Unmanned Aerial Vehicle Applications Research Articles

A Survey: Future Smart Cities Based on Advance Control of Unmanned Aerial Vehicles (UAVs)

This article presents a survey of unmanned aerial vehicle (UAV) applications in smart cities, emphasizing integration challenges. Smart cities leverage innovative technologies, including the Internet of Things (IoT) and UAVs, to enhance residents’ quality of life. The study highlights UAV applications, challenges, limitations, and future perspectives of smart city development. Advanced control methods for maximizing UAV benefits are discussed. Control theory challenges and issues for the deployment of UAVs are addressed. By concentrating on challenges, potential applications, and advanced control techniques, this paper offers insights into UAVs’ role in shaping the future of smart cities.

Unmanned Aerial Vehicles: The Possibility of Using Fuel Cell, Battery, and Turbocharger Technologies

The adverse impacts of fossil fuels on the environment have guided the authorities to set the official 100% phase out target for sales or registrations of new internal combustion engines (ICEs) that use fossil fuels as the prime mover. Batteries and fuel cells have proved to be an appropriate alternative for the ICEs to be used in cars, trucks, and buses. Although there have been limited number of studies to use batteries as the prime mover of small size planes, the possibility of using the integrated system of battery and fuel cell is yet to be investigated in the airplane industry.Due to the infancy of using batteries and fuel cells in airplane sector, the studies are limited to the domain of Unmanned Aerial Vehicles (UAVs). UAVs are designed to fulfill long and high-altitude flight missions. The flight duration and flight altitude have been proved to be improved using the proton exchange membrane fuel cells (PEMFCs) as the prime-mover of the UAVs. However, the low acceleration of the PEMFCs has triggered the idea of integrated PEMFCs and Lithium-Ion (Li-ion) battery. Thus, the integrated system can improve the low range of batteries by maintaining high acceleration and fast re-fueling time.Among different types of fuel cells, PEMFCs are established to be the best option for mobility applications, and the well-known Nickel Manganese Cobalt (NMC), and Lithium Titanite Oxide (LTO) Li-Ion batteries are considered as the most efficient types. LTO enjoys from the highest performance, lifespan, and safety by higher costs, whereas NMC provides acceptable performance with low costs and high specific energy.Published results have indicated that PEMFC can be utilized in an UAV, reducing the overall weight and leading to the longest flight endurance. However, better acceleration and simpler configurations can be achieved by Li-Ion battery packs, although the practical implementation is still limited by the actual chemistry. So far, it is believed that fuel cells can be suitable for cruise flight while batteries are more appropriate for other flight modes.In addition to the possibility of the integrated battery and fuel cell system to improve the overall performance of the UAVs, the integration of the turbochargers to the PEMFCs is believed to facilitate the transition from engines to more fuel cell/battery technologies. To reach the maximum thrust and minimum fuel consumption rate in an UAV, a hybrid propulsion system of a PEMFC and a single stage turbocharged engines is able to provide the maximum thrust, the minimum fuel consumption rate, and the thermal efficiency of 34.58 mg/s, 69.56N, and 62.41%, respectively.This study evaluates the possibility of integrating the fuel cell technology to the batteries as an alternative for fossil fuel-based combustion engines for the small sized UAV application. To further improve the performance of the UAV, a turbocharger is also integrated to reach higher thrusts and lower fuel consumption rates. The suggested integrated system, which is shown in Fig. 1., is designed to provided 6.8 kW power for a UAV with TWR (Thrust to Weight Ratio) of 0.4 and weight of 17.7kg. PEMFC will be used as the selected type of fuel cell while NMC is the considered type of Li-Ion battery. The turbocharger is also a hydrogen fueled ICE with the compression ratio of 11.15. The goal of this study is the performance evaluation of the current integrated system in an optimized condition considering the size, weight, hydrogen consumption, thrust, and overall efficiency as the output parameters. Figure 1

Designing UAV Swarm Experiments: A Simulator Selection and Experiment Design Process.

