Types Of Unmanned Aerial Vehicles Research Articles

Аналіз наявних методів та підходів до пошуку пошкодженої автобронетанкової техніки під час технічної розвідки в сучасних арміях світу

The article proposes the definition of directions and opportunities for the use of technical means of information in the performance of the tasks of rear support of troops in armed conflicts that are taking place today.
 Possible areas of application of unmanned aerial vehicles in the technical intelligence system of the National Guard of Ukraine and the Armed Forces of Ukraine are considered. The analysis of various types of unmanned aerial vehicles revealed the peculiarities of their use. The foreign experience of the use and development of unmanned technologies in the field of search and evacuation of damaged armored vehicles was analyzed. Conclusions were obtained regarding the combination of unmanned aerial vehicles with traditional means of technical intelligence, which showed their significant advantages. Prospective areas of application of aircraft in combination with technical intelligence systems of individual units of the National Guard of Ukraine and military units of the Armed Forces of Ukraine during operations and hostilities have been identified. The possibility of combining aviation technologies and means of technical intelligence in armed conflicts of modern armies of the world is also considered.
 Technical intelligence was analyzed from the point of view of the type of intelligence by tasks, forces, means and methods; and also as an organization of intelligence activities based on the use of technical means. The dependence on the nature of signals detected by technical means is shown, one of the types of technical intelligence is photo and infrared intelligence based on the interception of light waves (species intelligence).
 Species intelligence is considered as the collection of intelligence information based on the analysis of a large number of images obtained with the help of photographic, optical-electronic and radar equipment. According to its specificity, intelligence belongs to technical intelligence and includes air and space intelligence. Species reconnaissance uses photographic images taken in the visible range of the electromagnetic spectrum, infrared photography and multispectral photography. Radar images for species reconnaissance are created by equipment of various electromagnetic ranges synthesized in radar equipment. Spectral reconnaissance should be distinguished from electronic reconnaissance using optical-electronic or radar equipment that does not form an image.
 Military experts of developed countries believe that in the modern combat environment, unmanned aerial vehicles (UAVs) for conducting reconnaissance as a means of technical intelligence can more effectively and quickly solve the tasks of technical intelligence. At the same time, the time required to deliver the received intelligence information to the relevant management bodies is reduced.

DECISION-MAKING HETEROGENEOUS UAV SWARM SYSTEM WITH NEURAL NETWORK-ENHANCED REINFORCEMENT LEARNING

This article explores how artificial intelligence and automation are significantly impacting unmanned aerial vehicles (UAVs), moving from traditional roles to versatile applications. The paper addresses the problem of optimizing the composition of a UAV swarm for efficient task execution by proposing an expert decision-making system that integrates neural networks and reinforcement learning. This system dynamically selects the optimal configuration for heterogeneous UAV swarms, in particular, for searching for objects in unfamiliar terrain. In the experimental phase, an advanced level of system was implemented by combining neural networks and reinforcement learning, based on role-based and MADDPG algorithms for heterogeneous UAV swarms. Decentralized information fusion-based swarm decision making algorithm (IFDSDA) is presented to overcome communication obstacles. The experiment presents a concept for improving heterogeneous UAV swarms using a neural decision network based on reinforcement learning. The environment is represented by a three-dimensional space with objects to be searched in random locations. The neural network evolves its decision-making strategy during training episodes, having an architecture with an input layer that processes information about the UAV’s state, hidden layers, and an output layer that influences the swarm’s behavior. The paper describes the process of direct propagation, reward-based weight adjustment, and the role of the output layer in determining collective actions. The results demonstrate the effective distribution of UAV types by the swarm based on a neural network, reducing redundancy and resource waste, thereby increasing overall efficiency. The article highlights the optimal solution obtained during the experiment, accompanied by a visual representation of the reward results.

