenced by vegetation types. Biomass estimation using UAVs (r2 = 0.40, RMSE = 534.6 g m−2) showed better correlation with NDVI than a Landsat‐based approach (r2 = 0.26, RMSE = 596.8 g m−2). We found combining high‐resolution UAV AGB maps and Landsat NDVI to develop AGB models showed the highest correlation (r2 = 0.45, RMSE = 659.7 g m−2) and provided additional spatial information to aid scaling field data to satellite imagery. Overall, UAVs captured more spatial complexity in aboveground biomass at finer scales than is possible with moderate‐resolution Landsat pixels, indicating that UAVs can be used to characterize patterns of within‐marsh variability resulting from local‐scale (≤ 100s of meters) ecological processes.

Gravaglione represents one of the main swallow holes of the Canale di Pirro, low Murge, Apulia region, Italy. Here, after an intense rainstorm, a huge volume of rainwater accumulates at the surface. The drainage dynamics suggest that the Gravaglione could be part of a large, and potentially unknown, karst system. To verify this hypothesis and to acquire useful information on the possible karst environment features, an integrated aerial and geophysical multiscale and multimethod approach was applied. In particular, aerial photogrammetry, ground penetrating radar measurements and electrical resistivity tomography surveys were hence conducted and integrated to potentially detect the caves, define the subsurface volume possibly affected by karst systems and to verify the existence of links between the surficial morphology and the subsoil structure. The results provided interesting insights regarding the presence of a complex karst system extending up to 200 m b.g.l. and with a marked 3D nature. Overall, the Gravaglione case study demonstrates the geophysical approach validity and poses the basis for the development of an expeditive and low-cost high-resolution strategy for detecting and characterizing karst caves.

The UK aviation industry is committed to achieving net zero emissions by 2050 through sustainable measures and one of the key aspects of this effort is the implementation of Unmanned Traffic Management (UTM) systems. These UTM systems play a crucial role in enabling the safe and efficient integration of unmanned aerial vehicles (UAVs) into the airspace. As part of the Airspace of the Future (AoF) project, the development and implementation of UTM services have been prioritised. This paper aims to create an environment where routine drone services can operate safely and effectively. To facilitate this, a digital twin of the National Beyond Visual Line of Sight Experimentation Corridor has been created. This digital twin serves as a virtual replica of the corridor and allows for the synthetic testing of unmanned traffic management concepts. The implementation of the digital twin involves both simulated and hybrid flights with real drones. Simulated flights allow for the testing and refinement of UTM services in a controlled environment. Hybrid flights, on the other hand, involve the integration of real drones into the airspace to assess their performance and compatibility with the UTM systems. By leveraging the capabilities of UTM systems and utilising the digital twin for testing, the AoF project aims to advance the development of safer and more efficient drone operations. The Experimentation Corridor has been developed to simulate and test concepts related to managing unmanned traffic. The paper provides a detailed account of the implementation of the digital twin for the AoF project, including simulated and hybrid flights involving real drones.

In-depth knowledge about soil moisture dynamics is crucial for irrigation management in precision agriculture. This study evaluates the feasibility of high spatial resolution near-infrared remote sensing with unmanned aerial systems for soil moisture estimation to provide decision support for precision irrigation management. A new trapezoid model based on near-infrared transformed reflectance (NTR) and the normalized difference vegetation index (NDVI) is introduced and used for estimation and mapping of root zone soil moisture and plant extractable water. The performance of the proposed approach was evaluated via comparison with ground soil moisture measurements with advanced time domain reflectometry sensors. We found the estimates based on the NTR−NDVI trapezoid model to be highly correlated with the ground soil moisture measurements. We believe that the presented approach shows great potential for farm-scale precision irrigation management but acknowledge that more research for different cropping systems, soil textures, and climatic conditions is needed to make the presented approach viable for the application by crop producers.

With the increasing occurrence of cyanobacteria blooms, it is crucial to improve our ability to monitor impacted lakes accurately, efficiently, and safely. Cyanobacteria are naturally occurring in many waters globally. Some species can release neurotoxins which cause skin irritations, gastrointestinal illness, pet/livestock fatalities, and possibly additional complications after long-term exposure. Using a DJI M300 RTK Unmanned Aerial Vehicle equipped with a MicaSense 10-band dual camera system, six New Hampshire lakes were monitored from May to September 2022. Using the image spectral data coupled with in situ water quality data, a random forest classification algorithm was used to predict water quality categories. The analysis yielded very high overall classification accuracies for cyanobacteria cell (93%), chlorophyll-a (87%), and phycocyanin concentrations (92%). The 475 nm wavelength, normalized green-blue difference index—version 4 (NGBDI_4), and normalized green-red difference index—version 4 (NGRDI_4) indices were the most important features for these classifications. Logarithmic regressions illuminated relationships between single bands/indices with water quality data but did not perform as well as the classification algorithm approach. Ultimately, the UAS multispectral data collected in this study successfully classified cyanobacteria cell, chlorophyll-a, and phycocyanin concentrations in the studied NH lakes.

