Search Results (29)

In this paper, we propose a cooperative security method for unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) based on the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm to address the scenario of unauthorized rogue drones (RDs) intruding into an airport’s restricted airspace. The proposed method integrates artificial intelligence techniques with engineering solutions to enhance the autonomy and effectiveness of air–ground cooperation in airport security. Specifically, the MADDPG algorithm enables the Security Interception UAVs (SI-UAVs) to autonomously detect and counteract RDs by optimizing their decision-making processes in a multi-agent environment. Additionally, Particle Swarm Optimization (PSO) is employed for distance-based target assignment, allowing each SI-UAV to autonomously select intruder targets based on proximity. To address the challenge of limited SI-UAV flight range, a power replenishment mechanism is introduced, where each SI-UAV automatically returns to the nearest UGV for recharging after reaching a predetermined distance. Meanwhile, UGVs perform ground patrols across different airport critical zones (e.g., runways and terminal perimeters) according to pre-designed patrol paths. The simulation results demonstrate the feasibility and effectiveness of the proposed security strategy, showing improvements in the reward function and the number of successful interceptions. This approach effectively solves the problems of target allocation and limited SI-UAV range in multi-SI-UAV-to-multi-RD scenarios, further enhancing the autonomy and efficiency of air–ground cooperation in ensuring airport security. Full article

This work formulates and solves the actual problem of studying the logistics of unmanned aerial vehicle (UAV) operations in facility security planning. The study is related to security tasks, including perimeter control, infrastructure condition monitoring, prevention of unauthorized access, and analysis of potential threats. Thus, the topic of the proposed publication is relevant as it examines the sequence of logistical actions in the large-scale application of a swarm of drones for facility protection. The purpose of the research is to create a set of mathematical and simulation models that can be used to analyze the capabilities of a drone swarm when organizing security measures. The article analyzes modern problems of using a drone swarm: formation of the swarm, assessment of its potential capabilities, organization of patrols, development of monitoring scenarios, planning of drone routes and assessment of the effectiveness of the security system. Special attention is paid to the possibilities of wave patrols to provide continuous surveillance of the object. In order to form a drone swarm and possibly divide it into groups sent to different surveillance zones, the necessary UAV capacity to effectively perform security tasks is assessed. Possible security scenarios using drone waves are developed as follows: single patrolling with limited resources; two-wave patrolling; and multi-stage patrolling for complete coverage of the protected area with the required number of UAVs. To select priority monitoring areas, the functional potential of drones and current risks are taken into account. An optimization model of rational distribution of drones into groups to ensure effective control of the protected area is created. Possible variants of drone group formation are analyzed as follows: allocation of one priority surveillance zone, formation of a set of key zones, or even distribution of swarm resources along the entire perimeter. Possible scenarios for dividing the drone swarm in flight are developed as follows: dividing the swarm into groups at the launch stage, dividing the swarm at a given navigation point on the route, and repeatedly dividing the swarm at different patrol points. An original algorithm for the formation of drone flight routes for object surveillance based on the simulation modeling of the movement of virtual objects simulating drones has been developed. An agent-based model on the AnyLogic platform was created to study the logistics of security operations. The scientific novelty of the study is related to the actual task of forming possible strategies for using a swarm of drones to provide integrated security of objects, which contributes to improving the efficiency of security and monitoring systems. The results of the study can be used by specialists in security, logistics, infrastructure monitoring and other areas related to the use of drone swarms for effective control and protection of facilities. Full article

Traffic crashes and congestion generate high social and economic costs, yet traditional traffic monitoring methods, such as police patrols, fixed cameras, and helicopters, are costly, labor-intensive, and limited in spatial coverage. This paper presents a novel Drone Routing and Scheduling with Flexible Multiple Visits (DRSFMV) framework, an optimization model for planning drone-based highway monitoring under realistic operational constraints, including battery limits, variable monitoring durations, recharging at a depot, and target-specific inter-visit time limits. A mixed-integer nonlinear programming (MINLP) model and a linearized version (MILP) are presented to solve the problem. Due to the NP-hard nature of the underlying problem structure, a heuristic solver, Hexaly, is also used. A case study using real traffic census data from three Southern California counties tests the models across various network sizes and configurations. The MILP solves small and medium instances efficiently, and Hexaly produces high-quality solutions for large-scale networks. Results show clear trade-offs between drone availability and time-slot flexibility, and demonstrate that stricter revisit constraints raise operational cost. Full article