Unmanned Aerial Vehicles (UAVs) in Landslide Investigation and Monitoring: A Review

Over the past decade, Unmanned Aerial Vehicles (UAVs) have emerged as essential tools for landslide studies, particularly in on-site investigations. This paper reviews UAV applications in landslide studies, with a focus on static geological characteristics, monitoring temporal and spatial dynamics, and responses post-events. We discuss the functions and limitations of various types of UAVs and sensors (RGB cameras, multi-spectral cameras, thermal IR cameras, SAR, LiDAR), outlining their roles and data processing methods in landslide applications. This review focuses on the UAVs’ roles in landslide geology surveys, emphasizing landslide mapping, modeling and characterization. For change monitoring, it provides an overview of the temporal and spatial evolution through UAV-based monitoring, shedding light on dynamic landslide processes. Moreover, this paper underscores UAVs’ crucial role in emergent response scenarios, detailing strategies and automated detection using machine learning algorithms. The discussion on challenges and opportunities highlights the need for ongoing UAV technology advancements, addressing regulatory hurdles, hover time limitations, 3D reconstruction accuracy and potential integration with technologies like UAV swarms.

UAV Atmosphere Sounding for Rocket Launch Support.

One of the crucial branches of activity at the Łukasiewicz Research Network-Institute of Aviation is developing a suborbital rocket vehicle capable of launching small payloads beyond the Earth's atmosphere, reaching over 100 km in altitude. Ensuring safety is a primary concern, particularly given the finite flight zone and impact area. Crucial to safety analysis is the wind profile, especially in the very first seconds of a flight, when rocket velocity is of the same order as the wind speed. Traditional near-ground wind data sources, ranging from wind towers to numerical models of the atmosphere, have limitations. Wind towers are costly and unfeasible at many test ranges used for launches, while numerical modeling may not reflect the specific ground profile near the launcher due to their large cell size (2 to +10 km). Meteorological balloons are not favorable for such measurements as they aim to provide the launch operator with a wind profile at high altitudes, and are launched only 1-2 times per flight attempt. Our study sought to prototype a wind measurement system designed to acquire near-ground wind profile data. It focuses on measuring wind direction and speed at near-ground altitudes with higher flight frequency, offering data on demand shortly before launch to help ensure safety. This atmosphere sounding system consists of an Unmanned Aerial Vehicle (UAV) equipped with an onboard ultrasonic wind sensor. Some reports in the literature have discussed the possibility of using UAV-borne anemometers, but the topic of measurement errors introduced by placing the anemometer onboard an UAV remains under studied. Limited research in this area underlines the need for experimental validation of design choices-for specific types of UAVs, anemometers, and mounting. This paper presents a literature review, a detailed overview of the prototyped system, and flight test results in both natural (outdoor) and controlled (indoor, no wind) conditions. Data from the UAV system's anemometer was benchmarked against a stationary reference weather station, in order to examine the influence of the UAV's rotor on the anemometer readings. Our findings show a wind speed Root Mean Square Error (RMSE) of 5 m/s and a directional RMSE of below 5.3° (both averaged for 1 min). The results were also compared with similar UAV-based wind measurements. The prototyped system was successfully used in a suborbital rocket launch campaign, thus demonstrating the feasibility of integrating UAVs with dedicated sensors for performing regular meteorological measurements in automatic mode.

Towards sustainable UAV operations: Balancing economic optimization with environmental and social considerations in path planning

Unmanned Aerial Vehicles (UAVs), i.e., drones, are expected to be widely used in various applications, such as parcel delivery and passenger transport, with the benefits of mitigating traffic congestion and reducing carbon emissions. In this paper, we study a UAV path planning problem under uncertain weather conditions, and design a data-driven dynamic decision support system for multiple types of UAVs. To this end, we categorize all relevant costs into three types, namely, economic, environmental, and social costs, and formulate a nonlinear two-stage stochastic programming model to establish optimal paths for UAV missions under weather uncertainty. We then discretize the nonlinear model and propose a tight linear approximation for the discretized problem to allow for a near real-time implementation. To quantify weather uncertainty, we propose a weather scenario generation algorithm to map ensemble-based weather forecast information to airspace blockage maps. With comprehensive computational studies through simulations, we show that our proposed stochastic approach can lower operating costs by an average of around 6%, where the savings increase as weather conditions become more severe and complex. We also find that, for missions operated by small UAVs, it is not sufficient to determine a path solely based on economic cost minimization, but it should rather be through total cost minimization, which involves environmental and social costs. Considering only the economic cost in the optimization may lead to much higher non-economic costs. However, for missions operated by large UAVs, it is sufficient to determine paths through economic cost optimization, as including environmental and social costs in the optimization process does not result in solutions that are much different from those obtained by considering only the economic costs. For both small and large UAVs, a path established solely through environmental or social cost minimization may not be economically sustainable, as doing so would imply very high economic costs.

