Vision-based UAV Altitude Estimation using Deep Learning: A ResNet50 Approach on Nadir Images

Introduction. Due to their unique properties, unmanned aerial vehicles (UAVs) are widely used in the civil and military sectors. Due to the growing demand for specialist UAV operators, there is an increasing need to assess the working conditions at the workplaces of UAV operators. At the same time, special attention is paid to the sanitary and hygienic control of microbial contamination of the controls of unmanned vehicles. The study aims to execute a microbiological assessment of the qualitative and quantitative composition of microorganisms polluting the controls of unmanned aerial vehicles in order to further develop preventive measures aimed at preventing the spread of infectious diseases among operators of unmanned aerial vehicles. Materials and methods. The study assessed the quantitative and qualitative composition of microorganisms on control panels, keyboards, video cameras and video helmets of operators of unmanned aerial vehicles using both personal and multi-user equipment. Results. As a result of the study of microbial contamination of control panels, keyboards, video glasses and video helmets of operators of unmanned aerial vehicles, it was found that when operating equipment in multiuser access mode, the level of microbial contamination increases 2.9 times compared with personal use. In the course of the study, experts identified and ranked the main types of microorganisms living on the surfaces of video glasses, video helmets, keyboards and control panels of operators of unmanned aerial vehicles, with the aim of further developing and improving measures to preserve and promote the health of the aforementioned specialists. Limitations. It is necessary to study the effectiveness and safety of various disinfection methods for materials (plastic, glass, rubber, etc.) used to create UAV controls. To date, the impact of the use of glasses and helmets on the visual organ of operators of unmanned aerial vehicles in the long term of their use in the professional activities of specialists has not been fully studied, which does not allow fully assess the risks to their health. It is necessary to develop new solutions to minimize the impact of adverse factors on the organ of vision, including microbiological ones, as well as to minimize the manifestations of conjunctival symptoms. Conclusion. Multiuser equipment is most susceptible to microbial contamination, which increases the risk of infectious and inflammatory diseases among UAV operators. When using glasses and helmets, there are more pronounced manifestations of conjunctival symptoms due to their closer location to the organ of vision and the presence of additional fans in their design. The spectral composition of microorganisms contaminating UAV controls is mainly represented by representatives of opportunistic flora. The leading reason for contamination of equipment by opportunistic and pathogenic microorganisms is the limitation of the ability of UAV operators to comply with personal hygiene rules and insufficient treatment of equipment with disinfectant solutions. Ethics. The study was approved at a meeting of the Independent Ethics Committee at the Kirov Military Medical Academy on December 17, 2024 (Protocol No. 297 dated December 17, 2024). The study was conducted in compliance with ethical principles and the voluntary consent of the participants.

Operator Engagement During Prolonged Simulated UAV Operation

The unmanned aerial vehicles (UAVs) needs efficient energy management to ensure optimal performance and flight time.In this paper, the energy consumption and efficiency degradation in DJI Mini 2 drone batteries by the use of a deep neural network (DNN) for predictive analysis, was concern.The research conducted repeated flights and monitoring battery discharge from 100% to 27% over 20 trials.Experimental conditions, including flight duration and environmental factors, were controlled to ensure repeatability and to minimize any external influences on the recorded data.Data were stored onto AIRDATA (drone logbook) and then recollected for new labeling.The initial flights demonstrated similar, near constant performance, while following flights showed a gradual reduction in flight time (performance degradation).To ensure comparable power usage and minimize external influences, hover mode was selected for all flights.Next, on this data the DNN was trained using the metrics of mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), coefficient of variation of the root mean squared error (CVRMSE), and determination coefficient (R).The trained model achieved the MSE of 0.352%, RMSE of 0.593%, MAE of 0.324%, MAPE of 0.857%, CVRMSE of 0.743%, and R of 0.981.The obtained results show the DNN's ability to predict future power consumption for the UAV that in turn provides insights for energy management and extension of battery life.The paper contributes to the development of sustainable UAV operations by better knowledge about battery performance for in-flight conditions.

Unmanned aerial vehicles (UAVs) have been widely used in different tasks. Path planning is the most important factor in the operation of UAVs. This paper proposes a 3D path planning method based on an improved A* algorithm to train the UAV to fly in a real-world environment. The terrain data is first processed with a digital elevation model (DEM). Then the cost function, which balances route efficiency and environmental factors such as weather threats, is built. The available action space for the UAV is reduced considering the manoeuvrability of the vehicle. An improved A* algorithm is used to solve the search model, and the efficiency is enhanced by optimizing the data structure and building a weighted A* algorithm. Simulation results show that the method performed well in planning an optimal path in a 3D environment, and the algorithm is simple and efficient. This work contributes to path planning for UAVs in different missions.