The rapid advancement and increasing number of applications of Unmanned Aerial Vehicle (UAV) swarm systems have garnered significant attention in recent years. These systems offer a multitude of uses and demonstrate great potential in diverse fields, ranging from surveillance and reconnaissance to search and rescue operations. However, the deployment of UAV swarms in dynamic environments necessitates the development of robust experimental designs to ensure their reliability and effectiveness. This study describes the crucial requirement for comprehensive experimental design of UAV swarm systems before their deployment in real-world scenarios. To achieve this, we begin with a concise review of existing simulation platforms, assessing their suitability for various specific needs. Through this evaluation, we identify the most appropriate tools to facilitate one's research objectives. Subsequently, we present an experimental design process tailored for validating the resilience and performance of UAV swarm systems for accomplishing the desired objectives. Furthermore, we explore strategies to simulate various scenarios and challenges that the swarm may encounter in dynamic environments, ensuring comprehensive testing and analysis. Complex multimodal experiments may require system designs that may not be completely satisfied by a single simulation platform; thus, interoperability between simulation platforms is also examined. Overall, this paper serves as a comprehensive guide for designing swarm experiments, enabling the advancement and optimization of UAV swarm systems through validation in simulated controlled environments.

Enabling secure image transmission in unmanned aerial vehicle using digital image watermarking with H-Grey optimization

Drone technology, also known as Unmanned Aerial Vehicles (UAVs), has advanced rapidly in the last decade owing to the huge number of users. This approach has an immense opportunity in areas like healthcare, agriculture, and forestry. However, the transmission of digital images via drone technology is associated with vulnerabilities and risks, due to a lack of efficient security solutions. Digital Image Watermarking (DIW) has emerged as a viable solution for digital image transmission in UAV applications. However, achieving robustness, security, and imperceptibility in a watermarking system at the same time is a difficult task. To address the aforementioned concerns, this paper proposes a secure digital image watermarking scheme in the hybrid DWT-SVD domain. The scaling factor is chosen adaptively from the hybrid Harmony-Grey Wolf (H-Grey) optimization algorithm to maintain a balance between imperceptibility and robustness. In addition, a novel symmetric cryptographic-based four-level encryption approach PSMD (Partitioning, Substituting, Merging, Division) is proposed to address UAV image security concerns by using Fibonacci and prime number series to generate two random keys. The proposed encryption scheme provides high security at low computational cost. The experimental results show that the visual quality and robustness are both high (Avg. PSNR = 41.27 dB and Avg. SSIM = 0.99, NC = 0.99, and BER = 0.0008 calculated for 100 images). The subjective and objective experimental analysis indicates that the proposed encryption scheme is highly secure and the computational cost is also low. The average embedding and extraction time is 0.25 s and 0.09 s It is resistant to various image processing attacks. A comparison with some of the most recent popular schemes confirms the scheme’s effectiveness. The presented DIW can be used for copyright protection, UAV image transmission applications, military applications, and other purposes.

Quantitative Trait Locus (QTL) Mapping for Common Wheat Plant Heights Based on Unmanned Aerial Vehicle Images