The monitoring of soil moisture content (SMC) at very high spatial resolution (

Ancillary structures are essential for highways’ safe operationality but are mainly prone to environmental corrosion. The traditional way of inspecting ancillary structures is manned inspection, which is laborious, time-consuming, and unsafe for inspectors. In this paper, a novel image processing technique was developed for autonomous corrosion detection of in-service ancillary structures. The authors successfully leveraged corrosion features in the YCbCr color space as an alternative to the conventional red–green–blue (RGB) color space. The proposed method included a preprocessing operation including contrast adjustment, histogram equalization, adaptive histogram equalization, and optimum value determination of brightness. The effect of preprocessing was evaluated against a semantically segmented ground truth as a set of pixel-level annotated images. The false detection rate was higher in Otsu than in the global threshold method; therefore, the preprocessed images were converted to binary using the global threshold value. Finally, an average accuracy and true positive rate of 90% and 70%, respectively, were achieved for corrosion prediction in the YCbCr color space.

One of the most commonly used surveying techniques for landslide monitoring is a photogrammetric survey using an Unmanned Aerial System (UAS), where landslide displacements can be determined by comparing dense point clouds, digital terrain models, and digital orthomosaic maps resulting from different measurement epochs. A new data processing method for calculating landslide displacements based on UAS photogrammetric survey data is presented in this paper, whose main advantage is the fact that it does not require the production of the above-mentioned products, enabling faster and simpler displacement determination. The proposed method is based on matching features between the images from two different UAS photogrammetric surveys and calculating the displacements based only on the comparison of two reconstructed sparse point clouds. The accuracy of the method was analyzed on a test field with simulated displacements and on an active landslide in Croatia. Moreover, the results were compared with the results obtained with a commonly used method based on comparing manually tracked features on orthomosaics from different epochs. Analysis of the test field results using the presented method show the ability to determine displacements with a centimeter level accuracy in ideal conditions even with a flight height of 120 m, and on the Kostanjek landslide with a sub-decimeter level accuracy.

This article focuses on the modeling and controlling of Unmanned Aerial Manipulators (UAMs) in a leader/follower configuration performing a cooperative manipulation task. Each UAM consists of an Unmanned Aerial Vehicle (UAV) with an attached serial-link robotic manipulator. The Recursive Newton-Euler dynamics formulation is employed to account for the interaction between the UAV and its manipulator. The overall system consists of a couple UAMs with a carrying load. The coupling between these systems is due to the exerted forces by their manipulators through the object characterized by its stiffness matrix. A leader/follower control scheme is employed with a stability-analysis tailored to the UAM-pair. The leader UAM defines the trajectory of the moving object while the follower acts so as to reduce the system's internal reaction forces. Simulation studies are employed to validate the controller's performance while comparing the system's response against that derived from a classical nonlinear tracking controller.

This article presents the development of Autopilot (AP) blocks for a fixed-wing aircraft and the implementation of a Processor In the Loop (PIL) test platform. The AP entails the control, guidance, and navigation blocks. The control block design employed classical control theory considering a Piper 4 aircraft model, and the MILF-8785C design requirements standard. The PIL test platform is composed of an AP running in a microcontroller and a simulated aircraft on a PC (Personal Computer). The results of three missions carried out on the PIL platform are presented. In these missions, the performed maneuvers are such that lateral and longitudinal controls are required. In all three cases, all predefined waypoints are achieved, indicating the success of the developed AP project. The AP test implemented in a microcontroller that can be embedded onto an aircraft for a real flight mission indicates the usefulness of the developed PIL platform.

Vegetation on coastal dunes is a key element, as it promotes the growth and stabilization of these landforms while contributing significantly to biodiversity. Physical (e.g. impact of storms), ecological (e.g. animal grazing) and human-related (e.g. farming and recreation) factors may disturb coastal dune vegetation, changing dune dynamics and eventually inducing ecogeomorphic state shifts. Therefore, understanding vegetation dynamics and state turns crucial to predict dune evolution paths. The latter must be supported by observations combined with the development of tools (e.g. indexes) able to detect eventual changes and to automatically categorize the state of the vegetation. Here, a multi-step index to characterise the dune vegetation state (DUVES) was developed and tested in Barreta Island (South Portugal), where grey dune vegetation has declined in recent years. The index was computed using classified true colour orthophotos and orthomosaics derived from UAS (Unmanned Aerial Systems) surveys. Google Earth images were used as complementary data to analyse the evolution trends. The possible sources of disturbance (i.e. human-related activities and gull occupation) were also investigated by comparing their distribution with the vegetation changes. DUVES successfully identified different states of vegetation cover that expressed its stability, perturbation or growth based on temporal changes and allowed the analysis of their evolutionary trends. The distribution of perturbation was mostly associated with gull nesting areas, increasing over time, and to a less extent to human-related activities. The observed grey dune habitat loss was due to replacement of plants typical from this habitat by ruderal species promoted by the positive feedback established between gulls and vegetation. The developed index proved to be of great utility to define dune habitat evolution and understand the associated drivers, being a tool with a wide range of applications, namely for improving future coastal management actions aimed at conserving dune habitats. Moreover, DUVES is potentially transferable due to its easy adaptability depending on the particularities of each study site or goal.