We study the use of a homogeneous fleet of drones to design an unattended persistent drone-based patrolling system for vast circular areas. The drones follow flight missions supported by auxiliary on-ground charging stations, whose location and number must be determined. To this end, we first present a mixed integer non-linear programming model for defining cyclic schedules of drone flights considering the selection of the drone model from a set of candidate drone platforms. By imposing a minimum acceptable time between consecutive visits to any perimeter point, the objective consists of minimizing the total surveillance system deployment cost. The solution provides the best platform, the location of base stations, and the number of drones needed to monitor the perimeter, as well as the flight mission for each drone. We test five commercial platforms in six different scenarios whose radios vary between 1196 and 1696 m. In five of them, the MD4-100 Microdrones model achieves the lower cost solution, with values of EUR 66,800 and 83,500 for Scenarios 1 and 2 and EUR 116,900 for Scenarios 3, 4 and 5, improving its rivals in average percentages that vary between 8.46% and 70.40%. In Scenario number 6, the lower cost solution is provided by the TARTOT-500 model, with a total cost of EUR 168,000, improving by 20% the solution provided by the MD4-100. After obtaining the optimal solution, to evaluate the system robustness, we propose a discrete event simulation model incorporating uncertain flight times taking into account the possibility of accelerated depletion of drones’ Lithium-Ion polymer (Li-Po) batteries. Overall, our research investigates how various factors—such as the number of drones in the fleet and the division of the perimeter into sectors—impact the reliability of the system. Using Scenario number 3, our tests demonstrate that under a risk of battery failures of 2.5% and three UAVs per station, the surveillance system reaches a global percentage of punctually patrolled sectors of 92.6% and limits the number of delayed sectors (the relay UAV reaches the perimeter slightly above the required time, but it positively re-establishes the cyclic pattern for patrolling) to only a 5.6%. Our findings provide valuable insights for designing more robust and cost-effective drone patrol systems capable of operating autonomously over large planning horizons. Full article

Dams in their natural environment will gradually develop cracks and other forms of damage. If not detected and repaired in time, the structural strength of the dam may be reduced, and it may even collapse. Repairing cracks and defects in dams is very important to ensure their normal operation. Traditional detection methods rely on manual inspection, which consumes a lot of time and labor, while deep learning methods can greatly alleviate this problem. However, previous studies have often focused on how to better detect crack defects, with the corresponding image resolution not being particularly high. In this study, targeting the scenario of real-time detection by drones, we propose an automatic detection method for dam crack targets directly on high-resolution remote sensing images. First, for high-resolution remote sensing images, we designed a sliding window processing method and proposed corresponding methods to eliminate redundant detection frames. Then, we introduced a Gaussian distribution in the loss function to calculate the similarity of predicted frames and incorporated a self-attention mechanism in the spatial pooling module to further enhance the detection performance of crack targets at various scales. Finally, we proposed a pruning-after-distillation scheme, using the compressed model as the student and the pre-compression model as the teacher and proposed a joint distillation method that allows more efficient distillation under this compression relationship between teacher and student models. Ultimately, a high-performance target detection model can be deployed in a more lightweight form for field operations such as UAV patrols. Experimental results show that our method achieves an mAP of 80.4%, with a parameter count of only 0.725 M, providing strong support for future tasks such as UAV field inspections. Full article

The emerging working mode of coordinated trucks and drones has demonstrated significant practical potential in various fields, including logistics and delivery, intelligence surveillance reconnaissance, area monitoring, and patrol. The seamless collaboration between trucks and drones is garnering widespread attention in academia and has emerged as a key technology for achieving efficient and secure transportation. This paper provides a comprehensive and in-depth review of the research status on the routing problem for coordinated trucks and drones, covering aspects such as application background, cooperative modes, configurations, issues that have been taken into consideration, and solution methodologies. Full article