The application of unmanned aerial vehicle (UAV) remote sensing technology for high-throughput acquisition of phenotypic values of field plant height is of great significance in plant height monitoring of wheat varieties (lines). Based on the UAV, mounted with high-resolution digital cameras, a low-altitude remote sensing platform was built to obtain images of the Berkut/Worrakatta recombinant inbred lines (RIL) wheat population (297 lines). The optimal digital surface model (DSM) for plant height extraction was constructed by combining the actual wheat plant heights measured in the field. Using a wheat 50K SNP chip, quantitative trait locus (QTL) analysis was performed for field plant height of the RIL population using UAV plant height data to detect the locus related to plant height. We verified the effectiveness of QTL analysis using the UAV plant height data. The results of the DSM plant height extraction method showed that the plant height extraction models during the entire growth period had a highly significant correlation (p < 0.001); the measured plant height correlated with the predicted plant height, with prediction model accuracy R2 = 0.7527 and verification model accuracy R2 = 0.8214. QTL analysis was conducted on the manually and UAV-measured plant height traits. For the manually measured plant height phenotypes, one locus related to plant height was detected on wheat chromosome 6A and explained 13.12% of phenotypic variation. For the UAV-measured phenotypes, one locus related to plant height was also detected on wheat chromosome 6A and explained 9.62% of phenotypic variation. The QPH.xjau-6A locus on chromosome 6A, which featured in the results of the two measurement methods, is a stable locus, indicating that the mapping results obtained using the actual plant height values were in good agreement with those obtained using the UAV extracted values. Three candidate genes related to plant height were screened: they encode protein kinase, NAC domain protein, and cytochrome P450, respectively. Therefore, this study provided reference information for monitoring plant phenotype and growth of wheat and also for the extraction of plant height for wheat breeding in the future.

Qualification of a Wankel Rotary Engine for UAV Application - Airworthiness Challenges

An indigenous Wankel Rotary Engine of 40 kW class has been developed for Unmanned Aerial Vehicle (UAV) application. It should be capable of flying flawlessly for long duration at designated altitude and carrying the payload intended for the mission. The engine has to withstand high shock loads during launching from a mobile launcher and has to climb to the designated altitude within a specified time. To meet these requirements the engine needs to deliver required power while operating at its maximum speed during take-off and climb. Therefore qualification of a Wankel rotary engine for this type of application is a great challenge and needs thorough study of its component durability and engine performance and reliability. This paper presents the various qualification tests stipulated on the components, systems and on engine to establish its performance, integrity and reliability. The paper also highlights the basis of clearing the engine for its first flight. Engine made its maiden flight successfully on an UAV platform exhibiting satisfactory performance.

Performance evaluation of RW‐Quadrotor and Bi‐Quadrotor for payload delivery

AbstractPayload delivery is one of the many Unmanned Aerial Vehicle (UAV) applications that save time, energy, and human resources. The present simulation study compares rotary‐wing quadrotors (RW‐Quadrotors) and tail‐sitter biplane quadrotors (Bi‐Quadrotors) under payload delivery conditions that help to choose the best UAV for the particular mission. A comparative study is performed based on (i) time taken to accomplish the given mission, (ii) trajectory tracking performance, (iii) motor speed throughout the entire flight envelope, and (iv) remains SoC (State of Charge) after the mission. Furthermore, a Backstepping Controller (BSC) and an Adaptive Backstepping Controller (ABSC) are developed for both UAVs to handle mass changes during flight while Genetic Algorithm (GA) is used for the gain optimization. MATLAB Simulink‐based numerical simulation reveals that the Bi‐Quadrotor has a lower average rotor speed than the RW‐Quadrotor. However, the Bi‐Quadrotor's SoC after the mission is higher than the RW‐Quadrotor, and it took less time to accomplish the mission. The ABSC designed for RW‐Quadrotor and Bi‐Quadrotor effectively handles mass change during the mission. However, the BSC controller increases energy consumption for the Bi‐Quadrotor in the presence of wind gusts.

Effect of membrane electrode assembly characteristics on the performance of a proton exchange membrane fuel cell stack designed for unmanned aerial vehicle applications

ABSTRACT A proton exchange membrane fuel cell stack designed for unmanned aerial vehicle applications was used to evaluate the influence of several membrane electrode assembly (MEA) characteristics. MEAs prepared with catalytic layers (CLs) deposited directly on the membrane presented slightly better performance than those with CLs deposited in the gas diffusion layers (GDLs). Thinner membranes were beneficial to the stack performance due to lower activation and ohmic losses. Increasing the Pt loading on the cathode side from 0.5 to 2.0 mg Pt cm−2 did not improve the performance. Small differences were observed between carbon cloth and carbon paper GDLs, possibly due to the presence of a microporous layer (MPL). The hydrophobic treatment of the GDLs improved the performance due to a better humidification of the MEAs. The MPL was found fundamental for a high performance due to an improvement in the electric contact, decreasing ohmic losses. The stack operation under realistic conditions, consisting of a transient energy profile typical of an UAV flight, was also investigated. The stack presented higher dynamic performance than its steady-state one at low currents but slightly lower at high currents. Differences in temperature were observed, which probably contributed to this disparity.