Soil moisture is a key parameter in hydrological research and drought management. The inversion of soil moisture based on land surface temperature (LST) and NDVI triangular feature spaces has been widely used in various studies. Remote sensing provides regional LST data with coarse spatial resolutions which are insufficient for field scale (tens of meters). In this study, we bridged the data gap by adopting a Data Mining Sharpener algorithm to downscale MODIS thermal data with Vis-NIR imagery from Sentinel-2. To evaluate the downscaling algorithm, an unmanned aerial system (UAS) equipped with a thermal sensor was used to capture the ultra-fine resolution LST at three sites in the Tang River Basin in China. The obtained fine-resolution LST data were then used to calculate the Temperature Vegetation Dryness Index (TVDI) for soil moisture monitoring. Results indicated that downscaled LST data from satellites showed spatial patterns similar to UAS-measured LST, although discrepancies still existed. Based on the fine-resolution LST data, a 10-m resolution TVDI map was generated. Significant negative correlations were observed between the TVDI and in-situ soil moisture measurements (Pearson’s r of −0.67 and −0.71). Overall, the fine-resolution TVDI derived from the downscaled LST has a high potential for capturing spatial soil moisture variation.

Predictive modeling (PM) is a useful tool in selecting from a large number of target variables. Given the large quantities of data types generated from unmanned aerial system (UAS) platforms, uncovering the most appropriate candidate UAS data–based phenotype (UASDP) with a strong relationship to target crop traits may be a challenge with the traditional regression models. We hypothesized that by employing machine learning modeling techniques, the contributions of multiple vegetative indices as predictors of biomass and water-use could be defined and ranked. The objective of this study was to apply predictive machine techniques in determining the phase of spinach growth and UASDP variables that best predict yield and water-use efficiency. UASDPs were derived using red-green-blue and multi-spectral sensors mounted on a UAS platform flown weekly on 10 spinach genotypes under well-watered and partial water deficit conditions. Candidate UASDPs including mean and maximum plant height, canopy cover and volume, excess greenness index (ExG), chlorophyll red-edge (ChlRE), normalized difference vegetation index (NDVI) and normalized difference red-edge (NDRE) were generated. UASDPs were used to predict above-ground biomass-based fresh yield, biological dry yield, and field water-use efficiency (WUEf). This study highlights the use of bootstrap forest partitioning, and partition rank fraction as methods for selecting among many UASDPs as predictors of WUEf and yield. We also used a weighted geometric mean to integrate various model performance metrics to further refine the UASDPs rankings. UASDPs from mid to late growth stages were better predictors of yield and WUEf than UASDPs from earlier periods. Canopy volume, NDRE and ExG were the best predictors overall and were useful for distinguishing varieties, particularly in water-stress conditions. We expect that the approaches detailed here will improve reliability of discriminating the importance of UAS-acquired data types thereby improving UAS data-processing efficiency.

Small Unmanned Aerial Systems (UAS) usage is undoubtedly increasing at a significant rate. However, alongside this expansion is a growing concern that dependable low-cost counter measures do not exist. To mitigate a threat in a restricted airspace, it must first be known that a threat is present. With airport disruption from malicious UASs occurring regularly, low-cost methods for early warning are essential. This paper considers a low-cost early warning system for UAS detection and classification consisting of a BladeRF software-defined radio (SDR), wideband antenna and a Raspberry Pi 4 producing an edge node with a cost of under USD 540. The experiments showed that the Raspberry Pi using TensorFlow is capable of running a CNN feature extractor and machine learning classifier as part of an early warning system for UASs. Inference times ranged from 15 to 28 s for two-class UAS detection and 18 to 28 s for UAS type classification, suggesting that for systems that require timely results the Raspberry Pi would be better suited to act as a repeater of the raw SDR data, enabling the processing to be carried out on a higher powered central control unit. However, an early warning system would likely fuse multiple sensors. These experiments showed the RF machine learning classifier capable of running on a low-cost Raspberry Pi 4, which produced overall accuracy for a two-class detection system at 100% and 90.9% for UAS type classification on the UASs tested. The contribution of this research is a starting point for the consideration of low-cost early warning systems for UAS classification using machine learning, an SDR and Raspberry Pi.