Unmanned Aerial Vehicles (UAVs) have created safety problems for the publics. Assessments of the mental workload for UAV operations are essential to realize the causes of UAV accidents. To test the following hypotheses: i. mission difficulty in UAV operation affects both subjective and objective measures of mental workload; ii. mission difficulty affects number of failures in UAV operation. Fourteen male adults participated as UAV operators after attending a UAV training course. They performed four flight missions of different difficulty levels. During their flight missions, their heart rate and inter-beat interval (IBI) were collected. Upon completing each flight mission, the participants gave subjective ratings of mental workload using three commonly adopted assessment tools. The time of flight and number of failures in flight operations were also recorded. The results showed that mission difficulty affected the scores of all three assessment tools significantly. Mission difficulty also affected number of failures and IBI significantly. The scores of the three assessment tools were highly correlated (ρ= 0.7 to 0.83, p < 0.001) with one another. The results of the three subjective ratings were also consistent with that of the IBI data. High mental workload in UAV operation could lead to poor flight performance.

This article presents a cooperative multiagent deep reinforcement learning (MADRL) approach for unmanned aerial vehicle (UAV)-aided mobile edge computing (MEC) networks. An UAV with computing capability can provide task offlaoding services to ground Internet of Things devices (IDs). With partial observation of the entire network state, the UAV and the IDs individually determine their MEC strategies, i.e., UAV trajectory, resource allocation, and task offloading policy. This requires joint optimization of decision-making process and coordination strategies among the UAV and the IDs. To address this difficulty, the proposed cooperative MADRL approach computes two types of action variables, namely, message action and solution action, each of which is generated by dedicated actor neural networks (NNs). As a result, each agent can automatically encapsulate its coordination messages to enhance the MEC performance in the decentralized manner. The proposed actor structure is designed based on graph attention networks such that operations are possible regardless of the number of IDs. A scalable training algorithm is also proposed to train a group of NNs for arbitrary network configurations. Numerical results demonstrate the superiority of the proposed cooperative MADRL approach over conventional methods.

The future of low altitude urban airspace utilization is poised to be transformed by the increasing dominance of small Unmanned Aerial Vehicles (UAVs). Their application can essentially be categorized as a threefold problem: establishing the regulatory principles for UAV operation, identification of the infrastructure needed to monitor the UAVs and ensure they adhere to the rules, and laying out the minimum performance requirements of the platforms so that they are capable of meeting the flight standards. This work addresses the pressing challenge of enhancing the performance envelope of small UAVs. The primary aim of this research is to develop an innovative overactuated multirotor system that augments UAV maneuverability, fault tolerance and operational safety. A crucial aspect of this study is overcoming the inherent underactuation limitations in standard quadrotors. This challenge is addressed through active propeller tilting, investigated in-depth through extensive real-world flight tests and simulations. The developed quadrotor, aptly named QuadPlus, efficiently combines actuator redundancy and thrust vectoring via propeller rotation in a compact package, compared to its state-of-the-art counterparts. Furthermore, a cascade control methodology based on incremental control allocation is devised to ensure stable attitude and velocity tracking even in the presence of actuator saturation and external disturbances during flight operations. The study showcases the successful realization of the QuadPlus multirotor system with an impressive 12 degrees of freedom. The innovative design allows for independent bi-axial tilting of 115 ° and 180 ° about two perpendicular axes while keeping the mechanical complexity relatively low. Real-world flight tests demonstrate the efficacy of the control methodology, showcasing stable hover and trajectory tracking in a desired hyperplane with accurate attitude tracking. Different test cases demonstrating the capability of QuadPlus to achieve symmetric (20 ° roll and pitch) and asymmetric (20 ° roll, 10 ° pitch and vice versa) attitude with static hover are presented. It is also tested against Level 3 wind conditions of 4m/s and actuator saturation. The extensive bi-axial tilting capability of the system enables precise force production of up to 9N for contact-based tasks, presenting promising potential for inspection-related applications. The research also significantly contributes to ensuring fail-safe UAV operations. A centralized fault-tolerant control (FTC) framework based on nonlinear model predictive control (NMPC) is proposed to address potential propeller failures, effectively mitigating attitude deviations and enhancing overall system control. The FTC framework is validated through intensive simulations, proving its effectiveness in maintaining stable flight even in the presence of failures. The system is successfully tested against a single propeller failure while following an attitude of 30 ° each in roll and pitch. It is also successfully tested for failure recovery in aggressive flights following an 8-shaped trajectory with the attitude varying between ±30 ° for roll and pitch, and ±25 ° for yaw. The corresponding maximum deviations from the commanded trajectory are observed to be within acceptable limits. Additionally, the research introduces a novel approach for on-board state-of-health (SoH) diagnostics of the battery, a critical component of the multirotor system. The energy equilibrium approach enables better management of battery performance, ensuring safe and reliable UAV operations while avoiding rapid performance deterioration. Experimental work is performed to validate the model. However, further analysis onboard QuadPlus is presented as future work as part of realizing the fault tolerance capabilities. It is hypothesized that the integration of the SoH diagnostics into the flight controller will help keep the current drawn within the safe functional limits of the battery. In summary, this research substantially advances the field of UAV technology by presenting a groundbreaking solution to the challenges posed by low altitude urban airspace utilization. The innovative overactuated multirotor system, QuadPlus, coupled with the fault-tolerant control framework and on-board component diagnostics approach, holds immense potential for inspection-related applications and interactions with the physical environment.