Forest fires represent a significant menace to both the ecological equilibrium of forests and the safety of human life and property. Upon ignition, fires frequently generate billowing smoke. The prompt identification and management of fire sources and smoke can efficiently avert the occurrence of extensive forest fires, thereby safeguarding both forest resources and human well-being. Although drone patrols have emerged as a primary method for forest-fire prevention, the unique characteristics of forest-fire images captured from high altitudes present challenges. These include remote distances, small fire points, smoke targets with light hues, and complex, ever-changing background environments. Consequently, traditional target-detection networks frequently exhibit diminished accuracy when handling such images. In this study, we introduce a cutting-edge drone-based network designed for the detection of forest fires and smoke, named FSNet. To begin, FSNet employs the YOCO data-augmentation method to enhance image processing, thereby augmenting both local and overall diversity within forest-fire images. Next, building upon the transformer framework, we introduce the EBblock attention module. Within this module, we introduce the notion of “groups”, maximizing the utilization of the interplay between patch tokens and groups to compute the attention map. This approach facilitates the extraction of correlations among patch tokens, between patch tokens and groups, and among groups. This approach enables the comprehensive feature extraction of fire points and smoke within the image, minimizing background interference. Across the four stages of the EBblock, we leverage a feature pyramid to integrate the outputs from each stage, thereby mitigating the loss of small target features. Simultaneously, we introduce a tailored loss function, denoted as $L_{forest}$, specifically designed for FSNet. This ensures the model’s ability to learn effectively and produce high-quality prediction boxes. We assess the performance of the FSNet model across three publicly available forest-fire datasets, utilizing mAP, Recall, and FPS as evaluation metrics. The outcomes reveal that FSNet achieves remarkable results: on the Flame, Corsican, and D-Fire datasets, it attains mAP scores of 97.2%, 87.5%, and 94.3%, respectively, with Recall rates of 93.9%, 87.3%, and 90.8%, respectively, and FPS values of 91.2, 90.7, and 92.6, respectively. Furthermore, extensive comparative and ablation experiments validate the superior performance of the FSNet model. Full article

When working with objects on a smaller scale, higher detection accuracy and faster detection speed are desirable features. Researchers aim to endow drones with these attributes in order to improve performance when patrolling in controlled areas for object detection. In this paper, we propose an improved YOLOv7 model. By incorporating the variability attention module into the backbone network of the original model, the association between distant pixels is increased, resulting in more effective feature extraction and, thus, improved model detection accuracy. By improving the original network model with deformable convolution modules and depthwise separable convolution modules, the model enhances the semantic information extraction of small objects and reduces the number of model parameters to a certain extent. Pretraining and fine-tuning techniques are used for training, and the model is retrained on the VisDrone2019 dataset. Using the VisDrone2019 dataset, the improved model achieves an mAP₅₀ of 52.3% on the validation set. Through the visual comparative analysis of the detection results in our validation set, we find that the model shows a significant improvement in detecting small objects compared with previous iterations. Full article

Cities attract a large number of inhabitants due to their more advanced industrial and commercial sectors and more abundant and convenient living conditions. According to statistics, more than half of the world’s population resides in urban areas, contributing to the prosperity of cities. However, it also brings more crime risks to the city. Crime prediction based on spatiotemporal data, along with the implementation of multiple unmanned drone patrols and responses, can effectively reduce a city’s crime rate. This paper utilizes machine learning and data mining techniques, predicts crime incidents in small geographic areas with short timeframes, and proposes a random forest algorithm based on oversampling, which outperforms other prediction algorithms in terms of performance. The research results indicate that the random forest algorithm based on oversampling can effectively predict crimes with an accuracy rate of up to 95%, and an AUC value close to 0.99. Based on the crime prediction results, this paper proposes a multi-drone patrol response strategy to patrol and respond to predicted high-crime areas, which is based on target clustering and combined genetic algorithms. This strategy may help with the pre-warning patrol planning within an hourly range. This paper aims to combine crime event predictions with crowd-sourced cruise responses to proactively identify potential crimes, providing an effective solution to reduce urban crime rates. Full article