Recent developments in agriculture mechanization have generated significant challenges towards sustainable approaches to reduce the environmental footprint and improve food quality. This paper highlights the benefits of using unmanned aerial systems (UASs) for precision spraying applications of pesticides, reducing the environmental risk and waste caused by spray drift. Several unmanned aerial spraying system (UASS) operation parameters and spray system designs are examined to define adequate configurations for specific treatments. A hexarotor DJI Matrice 600 equipped with T-Motor “15 × 5” carbon fiber blades is tested numerically using computational fluid dynamics (CFD) and experimentally in a wind tunnel. These tests assess the aerodynamic interaction between the wake of an advancing multicopter and the fine droplets generated by atomizers traditionally used in agricultural applications. The aim of this research is twofold. First, we analyze the effects of parameters such as flight speed (0, 2, and 3 m·s−1), nozzle type (hollowcone and fan), and injection pressure (2–3 bar) on spray distribution. In the second phase, we use data from the experimental campaign to validate numerical tools for the simulation of rotor–droplet interactions necessary to predict spray’s ground footprint and to plan a precise guidance algorithm to achieve on-target deposition and reduce the well-known droplet drift problem.

Many public safety agencies in the US have initiated a UAS-based procedure to document and map crash scenes. In addition to significantly reducing the time taken to document evidence as well as ensuring first responder safety, UAS-based mapping reduces incident clearance time and thus the likelihood of a secondary crash occurrence. There is a wide range of cameras used on these missions, but they are predominantly captured by mid-priced drones that cost in the range of $2000 to $4000. Indiana has developed a centralized processing center at Purdue University that has processed 252 crash scenes, mapped using 29 unique cameras, from 35 public agencies over the past three years. This paper includes a detailed case study that compares measurements obtained from a traditional ground-based real-time kinematic positioning base station and UAS-based photogrammetric mapping. The case study showed that UAS derived scale errors were within 0.1 ft (3 cm) of field measurements, a generally accepted threshold for public safety use cases. Further assessment was done on the 252 scenes using ground control scale error as the evaluation metric. To date, over 85% of the measurement errors were found to be within 0.1 ft (3 cm). When substantial errors are identified by the Purdue processing center, they are flagged for further dialog with the agency. In most of the cases with larger errors, the ground control distance was incorrectly measured, which is easily correctable by returning to the scene and performing new distance control measurements.

Recent advances in computer vision and camera-equipped unmanned aerial systems (UAS) for 3D modeling enable UAS-based photogrammetry surveys with high spatial-temporal resolutions. To generate consistent and high-quality 3D models using UASs, understanding how influence factors (i.e., flight height, image overlap, etc.) affect the 3D modeling accuracy and their levels of significance are important. However, there is little to no quantitative analysis that studies how these influence factors interact with and affect the accuracy when changing the values of the influence factors. Moreover, there is little to no research that assesses more than three influence factors. Therefore, to fill this gap, this paper aims to evaluate and predict the accuracy generated by different flight combinations. This paper presents a study that (1) assessed the significance levels of five influence factors (flight height, average image quality, image overlap, ground control point (GCP) quantity, and camera focal lengths), (2) investigated how they interact and impact 3D modeling accuracy using the multiple regression (MR) method, and (3) used the developed MR models for predicting horizontal and vertical accuracies. To build the MR model, 160 datasets were created from 40 flight missions collected at a site with a facility and open terrain. For validating the prediction model, five testing datasets were collected and used at a larger site with a complex building and open terrain. The results show that the findings of this study can be applied to surveyors’ better design flight configurations that result in the highest accuracies, given different site conditions and constraints. The results also provide a reasonable prediction of accuracy given different flight configurations.

In the present work, three LiDAR technologies (Faro Focus 3D X130—Terrestrial Laser Scanner, TLS-, Kaarta Stencil 2–16—Mobile mapping system, MMS-, and DJI Zenmuse L1—Airborne LiDAR sensor, ALS-) have been tested and compared in order to assess the performances in surveying built heritage in vegetated areas. Each of the mentioned devices has their limits of usability, and different methods to capture and generate 3D point clouds need to be applied. In addition, it has been necessary to apply a methodology to be able to position all the point clouds in the same reference system. While the TLS scans and the MMS data have been geo-referenced using a set of vertical markers and sphere measured by a GNSS receiver in RTK mode, the ALS model has been geo-referenced by the GNSS receiver integrated in the unmanned aerial system (UAS), which presents different characteristics and accuracies. The resulting point clouds have been analyzed and compared, focusing attention on the number of points acquired by the different systems, the density, and the nearest neighbor distance.

Abstract A typical problem for magnetic surveys with small Unmanned Aerial Systems (sUAS) is the heading error caused by undesired magnetic signals that originate from the aircraft. This can be addressed by suspending the magnetometers on sufficiently long tethers. However, tethered payloads require skilled pilots and are difficult to fly safely. Alternatively, the magnetometer can be fixed on the aircraft. In this case, aircraft magnetic signals are removed from the recordings with a process referred to as magnetic compensation, which requires parameters estimated from calibration flights flown in an area with magnetically low‐gradients prior to the survey. We present open‐source software fully written in Python to process data and compute compensations for two fundamentally different magnetometer systems (scalar and vector). We used the software to compare the precision of two commercially available systems by flying dense grid patterns over a 135 × 150 m area using different suspension configurations. The accuracy of the magnetic recordings is assessed using both standard deviations of the calibration pattern and tie‐line cross‐over differences from the survey. After compensation, the vector magnetometer provides the lowest heading error. However, the magnetic field intensity recovered with this system is relative and needs to be adjusted with absolute data if absolute intensity values are needed. Overall, the highest accuracy of all suspension configurations tested was obtained by fixing the magnetometer 0.5 m below the sUAS onto a self‐built carbon‐fiber frame, which also offered greater stability and allowed fully autonomous flights in a wide range of conditions.