Research on Inertial Navigation Technology of Unmanned Aerial Vehicles with Integrated Reinforcement Learning Algorithm

With the continuous expansion of unmanned aerial vehicle (UAV) applications, traditional inertial navigation technology exhibits significant limitations in complex environments. In this study, we integrate improved reinforcement learning (RL) algorithms to enhance existing unmanned aerial vehicle inertial navigation technology and introduce a modulated mechanism (MM) for adjusting the state of the intelligent agent in an innovative manner [1,2]. Through interaction with the environment, the intelligent machine can learn more effective navigation strategies [3]. The ultimate goal is to provide a foundation for autonomous navigation of unmanned aerial vehicles during flight and improve navigation accuracy and robustness. We first define appropriate state representation and action space, and then design an adjustment mechanism based on the actions selected by the intelligent agent. The adjustment mechanism outputs the next state and reward value of the agent. Additionally, the adjustment mechanism calculates the error between the adjusted state and the unadjusted state. Furthermore, the intelligent agent stores the acquired experience samples containing states and reward values in a buffer and replays the experiences during each iteration to learn the dynamic characteristics of the environment. We name the improved algorithm as the DQM algorithm. Experimental results demonstrate that the intelligent agent using our proposed algorithm effectively reduces the accumulated errors of inertial navigation in dynamic environments. Although our research provides a basis for achieving autonomous navigation of unmanned aerial vehicles, there is still room for significant optimization. Further research can include testing unmanned aerial vehicles in simulated environments, testing unmanned aerial vehicles in realworld environments, optimizing the design of reward functions, improving the algorithm workflow to enhance convergence speed and performance, and enhancing the algorithm's generalization ability. It has been proven that by integrating reinforcement learning algorithms, unmanned aerial vehicles can achieve autonomous navigation, thereby improving navigation accuracy and robustness in dynamic and changing environments [4]. Therefore, this research plays an important role in promoting the development and application of unmanned aerial vehicle technology.

Robust Predictor-Based Control for Multirotor UAV With Various Time Delays

This paper presents a robust predictor-based sliding mode control (RPSMC) for multirotor unmanned aerial vehicles (UAVs) to ensure desired tracking control under time delays which appear in practice by communications, complex computation, and actuator delays. Many UAV applications have difficulty in control and operation due to various types of time delays, resulting in repeated commands, severe control instability, and then mission failure. However, existing controllers have limitations in solving time delay problems. In this paper, the RPSMC with the prediction of future disturbance and reference trajectory is developed for the multirotor UAV to minimize the effects of time delay, robustly deal with external disturbances, and further achieve desired tracking control. The performance of RPSMC for the multirotor UAV is verified under various time delays and disturbances in numerical simulations. The results show the robustness and fast control convergence compared with proportional-integral-differential and conventional predictor-based controllers. Experimental results with step response and tracking of circular trajectory demonstrate the feasibility and performance of the RPSMC for UAVs in the presence of various time delays and disturbances.