BisSiam: Bispectrum Siamese Network Based Contrastive Learning for UAV Anomaly Detection

In recent years, a surging number of unmanned aerial vehicles (UAVs) are pervasively utilized in many areas. However, the increasing number of UAVs may cause privacy and security issues such as voyeurism and espionage. It is critical for individuals or organizations to manage their behaviors and proactively prevent the misbehaved invasion of unauthorized UAVs through effective anomaly detection. The UAV anomaly detection framework needs to cope with complex signals in noisy-prone environments and to function with very limited labeled samples. This paper proposes BISSIAM, a novel framework that is capable of identifying UAV presence, types, and operation modes. BISSIAM converts UAV signals to bispectrum as the input and exploits a siamese network-based contrastive learning model to learn the vector encoding. A sampling mechanism is proposed for optimizing the sample size involved in the model training whilst ensuring the model accuracy without compromising the training efficiency. Finally, we present a similarity-based fingerprint matching mechanism for detecting unseen UAVs without the need of retraining the whole model. Experimental results show that our approach outperforms other baselines and can reach 92.85% accuracy of UAV type detection in unsupervised learning scenarios, and 91.4% accuracy for detecting the UAV type of the out-of-sample UAVs.

FIRING EFFICIENCY OF ANTI-AIRCRAFT MISSILE COMPLEX “STRILA-10” IN TARGET DETECTION THROUGH THE OPTICAL VISOR AND AUTOMATED TARGET POINTER

When preparing for and conducting combat operations of the “SA-13 Gopher” anti-aircraft missile system, commanders and combat personnel of the system must know the missile firing efficiency indicators [1–4]. As an indicator of missile firing efficiency, the probability of hitting a target with one missile is usually used. Massive homing is carried out via the contrast channel or the infrared channel of the missile homing head. Under various conditions of use, it is necessary to evaluate the detection ranges that are realised and the probability of timely target detection, launch range and missile launch probability. It is also necessary to imagine the possible values of the inclined distances to the far edge of the kill zone of the complex. The probability of hitting a target with a single missile is affected by the geometric dimensions of the targets, their colour and contrast in the infrared wavelength range, the presence of target designation, meteorological range of visibility etc.
 The task of determining the firing efficiency of the “SA-13 Gopher” anti-aircraft missile system when detecting targets through the optical sight and automated targeting is therefore relevant and important.
 Research presented in the published [1–4; 12–13] do not allow determining the value of the probability of hitting different air targets in various conditions of the complex’s use. This literature provides only general approaches to solving this problem for typical conditions of use of an anti-aircraft missile system. A numerical modeling technique for determining the firing efficiency indicator in the form of values of the probability of hitting targets by a missile has been developed. The modeling took into account the speed of the targets, their plane in the shooting picture, their radiating capabilities and meteorological visibility range, the quality of the combat work of the operators of the combat vehicle, the sun exposure and the closing angles of the combat vehicle. Found possible detection ranges and detection probabilities of a typical target and a small unmanned aerial vehicle (UAV). A high probability of the operator of a combat vehicle missing a typical target at low and medium flight altitudes was noted. Considered launch ranges of missiles that are implemented and the probability of launching a missile when targets are captured in the contrast and infrared channels of the homing head. The largest inclined ranges to the far boundary of the zone of damage of the complex were calculated. The values of the probability of hitting a typical target and UAV in various firing conditions are determined. Sufficient firing efficiency has been proven when capturing and launching a missile in the contrast guidance channel and firing at a typical target. A comparison of the contrast and infrared channels of the homing missile head when firing at UAVs was carried out. Analytical expressions for the calculation of missile firing efficiency indicators and corresponding graphic material are presented.