Abstract In adaptive management of rangelands, monitoring is the vital link that connects management actions with on‐the‐ground changes. Traditional field monitoring methods can provide detailed information for assessing the health of rangelands, but cost often limits monitoring locations to a few key areas or random plots. Remotely sensed imagery, and drone‐based imagery in particular, can observe larger areas than field methods while retaining high enough spatial resolution to estimate many rangeland indicators of interest. However, the geographic extent of drone imagery products is often limited to a few hectares (for resolution ≤1 cm) due to image collection and processing constraints. Overcoming these limitations would allow for more extensive observations and more frequent monitoring. We developed a workflow to increase the extent and speed of acquiring, processing, and analyzing drone imagery for repeated monitoring of two common indicators of interest to rangeland managers: vegetation cover and vegetation heights. By incorporating a suite of existing technologies in drones (real‐time kinematic GPS), data processing (automation with Python scripts, high performance computing), and cloud‐based analysis (Google Earth Engine), we greatly increased the efficiency of collecting, analyzing, and interpreting high volumes of drone imagery for rangeland monitoring. End‐to‐end, our workflow took 30 d, while a workflow without these innovations was estimated to require 141 d to complete. The technology around drones and image analysis is rapidly advancing which is making high volume workflows easier to implement. Larger quantities of monitoring data will significantly improve our understanding of the impact management actions have on land processes and ecosystem traits.

Biomass is an important indicator for evaluating crops. The rapid, accurate and nondestructive monitoring of biomass is the key to smart agriculture and precision agriculture. Traditional detection methods are based on destructive measurements. Although satellite remote sensing, manned airborne equipment, and vehicle-mounted equipment can nondestructively collect measurements, they are limited by low accuracy, poor flexibility, and high cost. As nondestructive remote sensing equipment with high precision, high flexibility, and low-cost, unmanned aerial systems (UAS) have been widely used to monitor crop biomass. In this review, UAS platforms and sensors, biomass indices, and data analysis methods are presented. The improvements of UAS in monitoring crop biomass in recent years are introduced, and multisensor fusion, multi-index fusion, the consideration of features not directly related to monitoring biomass, the adoption of advanced algorithms and the use of low-cost sensors are reviewed to highlight the potential for monitoring crop biomass with UAS. Considering the progress made to solve this type of problem, we also suggest some directions for future research. Furthermore, it is expected that the challenge of UAS promotion will be overcome in the future, which is conducive to the realization of smart agriculture and precision agriculture.

Unoccupied aerial systems (UAS) have become common tools for ecological monitoring and management. However, UAS use has the potential to negatively affect wildlife. Both policy makers and practitioners require data about the potential impacts of UAS on natural biota, but few studies exist and some of the published results conflict. We conducted two experiments to assess the responses of chinstrap penguins (Pygoscelis antarcticus), Antarctic fur seals (Arctocephalus gazella), and leopard seals (Hydrurga leptonyx) to UAS overflights. First, to provide a baseline for assessing disturbance from UAS operations, we compare behavioral responses from UAS flights to those from traditional, ground surveys. Second, to inform users and policy makers about preferred flight methods, we assess behavioral and physiological responses to UAS flown at specific altitudes, during different stages of breeding chronology, and with other site factors. Between January 2017 and March 2018 we conducted 268 UAS flight approaches and 36 ground-based surveys at Cape Shirreff, Antarctic Peninsula. We applied generalized linear mixed effects models and Kruskal-Wallis tests to 10,164 behavioral scores obtained from three independent observers. When directly compared, behavioral responses by all species to UAS overflights at 30 m were not different from control periods, while responses to ground surveys were significantly more intense. Behavioral responses generally increased as UAS flew lower, and for penguins those increases intensified as the breeding season progressed (i.e., guard and molt stages). We argue that results from UAS wildlife response studies need to be assessed relative to the impacts of alternative methods, and within the ecological context of the target species. Finally, we suggest data-driven best practices for both UAS use and for the design of future UAS-wildlife response studies.

The scale and scope of global logistics systems make real-time visibility of individual assets in global logistics systems difficult. Aggregating global logistics data to a manageable level where interruptions and disruptions can be anticipated or resolved is high on the wish list of global logistics managers and decision makers. Asset tracking and condition visibility in global supply chains is also difficult because few standards or metrics have been assessed in a supply chain, particularly when new technology is introduced, such as unmanned aerial systems in global supply chains. In this paper, we describe the integration of an unmanned aerial system in a global logistics system, and the metrics used to assess the integrated system. We highlight the importance of supply chain process, business impact, societal and environmental sustainability metrics, in addition to economic and supply chain performance metrics, in evaluating the integrated system.