Geographic Artificial Intelligence and Unmanned Aerial Vehicles Application for Correlation Analysis of Settlement Density and Land Surface Temperature in Panggang Island Jakarta

This research aims to understand the relationship between high settlement density and land surface temperature associated with the Urban Heat Island (UHI) phenomenon. The data processing method involves collecting settlement density data using UAVs equipped with thermal sensors, as well as using GeoAI, namely GEE, to analyze LST on Panggang Island. The results showed a positive relationship between settlement density and LST on Panggang Island, with high settlement density contributing to an increase in ground surface temperature. The benefits of the application of GeoAI and UAVs in this analysis include accurate mapping, understanding the impacts of urbanization, sustainable urban planning, and fact-based decision-making. It is hoped that this research can contribute to better urban management and reduction of environmental impacts in Pulau Panggang, DKI Jakarta.

UAV-based emergency treatment plan for flood disasters at the Hongyanhe nuclear power plant

This paper explores the potential of the joint application of unmanned aerial vehicle(UAV)-based the belief-desire-intention (BDI) algorithm constructed for rescue by NetLogo and the flood model of the Hongyanhe Nuclear Power Plant based on the Global Surface Water (GSW) dataset. The study revealed the high potential of the GSW dataset provides useful data for detecting the occurrence of flood events about the Hongyanhe Nuclear Power Plant, the major features are as: the areas affected by the floods in the grid system for emergency treatment at the nuclear power station were: grid area718, 758, 759, 798, 799, and 839;the most affected grid areas were grid areas 758–759 and 798–799; these were the important areas, incorporating the storage area for wastewater, waste liquid and waste solids, the comprehensive office building, and No. 6 shelter. The study also reveals that the success rate of UAV rescue using a single drone was 59.8% (standard deviation 0.1657, standard error 0.0166, 95% Confidence Interval),and there was no significant difference in the results between the two groups(significance of the one-way analysis of variance P = 0.662 > 0.05); the success rate of UAV rescue using two drones was 58.8% (standard deviation 0.1572, standard error 0.0157, 95% Confidence Interval); the simulated time of completing the rescue task using a single drone was 5484.09 (standard deviation 2447.7519, standard error 244.7752, 95% Confidence Interval); and the simulated time of completing the rescue task using two drones was 4679.22 (standard deviation 3109.4676, standard error 310.9468, 95% Confidence Interval). Thus, the rescue efficiency was higher using two drones than using one drone (level of significance from the one-way analysis of variance P = 0.043 < 0.05).

Design of a novel fully-active PEMFC-Lithium battery hybrid power system based on two automatic ON/OFF switches for unmanned aerial vehicle applications

High-performance unmanned aerial vehicles (UAVs) have become increasingly popular in applications across industrial inspection, forest patrol, and aerial photography. For high-performance UAVs, hybrid power systems consisting of proton exchange membrane fuel cells (PEMFCs) and Lithium batteries are usually recommended for improving their flight endurance. In this paper, a novel fully-active PEMFC-Lithium battery hybrid power system is designed for UAV applications. The proposed construction employs an automatic ON/OFF switch in parallel with a DC-DC converter to control the PEMFC and a second automatic ON/OFF switch replacing the commonly used DC-DC converter to control the Lithium battery. With this design, we enable the two energy sources to directly supply power through control of the ON/OFF state of the two automatic switches, thereby significantly avoiding energy loss in the DC-DC converter. The design further allows for separate or joint operation modes of the PEMFC-Lithium battery hybrid power system to be actively selected through combined control of the automatic ON/OFF switches and the DC-DC converter. On this basis, a simple and effective rule-based strategy can be designed for the energy management between the PEMFC and Lithium battery. Experimental results successfully demonstrate the feasibility of the proposed fully-active PEMFC-Lithium battery hybrid power system, with numerous simulations results showing that the proposed fully-active hybrid power system reduces hydrogen consumption by 5.6 g and improves energy utilization by 7% in UAV applications when compared with the conventional fully-active hybrid power system under the same power requirement. Additionally, the active mode switch control can effectively alleviate voltage degradation of the PEMFC as well as improve the output performance of the Lithium battery, which would effectively extend the service-life of the hybrid power system. This paper successfully demonstrates the advantages of the proposed fully-active PEMFC-Lithium hybrid power system in UAV applications.