Experimental Study on Modal Characteristics of Large Flexible Aircraft

High altitude long endurance unmanned aerial vehicle (UAV) is a type of UAV that operates at an altitude of around 10000 meters and can fly for several days, relying on solar panels.Usually, this type of aircraft has the structural characteristics of high aspect ratio, large wingspan, low wing load, and high flexibility.In engineering applications, it is often necessary to measure the vibration modes of such aircraft, and the modal measurements of such aircraft are often not accurately obtained.This article conducts relevant research on the issue of inaccurate measurement in modal experiments of such aircraft. By establishing a finite element model, the reasons for the inaccurate measurement of low order frequencies in modal experiments of such aircraft are analyzed, and the magnitude of data errors in different situations is quantitatively calculated.Based on the above research, this article proposes two methods for solving modal tests of such aircraft. This method has important guiding significance for ground modal testing of large flexible structures, which is beneficial for dynamic analysis and flight control research of such aircraft.

A Conceptual Digital Forensic Investigation Model Applicable to the Drone Forensics Field

Although there is a considerable amount of studies in drone forensics that describe numerous practical and technical perspectives, there is a lack of a comprehensive investigation framework. This study used design science research methodology to design a conceptual model for the comprehensive investigation of Unmanned Aerial Vehicles (UAVs) under forensic conditions. This model can identify, capture, preserve, analyze, and document UAV incidents. The proposed model consists of four stages: preparation, data collection, analysis, and documentation. In the preparation stage, data are collected and analyzed about UAV-related resources, including the origin and model of the aircraft, any software or hardware installed onboard, and the legal framework and regulations in place. The data collection stage involves the completion of the collection process, where participants gather parts of the UAV and the data needed, such as the flight controller, flight log, and memory cards. The analysis stage involves analyzing the collected evidence. Lastly, the documentation stage involves documenting relevant evidence, analysis results, and any conclusions derived. This model provides a comprehensive process to forensically investigate UAV incidents and provides an efficient and effective approach to the analysis of UAV evidence, ensuring that evidence was collected and analyzed according to accepted forensic techniques. The proposed model can be applied to any UAV type and legal framework.

Lightweight and attention-based CNN architecture for wildfire detection using UAV vision data

ABSTRACT Unmanned aerial vehicles (UAVs) are invaluable technologies concerning their remote control and monitoring capabilities. Convolutional neural networks (CNNs), known for their high pattern recognition capabilities, are appropriate for forest fire detection with UAVs. Deep convolutional neural networks show substantial performance on hardware with high processing capabilities. While these networks can be operated in unmanned aerial vehicles controlled from ground control stations equipped with GPU-supported hardware, the execution on a typical UAV’s limited computational resources necessitates the use of lightweight, small-sized networks. To overcome these impediments, this article presents a lightweight and attention-based approach for performing forest fire detection tasks using UAV vision data (images acquired by cameras mounted on UAVs). In this paper, we also present comprehensive research for different approaches such as transfer learning, deep CNNs, and lightweight CNNs. Among the experimented models, the attention-based EfficientNetB0 backboned model emerged as the most successful architecture for forest fire detection. With the test accuracy of , the F1-score of , the recall of , and the precision of have strongly reinforced the efficiency of the EfficientNetB0-based model in wildfire recognition. Moreover, the network has a less parameter size than the experimented networks. It proves the model’s suitability for wildfire detection with UAVs having limited hardware resources.