The use of geospatial sciences and technologies for the management of grazinglands has fostered a plethora of applications related to ecology, wildlife, vegetation science, forage productivity and quality, and animal husbandry. Some of the earliest use of remote sensing dates to the proliferation of aerial photography in the 1930s. Today, remote sensing using satellite imagery, global navigation satellite systems (GNSS), and internet-connected devices and sensors allow for real- and near real-time modeling and observation of grazingland resources. In this special issue of Remote Sensing, we introduce nine original publications focusing on varying aspects of grazingland management, such as animal health and telemetry, climate change, soil moisture, herbaceous biomass, and vegetation phenology. The work in this issue spans a diverse range of scale from satellite to unmanned aerial systems imagery, as well as ground-based measurements from mounted cameras, telemetry devices, and datalogging devices. Remote sensing-based technologies continue to evolve, allowing us to address critical issues facing grazingland management such as climate change, restoration, forage abundance and quality, and animal behavior, production, and welfare.

Generating accurate estimates of group sizes or behaviours of cetaceans from boat-based surveys can be challenging because much of their activity occurs below the water surface and observations are distorted by horizontal perspectives. Automated observation using drones is an emerging research tool for animal behavioural investigations. However, drone-based and boat-based survey methods have not been quantitatively compared for small, highly mobile cetaceans, such as Delphinidae. Here, we conduct paired concurrent boat-based and drone-based surveys, measuring the number of individuals in 21 groups and the behaviour within 13 groups of bottlenose dolphin (Tursiops truncatus). We additionally assessed the ability to detect behaviour events by the drone that would not be detectable from the boat. Drone-derived abundance counts detected 26.4% more individuals per group on average than boat-based counts (p = 0.003). Drone-based behaviour observations detected travelling 55.2% more frequently and association in subgroups 80.4% more frequently than boat-based observations (p < 0.001 for both comparisons). Whereas foraging was recorded 58.3% and resting 15.1% less frequently by the drone than by boat-based surveys, respectively (p = 0.014 and 0.024). A considerable number of underwater behaviours ranging from individual play activities to intra- and inter-species interactions (including those with humans) were observed from the drone that could not be detected from the boat. Our findings demonstrate that drone surveys can improve the accuracy of population counts and behavioural data for small cetaceans and the magnitude of the discrepancies between the two methods highlights the need for cautious interpretation of studies that have relied on boat-derived data.

Agriculture is considered a hotspot for wireless sensor network (WSN) facilities as they could potentially contribute towards improving on-farm management and food crop yields. This study proposes six designs of unmanned aerial system (UAS)-enabled data ferries with the intent of communicating with stationary sensor node stations in maize. Based on selection criteria and constraints, a proposed UAS data ferrying design was shortlisted from which a field experiment was conducted for two growing seasons to investigate the adoptability of the selected design along with an established WSN system. A data ferry platform comprised of a transceiver radio, a mini-laptop, and a battery was constructed and mounted on the UAS. Real-time monitoring of soil and temperature parameters was enabled through the node stations with data retrieved by the UAS data ferrying. The design was validated by establishing communication at different heights (31 m, 61 m, and 122 m) and lateral distances (0 m, 38 m, and 76 m) from the node stations. The communication success rate (CSR) was higher at a height of 31 m and within a lateral distance of 38 m from the node station. Lower communication was accredited to potential interference from the maize canopy and water losses from the maize canopy.

The development of high-throughput genotyping and phenotyping has provided access to many tools to accelerate plant breeding programs. Unmanned Aerial Systems (UAS)-based remote sensing is being broadly implemented for field-based high-throughput phenotyping due to its low cost and the capacity to rapidly cover large breeding populations. The Structure-from-Motion photogrammetry processes aerial images taken from multiple perspectives over a field to an orthomosaic photo of a complete field experiment, allowing spectral or morphological trait extraction from the canopy surface for each individual field plot. However, some phenotypic information observable in each raw aerial image seems to be lost to the orthomosaic photo, probably due to photogrammetry processes such as pixel merging and blending. To formally assess this, we introduced a set of image processing methods to extract phenotypes from orthorectified raw aerial images and compared them to the negative control of extracting the same traits from processed orthomosaic images. We predict that standard measures of accuracy in terms of the broad-sense heritability of the remote sensing spectral traits will be higher using the orthorectified photos than with the orthomosaic image. Using three case studies, we therefore compared the broad-sense heritability of phenotypes in wheat breeding nurseries including, (1) canopy temperature from thermal imaging, (2) canopy normalized difference vegetation index (NDVI), and (3) early-stage ground cover from multispectral imaging. We evaluated heritability estimates of these phenotypes extracted from multiple orthorectified aerial images via four statistical models and compared the results with heritability estimates of these phenotypes extracted from a single orthomosaic image. Our results indicate that extracting traits directly from multiple orthorectified aerial images yielded increased estimates of heritability for all three phenotypes through proper modeling, compared to estimation using traits extracted from the orthomosaic image. In summary, the image processing methods demonstrated in this study have the potential to improve the quality of the plant trait extracted from high-throughput imaging. This, in turn, can enable breeders to utilize phenomics technologies more effectively for improved selection.