Binary Offset Carrier (BOC) and Binary Phase Shift Keying (BPSK) Modulation in Indoor Drones GNSS Recievers using Multipath Error Envelope MEE Technique

The applications of unmanned aerial vehicles (UAVs) have greatly eased the lives of human being. Due to the mass commerial production of UAVs (or Drones) and the support of the ongoing scientific research, it can now be used in disaster monitoring, surviellance, fire conrolling and many outdoor usages. However, the indoor applications have stricter requirements for the autonomous positioning capability of UAVs, requiring its positioning accuracy to be within the narrower areas and aisles. The Global Navigation Satellite System (GNSS) is currently the only guidance and autorecovery method that can be applied directly and consistently to UAV positioning. The current GPS L1C and/or Galileo E1 open services signals are being widely used for this purpose. Moreover, the Binary Offset Carrier (BOC) techniques have been recently adopted in Galileo and GPS BIII as one of the efficient mitigation methods to decrease the interference/multipath delays and to increase both the Position Accuracy and the immunity of GNSS interferences. This research aims to present a software method of assessing the improvements of the Accuracy, and yet the Availability, by producing the MEE for BOC technique. The used methodology is analyzing the theoretical equations behind the interfernece multipath error envelopes in both BPSK and the new BOC signals. The Results show a higher performance of BOC GNSS Recievers over BPSK ones, and less delay.

Application of Unmanned Aerial Vehicles in Precision Agriculture

Agriculture has long been an important part of human activities [...]

High-Power Microwave Pulse-Induced Failure on Unmanned Aerial Vehicle System

With the wide application of unmanned aerial vehicles (UAVs), more attention has been paid to the safe operation of UAVs in complex electromagnetic environments. In this article, we investigate the impact of high-power microwave (HPM) pulses on three types of typical commercial UAVs. We established the HPM irradiation experiment system first and obtained the responses of UAVs under the HPM attack with different electric field strengths. The flight logs and the damage analysis reports indicate that the damaged components of the type I UAV and the type II UAV are the rotor motor and the electronic speed controller. The experiment results prove that L-band narrowband HPM microwave has strong backdoor coupling effects on UAVs. The damage mechanism of UAVs by the HPM is studied by simulation. The simulation results show that an HPM pulse will cause a high-voltage pulse coupled to the cable connecting the electronic speed controller and the rotor motor, which caused the electronic speed controller to send out a false signal that caused the rotor motor speed to rise sharply, eventually making the rotor motor to burn out. Experiment results show that strengthening electromagnetic protection on the cable can effectively prevent HPM damage to UAVs. The simulation results match the experiment results and the damage analysis reports. Finally, some suggestions are provided to improve the UAV’s anti-HPM capability.

Stability Analysis of 3D UAV Gimbal using Flexible Body

Unmanned aerial vehicle (UAV) application in visual object tracking requires a gimbal to stabilize the camera in following the object’s movement. External disturbance and gimbal stability are the issues in this application. This study aims to analyze disturbance effects and study the stress and modal analysis on UAV gimbal using flexible body concepts. The three-dimensional (3D) gimbal is modeled using the RecurDyn software consisting of 3 arms and a camera. Each of the arms is connected using a revolute joint and a rotational force to represent a motor. The considered disturbances are step, pulse, ramp, and sine wave input. The PID controller is used to stabilize the gimbal arm from the gravity of the camera and external disturbance. The result shows that the PID controller is robust to step, pulse, and ramp disturbance, but not to the sin wave disturbance. In addition, the second arm of the gimbal is the most stressed component and is prone to vibration.