Missing data reconstruction in attitude for quadrotor unmanned aerial vehicle based on deep regression model with different sensor failures

The article identifies the conditions and factors that will affect the operation of unmanned aerial vehicles during its use in emergencies. The purpose of the study is to systematize the conditions and factors that will affect the functioning of specific technical means of surveillance installed on the unmanned aerial vehicle. The study of conditions and factors was carried out with a preliminary determination of the general indicators of efficiency of specific technical means of observation installed on the drone, which include: the probability of detection and recognition of the object by the operator (pilot-operator) on the display screen; the scale of the image on the display screen; viewing angle of the species technical means of observation; transverse capture of the area during aerial surveillance with the help of specific technical means; spectral range of operation of the species technical means of observation and the height of observation. According to the results of research, the conditions and factors include: unmanned aerial vehicle; set task for observation (monitoring); terrain features; control by the pilot (pilot-operator); weather conditions; gusts of wind; season; time of day; drone vibrations during flight; atmospheric turbulence; control of specific technical means of observation by the operator (pilot-operator); temperature regime; the presence of a library of signatures of the object of observation. It is concluded that the results of the study of conditions and factors that will affect the use of specific technical means of observation installed on an unmanned aerial vehicle are the basis for developing relevant requirements for specific technical means of observation of visible and infrared wavelengths of the electromagnetic spectrum. Further research should be carried out in the following areas: development or purchase of unmanned aerial vehicles for observation during the prevention and elimination of emergencies; development of libraries of signatures of objects of observation; research of possibilities of application of spectrozonal or multispectral species equipment for observation; creation of special training grounds for training and retraining of pilots and operators of unmanned aerial vehicles, etc.

In this study, we generated Digital Elevation Model (DEM) from Unmanned Aerial Vehicle (UAV) to confirm the suitability of UAV to the tidal flat study. The DEMs were generated from aerial triangulation method using rotary-wing UAV. For the accurate generation of mosaic images and DEM, the distorted images occurred by interior and exterior orientation were corrected using camera calibration. In addition, we set up a Ground Control Points (GCPs) in order to correct of the UAV position error. Therefore, the DEM was obtained with geometric error less than 30 cm. The height of generated DEM by UAV was compared with the levelled elevation by RTK-GPS and TanDEM-X radar satellite DEM. From this study, we could confirm that accurate DEM of the tidal flat can be generated using UAV and these detailed spatial information about tidal flat will be widely used for tidal flat management.

Мета: Відсутність алгоритму рекомендацій дій оператора БПЛА в надзвичайних ситуаціях, розкладання процесу прийняття рішень оператором БПЛА в надзвичайних ситуаціях, розробка структури системи підтримки прийняття рішень для оператора БПЛА, розробка моделей прийняття рішень оператором БПЛА. Методи: Створення алгоритму прийняття рішень за допомогою критеріїв Вальда, Гурвіца, Ла Пласа. Результати: програма "UAV_AS", що дає рекомендації оператору БПЛА про те, як діяти в разі виникнення надзвичайної ситуації. Авторами розроблено алгоритм дій рекомендації оператора БПЛА в надзвичайних ситуаціях. Представлений процес прийняття рішень оператором БПЛА в надзвичайних ситуаціях у вигляді детермінованої моделі із застосуванням методів мережевого моделювання. Побудовано мережевий граф, за яким визначено критичний час на парирування надзвичайної ситуації у разі польоту БПЛА. Розроблено структуру розподіленої СППР для колективного управління пілотованих та безпілотних літальних апаратів оператором. Авторами розроблено бази даних для локальних СППР операторів управління БПЛА, Розроблено моделі прийняття рішень оператором БПЛА в умовах ризику та в умовах невизначеності у разі аварійної посадки БПЛА. Обговорення: У статті розглядаються проблеми польотів безпілотних літальних апаратів для вирішення різних завдань в надзвичайній ситуації, засноване на статистичних даних від аналізу типів надзвичайних ситуацій для безпілотних літальних апаратів. Визначається послідовність дій оператора БПЛА в разі виникнення надзвичайних ситуацій.