An Operational Capacity Assessment Method for an Urban Low-Altitude Unmanned Aerial Vehicle Logistics Route Network

The Federal Aviation Administration introduced the concept of urban air mobility (UAM), a new three-dimensional transport system that operates with a fusion of manned/unmanned aerial vehicles on an urban or intercity scale. The rapid development of UAM has brought innovation and dynamism to many industries, especially in the field of logistics. Various types of unmanned aerial vehicles (UAVs) for use in transport logistics are being designed and produced. UAV logistics refers to the use of UAVs, usually carrying goods and parcels, to achieve route planning, identify risk perception, facilitate parcel delivery, and carry out other functions. This research provides a method for assessing the operational capacity of a UAV logistics route network. The concept of “logistics UAV route network operation capacity” is defined, and a bi-objective optimization model for assessing the route network’s operating capacity is developed. The first objective is to maximize the number of UAV logistics delivery plans that can be executed in a fixed operation time. The second objective is to minimize the total operational impedance value in a fixed operation time. To solve the bi-objective optimization model, the Non-dominated Sorting Genetic Algorithm-II (NSGA-II) is utilized. A UAV logistics route network with 62 nodes is developed to assess the rationale and validity of the proposed concept. The experiments show that with an increase in operation time, the route network’s optimal operational capacity gradually increases, the convergence speed of the algorithm slows down, and the optimization magnitude gradually reduces. Two key parameters—operational safety interval and flight speed—are further analyzed in the experiments. According to the experiments, as the safety interval increases, the route network’s average operational capacity steadily diminishes, as does its sensitivity to the safety interval. The average operational capacity steadily rose with the rise in flight speed, especially when the UAV logistics flight speed was between 10 m/s and 10.5 m/s. In that range, the operational capacity of the route network was substantially impacted by the flight speed.

Auto-Landing of Moving-Mass Actuated Unmanned Aerial Vehicles Based on Linear Active Disturbance Rejection Control

Unlike the roll motion of the unmanned aerial vehicle (UAV) controlled by the ailerons, the moving-mass actuated unmanned aerial vehicle (MAUAV) uses the motion of the mass block inside the wing to generate the roll moment. The light weight and severe coupling of lateral and longitudinal motion of this type of small UAV make its landing control a challenging task. Considering the above problems, the dynamic models of MAUAV are first established. Then, forward velocity, altitude, attitude, and moving-mass position controllers are designed separately to make the MAUAV track a given path during the landing process. Linear active disturbance rejection control (LADRC) is introduced in the design process of all four controllers, compensating for unknown disturbances in the system. Simulation results show that the proposed control scheme can achieve fast and accurate tracking of forward velocity and flight trajectory commands with good robustness to model uncertainties.

DEZVOLTAREA UNUI UAV TRICOPTER-HEXAROTOR DESTINAT REALIZĂRII LUCRĂRILOR DE STROPIRE DE PRECIZIE

The utilization of unmanned aerial vehicles (UAVs) for chemical application has become a topic of great interest to both researchers and the market in recent times. Producers have proposed the use of drones for joint spraying as a means of enhancing precision agriculture and productivity. However, chemical spraying is a highly specialized field where the quality of the spray is critical so the used UAV have to be precise and reliable. Within this paper will be presented the process of developing a tricopter-hexarotor chassis type UAV, electrically driven, fitted with a 66 litres tank and 6 anti-drift nozzles which will be used for precision spraying of agricultural crops based on preprogrammed flying missions.

Development of a Hybrid Fixed-Wing UAV Aeromagnetic Survey System and an Application Study in Chating Deposit

In recent years, the development of airborne magnetic survey technology based on unmanned aerial vehicles (UAVs) has been rapidly advancing. The commonly used systems are the fixed-wing UAV-based, multi-rotors UAV-based, and unmanned helicopters-based magnetic survey systems. And, a type of hybrid UAV that uses a vertical takeoff and landing (VTOL) and fixed-wing cruise mode is increasingly being used to carry airborne magnetic survey systems. To meet the requirements of most UAVs for small-sized and lightweight payloads, a miniature magnetic survey system was developed and integrated into a hybrid fixed-wing UAV and formed an aeromagnetic survey system. And, a peripheral mineral exploration test was conducted in a known porphyry copper–gold deposit in southeastern China using the system. By processing the collected magnetic data with 3D inversion of susceptibilities, potential ore-bearing intrusive rocks were quickly identified and delineated, providing clues for peripheral mineral exploration in the mining area.