Unmanned Aerial Vehicles for Search and Rescue: A Survey

In recent years, unmanned aerial vehicles (UAVs) have gained popularity due to their flexibility, mobility, and accessibility in various fields, including search and rescue (SAR) operations. The use of UAVs in SAR can greatly enhance the task success rates in reaching inaccessible or dangerous areas, performing challenging operations, and providing real-time monitoring and modeling of the situation. This article aims to help readers understand the latest progress and trends in this field by synthesizing and organizing papers related to UAV search and rescue. An introduction to the various types and components of UAVs and their importance in SAR operations is settled first. Additionally, we present a comprehensive review of sensor integrations in UAVs for SAR operations, highlighting their roles in target perception, localization, and identification. Furthermore, we elaborate on the various applications of UAVs in SAR, including on-site monitoring and modeling, perception and localization of targets, and SAR operations such as task assignment, path planning, and collision avoidance. We compare different approaches and methodologies used in different studies, assess the strengths and weaknesses of various approaches, and provide insights on addressing the research questions relating to specific UAV operations in SAR. Overall, this article presents a comprehensive overview of the significant role of UAVs in SAR operations. It emphasizes the vital contributions of drones in enhancing mission success rates, augmenting situational awareness, and facilitating efficient and effective SAR activities. Additionally, the article discusses potential avenues for enhancing the performance of UAVs in SAR.

A hybrid fast inference approach with distributed neural networks for edge computing enabled UAV swarm

Nowadays, unmanned aerial vehicle (UAV) swarm supported by mobile edge computing is attracting more and more attention, such as smart agriculture, smart transportation, smart security surveillance, and smart environmental monitoring. Nevertheless, small UAV devices cannot deploy advanced best-performing DNN models with insufficient computing and storage resources. Aerial computing centers serving as edges can collect tasks from UAVs, reduce UAV load, and provide high-accuracy inference. Limited radio resources in offloading and load imbalance in computing centers could lead to significant latency in task inference. Considering image processing in UAV applications, we propose a hybrid inference framework combining early exit and task offloading based on distributed neural network. The bottleneck-designed DNN provides a smaller intermediate data size than the original input. By deploying shallow neural networks at the drone terminals and advanced DNNs at the aerial computing servers, latency-constrained tasks can be exited after local inference, offloaded to the edge after local computation, or directly offloaded to the edge. Compared to binary or partial offloading, the early-exit-based framework provides fast inference as the workflow stops when a confident result is obtained. Combining task offloading into the inference process offers a more flexible option to release the UAV computing load. The proposed approach efficiently utilizes wireless transmission and balances the computation load to maximize the system inference accuracy within the latency constraints. Practical task inference simulations demonstrate that our approach obtains higher system accuracy under different radio resources, task generation rates, and terminal computing capacities conditions.

Lightweight Small Target Detection Algorithm with Multi-Feature Fusion

Unmanned aerial vehicles (UAVs) are a highly sought-after technology with numerous applications in both military and non-military uses. The identification of targets is a crucial aspect of UAV applications, but there are challenges associated with complex detection models and difficulty in detecting small targets. To address these issues, this study proposes the lightweight L-YOLO algorithm for target detection tasks from a UAV perspective. The L-YOLO algorithm improves on YOLOv5 by improving the model’s detection performance for small targets while reducing the number of parameters and computational effort. The GhostNet module replaces the relevant convolution in the YOLOv5 model to create a lightweight model. The EIoU loss is used as the loss function of the algorithm to accelerate convergence and improve regression accuracy. Furthermore, feature-level extensions based on YOLOv5 are implemented, and a new detection head is proposed to improve the model’s detection accuracy for small targets. The size of the anchor boxes is redesigned to suit the small targets using the K-means++ clustering algorithm. The experiments were conducted on the VisDrone-2022 dataset, and the L-YOLO algorithm demonstrated a reduction in computational effort by 42.42% and number of parameters by 48.6% compared to the original algorithm. Furthermore, recall and mAP@0.5 improved by 2.1% and 1.4%, respectively. These results demonstrate that the L-YOLO algorithm not only has better detection performance for small targets but is also a lighter model, indicating promising prospects for target detection from a UAV perspective.