Autonomous Landing of an Unmanned Aerial Vehicle using Image-Based Fuzzy Control

In this paper, the aerodynamic performance of a Hex-rotor unmanned aerial vehicle (UAV) with different rotational speeds (1500–2300 RPM) considering the horizontal airflow conditions is analyzed by both simulations and experiments. A low-speed wind tunnel experiments platform is applied to measure the thrust, torque, and power consumption of a Hex-rotor UAV with different rotational speeds in horizontal airflow, which varied from 0 m/s–4 m/s. First, this paper introduces the effect of horizontal airflow on a UAV. Then, the low-speed wind tunnel experiments were carried out on a Hex-rotor UAV (D/L = 0.56) with different horizontal velocities to determine the hover performance. Finally, numerical simulations were obtained with the streamline distributions, pressure distributions, velocity contour, and vortex distributions at different horizontal airflow conditions to describe the aerodynamic interference effect of different horizontal airflows. Combined with the experimental results and numerical simulations results, the horizontal airflow proved to have a significant influence on the aerodynamic performance of the Hex-rotor UAV with an increase in thrust and power. Indeed, the streamlines in the flow field were coupled to each other at the presence of the incoming airflow. Especially when the incoming airflow was larger, the Hex-rotor UAV could properly use low-speed flight to maintain high power loading. Finally, it is inferred that the aerodynamic performance of the Hex-rotor UAV is also related to the movement and deformation of the vortex at the tip of the rotor.

Wireless communication is an essential technology to unlock the full potential of unmanned aerial vehicles (UAVs) in numerous applications and has thus received unprecedented attention recently. Although technologies such as direct link, WiFi, and satellite communications are still useful in some remote scenarios where cellular services are unavailable, it is believed that exploiting the thriving 5G and beyond cellular networks to support UAV communications is the most promising and cost-effective approach, especially when the number of UAVs grows dramatically. On the one hand, to guarantee safe and efficient flight operations of multiple UAVs, it is of paramount importance to provide secure and ultra-reliable communication links between the UAVs and their ground pilots or control stations for conveying command and control signals, especially in beyond-visual-line-of-sight (BVLOS) scenarios. On the other hand, because of advances in communication equipment miniaturization as well as UAV manufacturing, mounting compact and lightweight base stations (BSs) or relays on UAVs becomes increasingly feasible. This has led to two promising research paradigms for UAV communications, namely, UAV-assisted cellular communications and cellular-connected UAVs, where UAVs are integrated into cellular networks as aerial communication platforms and aerial users, respectively. As such, integrating UAVs into cellular networks is believed to be a win-win technology for both UAV-related industries and cellular network operators, which not only creates plenty of new business opportunities but also benefits the communication performance of 3-D wireless networks. In addition, UAV related sensing and computing are also helpful for achieving efficient and reliable communication (e.g., in avoiding coverage holes) as well as smart UAV coordination, positioning, and trajectory design. However, 5G and beyond wireless networks with UAVs significantly differs from traditional communication systems, because of the high altitude and high maneuverability of UAVs, the unique UAV-ground channels, the diversified quality of service (QoS) requirements for downlink command and control (C&C) and uplink mission-related data transmission, the stringent constraints imposed by the size, weight, and power (SWAP) limitations of UAVs, as well as the new design degrees of freedom enabled by joint UAV mobility control and communication resource allocation.

Abstract. Unmanned Aerial Vehicles (UAVs) are gaining popularity with agriculture producers to perform daily activities such as monitoring livestock and plant health. With the expense of a UAV, producers require more than just pictures and video to be cost effective. This project sought to retrofit a device on a UAV to capture temperature data emitted from ear tags on cattle. The questions to be answered were: (1) can the UAV handle the weight of the recorder without negatively affecting flight time, (2) what is the flying height that could be used for data recording, and (3) does the UAV negatively affect cattle behavior when flying over? UAV performance while taking livestock data was recorded daily for two months. Results indicated the UAV could carry the data receiver and its power source with a negligible reduction in flight time. Battery life was reduced from an average of 19 minutes to 17.5 minutes. Also, data was accurately recorded at flying heights of 50-400 feet, with the optimal altitude determined to be 150 feet. At this altitude, a producer could potentially monitor cattle on 38 acres per hour. At all flying heights, the animals appeared unaffected by the UAV. The ability to remotely monitor livestock health adds to the advantages a producer can reap from utilizing UAV technology.