Data-Driven Insights into Population Exposure Risks: Towards Sustainable and Safe Urban Airspace Utilization by Unmanned Aerial Systems

With the rapid increase in unmanned aerial vehicles (UAVs), ensuring the safety of airspace operations and promoting sustainable development of airspace systems have become paramount concerns. However, research dedicated to investigating the population exposure risks of UAV operations in urban areas and their spatial pattern is still missing. To address this gap, this study evenly divides the urban space into uniform grids and calculates critical areas for two UAV types within each grid. By integrating geospatial data, including buildings, land use, and population, data-driven risk maps are constructed to assess the spatial distribution patterns and potential population exposure risks of two UAV types and compare them with commonly used census units. The results indicate that the mean time between failures (MTBF) for the selected generic and rotary-type UAVs can be up to 9.04 × 108 h and 1.22 × 108 h, respectively, at acceptable risk levels, considering uncertainties. The spatial pattern of population exposure risk exhibits spatial heterogeneity and multi-scale effects in urban areas, aligning with population distribution. High-risk areas concentrate in regions characterized by high population mobility, such as transport hubs, commercial service areas, residential zones, and business districts. Additionally, the comparation emphasizes the potential bias introduced by using census units in risk assessment, especially in regions with significant urban build-up. This framework enables the evaluation of safety and acceptability across diverse urban land use areas and offers guidance for airspace management in megacities, ensuring the safe integration of UAVs in urban environments.

Effect on damage of aircraft windshield impacted by light UAV with different postures

Abstract To study the performance of the main windshield of a commercial aircraft that has been verified to be airworthy by bird-strike tests against unmanned aerial vehicle (UAV) impact at high-speed, a typical light UAV with various possible flight postures and the main windshield of a commercial aircraft are considered. The transient impact responses at critical moments, energy change and contact force of a multi-layer windshield impacted by a UAV with different postures are investigated using a simulation method based on the models verified by the high-speed impact test between the whole UAV and the full-size nose. This study shows that the flight posture of the UAV has a significant effect on the damage to the windshield. When the abdomen of a typical light UAV maintains a posture parallel to the plane of the windshield, the high-speed impact would cause catastrophic damage to the windshield and no longer be airworthy. Simultaneously, the damage to the aircraft windshield caused by UAV collision is far more serious than that caused by bird strikes under similar collision conditions. The mass-concentrated components of the UAV and their high-hardness characteristics are the main factors of affecting multi-layer glass of windshield damage. The degree of damage to the windshield is positively related to the absorbed energy rather than the impact contact force. In this study, the impact simulation results between the windshield and UAV with different flight postures are verified qualitatively by testing, which provides a rational understanding and technical pre-research support for emerging and increasingly frequent potential safety hazards in air transport practice.

ANALYSIS AND DESIGN OF VTOL UAV WING USING FINITE ELEMENT ANALYSIS

Unmanned Aircraft (PTTA) or Unmanned Aerial Vehicle (UAV) is an aircraft that flies without a pilot and is controlled automatically. UAVs have a wide range of applications, including remote sensing, communication links, and natural disaster monitoring. One type of UAV being developed is LSU 05, using a composite of carbon fiber (CFRP) and glass fiber (GFRP) for its wing structure. This study aims to create a 3D model of the UAV VTOL wing, analyze the effect of material variations on the strength of the wing structure, and find the optimal wing design using the finite element analysis method. This study uses the SolidWorks 2021 software with maneuver conditions loading. The results of the analysis show that carbon fiber provides higher strength than fiberglass. In a fiberglass wing, the maximum stress is 229 MPa, while in a carbon fiber wing it is 197 MPa. The biggest displacement occurred in the fiberglass wing, which was 2.661 mm. The safety factor of the carbon fiber wing is 2.7, while that of the fiberglass is 1.9. This research provides an in-depth understanding of the behavior of VTOL UAV wing structures and contributes to the development of better wing designs. The optimized design can increase the performance, efficiency and reliability of VTOL UAV.