Machine learning and computer vision technologies based on high-resolution imagery acquired using unmanned aerial systems (UAS) provide a potential for accurate and efficient high-throughput plant phenotyping. In this study, we developed a sorghum panicle detection and counting pipeline using UAS images based on an integration of image segmentation and a convolutional neural networks (CNN) model. A UAS with an RGB camera was used to acquire images (2.7 mm resolution) at 10-m height in a research field with 120 small plots. A set of 1,000 images were randomly selected, and a mask was developed for each by manually delineating sorghum panicles. These images and their corresponding masks were randomly divided into 10 training datasets, each with a different number of images and masks, ranging from 100 to 1,000 with an interval of 100. A U-Net CNN model was built using these training datasets. The sorghum panicles were detected and counted by a predicted mask through the algorithm. The algorithm was implemented using Python with the Tensorflow library for the deep learning procedure and the OpenCV library for the process of sorghum panicle counting. Results showed the accuracy had a general increasing trend with the number of training images. The algorithm performed the best with 1,000 training images, with an accuracy of 95.5% and a root mean square error (RMSE) of 2.5. The results indicate that the integration of image segmentation and the U-Net CNN model is an accurate and robust method for sorghum panicle counting and offers an opportunity for enhanced sorghum breeding efficiency and accurate yield estimation.

To understand how predators optimize foraging strategies, extensive knowledge of predator behavior and prey distribution is needed. Blue whales employ an energetically demanding lunge feeding method that requires the whales to selectively feed where energetic gain exceeds energetic loss, while also balancing oxygen consumption, breath holding capacity, and surface recuperation time. Hence, blue whale foraging behavior is primarily driven by krill patch density and depth, but many studies have not fully considered surface feeding as a significant foraging strategy in energetic models. We collected predator and prey data on a blue whale (Balaenoptera musculus brevicauda) foraging ground in New Zealand in February 2017 to assess the distributional and behavioral response of blue whales to the distribution and density of krill prey aggregations. Krill density across the study region was greater toward the surface (upper 20 m), and blue whales were encountered where prey was relatively shallow and more dense. This relationship was particularly evident where foraging and surface lunge feeding were observed. Furthermore, New Zealand blue whales also had relatively short dive times (2.83 ± 0.27 SE min) as compared to other blue whale populations, which became even shorter at foraging sightings and where surface lunge feeding was observed. Using an unmanned aerial system (UAS; drone) we also captured unique video of a New Zealand blue whale’s surface feeding behavior on well-illuminated krill patches. Video analysis illustrates the whale’s potential use of vision to target prey, make foraging decisions, and orient body mechanics relative to prey patch characteristics. Kinematic analysis of a surface lunge feeding event revealed biomechanical coordination through speed, acceleration, head inclination, roll, and distance from krill patch to maximize prey engulfment. We compared these lunge kinematics to data previously reported from tagged blue whale lunges at depth to demonstrate strong similarities, and provide rare measurements of gape size, and krill response distance and time. These findings elucidate the predator-prey relationship between blue whales and krill, and provide support for the hypothesis that surface feeding by New Zealand blue whales is an important component to their foraging ecology used to optimize their energetic efficiency. Understanding how blue whales make foraging decisions presents logistical challenges, which may cause incomplete sampling and biased ecological knowledge if portions of their foraging behavior are undocumented. We conclude that surface foraging could be an important strategy for blue whales, and integration of UAS with tag-based studies may expand our understanding of their foraging ecology by examining surface feeding events in conjunction with behaviors at depth.

In forestry, aerial photogrammetry by means of Unmanned Aerial Systems (UAS) could bridge the gap between detailed fieldwork and broad-range satellite imagery-based analysis. However, optical sensors are only poorly capable of penetrating the tree canopy, causing raw image-based point clouds unable to reliably collect and classify ground points in woodlands, which is essential for further data processing. In this work, we propose a novel method to overcome this issue and generate accurate a Digital Terrain Model (DTM) in forested environments by processing the point cloud. We also developed a highly realistic custom simulator that allows controlled experimentation with repeatability guaranteed. With this tool, we performed an exhaustive evaluation of the survey and sensor settings and their impact on the 3D reconstruction. Overall, we found that a high frontal overlap (95%), a nadir camera angle (90°), and low flight altitudes (less than 100 m) results in the best configuration for forest environments. We validated the presented method for DTM generation in a simulated and real-world survey missions with both fixed-wing and multicopter UAS, showing how the problem of structural forest parameters estimation can be better addressed. Finally, we applied our method for automatic detection of selective logging.

Unmanned Aerial Systems, or drones, are very helpful tools for managing open-pit mining operations and developing ecological restoration activities. This article presents a method for identifying water erosion processes in active quarries by means of drone imagery remote sensing, in the absence of pre-existing imagery or mapping for comparison. A Digital Elevation Model (DEM) with a spatial resolution (SR) >10 cm and an orthophoto with an SR >2.5 cm were generated from images captured with a drone and their subsequent photogrammetric processing. By using Geographical Information Systems tools to process the DEM, a detailed drainage network was obtained, the areas of detected water erosion were separated, and the watersheds in the gullies identified. Subsequently, an estimated DEM before the erosive processes was reconstructed by interpolating the gully ridges; this DEM serves as a reference for the relief before the erosion. To calculate the volume of eroded material, the DEM of Differences was calculated, which estimates the volume difference between the previously estimated DEM and the current DEM. Additionally, we calculated the material necessary for the geomorphological adaptation of the quarry and the slope map, which are two valuable factors closely related to the monitoring of erosive processes. The results obtained allowed us to identify the erosion factors quickly and accurately in this type of mining. In the case of water-filled quarries, it would be important to characterize the subsurface relief. Essentially, the presented method can be applied with affordable and non-invasive materials to create digital grid maps at 10 cm resolution, obtaining data ready for 3D metrics, being a very practical landscape modelling tool for characterizing the restoration evolution of open-pit mining spaces.

Limited range of emerging delivery vehicles, specifically unmanned aerial systems, creates gaps in last mile retail distribution networks and excludes significant numbers of potential customers. Implementing a two-stage distribution process is one approach that eliminates this deficit. However, dispatchers must load the vehicles hauling orders in the initial stage of the process and assign orders for delivery to transfer points in a way that ensures timely arrival while utilizing only the minimum required resources. This paper investigates the capacitated transfer point covering problem which has, until now, been unstudied. The research proposes a mathematical programming model to provide an optimal solution. The model is then applied to a realistic delivery situation using a distribution case study on the eastern coast of the U.S. and solved using a commercial optimization software. The results are analyzed, insights provided, and areas are identified for future research.

This study aimed to identify the optimal sets of spectral bands for monitoring multiple grapevine nutrients in vineyards. We used spectral data spanning 400–2500 nm and leaf samples from 100 Concord grapevine canopies, lab-analyzed for six key nutrient values, to select the optimal bands for the nutrient regression models. The canopy spectral data were obtained with unmanned aerial systems (UAS), using push-broom imaging spectrometers (hyperspectral sensors). The novel use of UAS-based hyperspectral imagery to assess the grapevine nutrient status fills the gap between in situ spectral sampling and UAS-based multispectral imaging, avoiding their inherent trade-offs between spatial and spectral resolution. We found that an ensemble feature ranking method, utilizing six different machine learning feature selection methods, produced similar regression results as the standard PLSR feature selection and regression while generally selecting fewer wavelengths. We identified a set of biochemically consistent bands (606, 641, and 1494 nm) to predict the nitrogen content with an RMSE of 0.17% (using leave-one-out cross-validation) in samples with nitrogen contents ranging between 2.4 and 3.6%. Further studying is needed to confirm the relevance and consistency of the wavelengths selected for each nutrient model, but ensemble feature selection showed promise in identifying stable sets of wavelengths for assessing grapevine nutrient contents from canopy spectra.

Forest disturbances—driven by pests, pathogens, and discrete events—have led to billions of dollars in lost ecosystem services and management costs. To understand the patterns and severity of these stressors across complex landscapes, there must be an increase in reliable data at scales compatible with management actions. Unmanned aerial systems (UAS or UAV) offer a capable platform for collecting local scale (e.g., individual tree) forestry data. In this study, we evaluate the capability of UAS multispectral imagery and freely available National Agricultural Imagery Program (NAIP) imagery for differentiating coniferous healthy, coniferous stressed, deciduous healthy, deciduous stressed, and degraded individual trees throughout a complex, mixed-species forests. These methods are first compared to assessments of crown vigor in the field, to evaluate the potential in supplementing this resource intensive practice. This investigation uses the random forest and support vector machine (SVM) machine learning algorithms to classify the imagery into the five forest health classes. Using the random forest classifier, the UAS imagery correctly classified five forest Health classes with an overall accuracy of 65.43%. Using similar methods, the high-resolution airborne NAIP imagery achieved an overall accuracy of 50.50% for the five health classes, a reduction of 14.93%. When these classes were generalized to healthy, stressed, and degraded trees, the accuracy improved to 71.19%, using UAS imagery, and 70.62%, using airborne imagery. Further analysis into the precise calibration of UAS multispectral imagery, a refinement of image segmentation methods, and the fusion of these data with more widely distributed remotely sensed imagery would further enhance the potential of these methods to more effectively and efficiently collect forest health information from the UAS instead of using field methods.