Usage of Uavs/ Drones Based on Their Categorisation: A Review

Unmanned Aerial Vehicle (UAV)/Drones earlier used in the military have now started their entry into each facet of life. Their entry has made them a subject of great relevance in the present technological era. The span of UAVs is, however, very broad due to various kinds of usages leading to different types of UAVs. This paper, therefore, presents all possible categorizations in respect of UAVs existing to date. The study of these categorizations in detail will lead everyone to study a Drone/ UAV for a specific field of usage as per requirement. This paper starts with the definition and description of a UAV in brief, followed by history and the underlying technology used for making different Drones. Subsequently, brief elucidation of different categorizations in respect of various UAVs expressed, followed by summary and conclusion.

Ungulate responses and habituation to unmanned aerial vehicles in Africa's savanna.

This article tests the hypothesis that "the likelihood that the species will react and level at which they do to the unmanned aerial vehicle (UAV) is related to the altitude, number of passes, sound intensity, type of UAV, takeoff distance, and species." This paper examined the behavioral responses of a group of free ranging ungulate species (Oryx, Kudu, Springbok, Giraffe, Eland, Hartebeest, and Impala) found in an animal reserve in Namibia to the presence of different in-flight UAV models. The study included 397 passes (trials) over 99 flights at altitudes ranging from 15 to 55 meters in three categories of response level: No response, Alert, and Movement. The ungulates were unhabituated to the UAVs and the study was conducted in the presence of stress-inducing events that occur naturally in the environment. Certain species were found to be more reactive than others, in addition to several displaying different response levels in single or mixed herd environments. Zebras were found to be less responsive in mixed herd environments while Oryx were present, as compared to when the Oryx were not; suggesting that some species may respond based on other species perception of threat or their relative fitness levels. The UAVs also produced inconsistent response rates between movement and alert behavior. The reference vehicle, Phantom 3 was much more likely than the Mavic to induce an alert response, while both having similar probabilities of inducing a movement response. Furthermore, the Custom X8 showed significantly more alert and movement responses than the other UAVs. This shows there may be several aspects to the UAVs that affect the responses of the ungulates. For instance, the sound intensity may alert the species more often, but close proximity may induce a movement response. More generally, the data shows that when the UAV is flying above 50 meters and has a measured sound intensity below 50 dB, the likelihood of inducing a movement response on an ungulate species is below 6% regardless of the vehicle on the first pass over the animals. Additionally, with each subsequent pass the likelihood of response dropped by approximately 20 percent. The results suggest a stronger correlation between flight altitude and response across the different ungulates, and the evidence suggests rapid habituation to the UAVs.

Ensemble Transfer Learning Based Cross-Domain UAV Actuator Fault Detection

The actuator is critical for controlling the unmanned aerial vehicle (UAV) attitude. With the development of sensing technology, the convenience of collecting high-quality data by sensors has provided favorable conditions for data-driven based UAV actuator fault detection methods. However, there are less historical data on a new type of UAV, making it challenging to accurately obtain data-driven fault detection models. The data domain distribution of the new UAV (target domain) is different from that of existing UAVs (source domains), and the basic data-driven methods are difficult to achieve cross-domain fault detection. The transfer learning of a single source domain method can solve this problem, but it has a strong dependence on the single source domain data characteristics. Therefore, we propose the ensemble transfer learning method, which learns multiple UAV data characteristics to address the accuracy degradation issue caused by single-source domain insufficient characteristics and realize cross-domain actuator fault detection with limited data for the new type of UAV. In the method, the basic actuator prediction models are established based on the bidirectional long short-term memory (BiLSTM) algorithm. Then we combine the transfer learning method to realize the cross-domain fault detection model establishment of the new type of UAV. In addition, a cross-domain similarity metric based on R-score is creatively proposed for the ensemble to avoid excessive dependence on a single-source domain. We conduct experiments based on the ArduPilot open-source simulation platform, and the results show that our proposed method can adaptively obtain more accurate fault detection performance.