With the rapid development of smart grids, the deployment number of transmission lines has significantly increased, posing significant challenges to the detection and maintenance of power facilities. Unmanned aerial vehicles (UAVs) have become a common means of power inspection. In the context of drone power inspection, drone clusters are used as relays for long-distance communication to expand the communication range and achieve data transmission between patrol drones and base stations. Most of the communication occurs in the air-to-air channel between UAVs, which requires high reliability of communication between drone relays. Therefore, the main focus of this paper is on decoding schemes for drone air-to-air channels. Given the limited computing resources and battery capacity of a drone, as well as the large amount of power data that needs to be transmitted between drone relays, this paper aims to design a high-accuracy and low-complexity decoder for LDPC long-code decoding. We propose a novel shared-parameter neural-network-normalized minimum sum decoding algorithm based on codebook quantization, applying deep learning to traditional LDPC decoding methods. In order to achieve high decoding performance while reducing complexity, this scheme utilizes codebook-based weight quantization and parameter sharing methods to improve the neural-network-normalized minimum sum (NNMS) decoding algorithm. Simulation experimental results show that the proposed method has a better BER performance and low computational complexity. Therefore, the LDPC decoding algorithm designed effectively meets the drone characteristics and the high channel decoding performance requirements. This ensures efficient and reliable data transmission on the data link between drone relays. Full article

The effectiveness of human security-based guard patrol systems often faces challenges related to the consistency of perimeter checks regarding timing and patterns. Some solutions use autonomous drones for monitoring assistance but primarily optimize their camera-based object detection capabilities for favorable lighting conditions. This research introduces an innovative approach to address these limitations—a flying watchdog designed to augment patrol operations with predetermined flight patterns, enabling checkpoint identification and position verification through vision-based methods. The system has a laser-based data transmitter to relay real-time location and timing information to a receiver. The proposed system consists of drone and ground checkpoints with distinctive shapes and colored lights, further enhanced by solar panels serving as laser data receivers. The result demonstrates the drone’s ability to detect four white dot LEDs with square configurations at distances ranging from 18 to 20 m, even under deficient light conditions based on the OpenCV detection algorithm. Notably, the study underscores the significance of achieving an even distribution of light shapes to mitigate light scattering effects on readings while also confirming that ambient light levels up to a maximum of 390 Lux have no adverse impact on the performance of the sensing device. Full article

Transmission line fault detection using drones provides real-time assessment of the operational status of transmission equipment, and therefore it has immense importance in ensuring stable functioning of the transmission lines. Currently, identification of transmission line equipment relies predominantly on manual inspections that are susceptible to the influence of natural surroundings, resulting in sluggishness and a high rate of false detections. In view of this, in this study, we propose an insulator defect recognition algorithm based on a YOLOv5 model with a new lightweight network as the backbone network, combining noise reduction and target detection. First, we propose a new noise reduction algorithm, i.e., the adaptive neighborhood-weighted median filtering (NW-AMF) algorithm. This algorithm employs a weighted summation technique to determine the median value of the pixel point’s neighborhood, effectively filtering out noise from the captured aerial images. Consequently, this approach significantly mitigates the adverse effects of varying noise levels on target detection. Subsequently, the RepVGG lightweight network structure is improved to the newly proposed lightweight structure called RcpVGG-YOLOv5. This structure facilitates single-branch inference, multi-branch training, and branch normalization, thereby improving the quantization performance while simultaneously striking a balance between target detection accuracy and speed. Furthermore, we propose a new loss function, i.e., Focal EIOU, to replace the original CIOU loss function. This optimization incorporates a penalty on the edge length of the target frame, which improves the contribution of the high-quality target gradient. This modification effectively addresses the issue of imbalanced positive and negative samples for small targets, suppresses background positive samples, and ultimately enhances the accuracy of detection. Finally, to align more closely with real-world engineering applications, the dataset utilized in this study consists of machine patrol images captured by the Unmanned Aerial Systems (UAS) of the Yunnan Power Supply Bureau Company. The experimental findings demonstrate that the proposed algorithm yields notable improvements in accuracy and inference speed compared to YOLOv5s, YOLOv7, and YOLOv8. Specifically, the improved algorithm achieves a 3.7% increase in accuracy and a 48.2% enhancement in inference speed compared to those of YOLOv5s. Similarly, it achieves a 2.7% accuracy improvement and a 33.5% increase in inference speed compared to those of YOLOv7, as well as a 1.5% accuracy enhancement and a 13.1% improvement in inference speed compared to those of YOLOv8. These results validate the effectiveness of the proposed algorithm through ablation experiments. Consequently, the method presented in this paper exhibits practical applicability in the detection of aerial images of transmission lines within complex environments. In future research endeavors, it is recommended to continue collecting aerial images for continuous iterative training, to optimize the model further, and to conduct in-depth investigations into the challenges associated with detecting small targets. Such endeavors hold significant importance for the advancement of transmission line detection. Full article

The wave-impacted waters off exposed sandy beaches support marine megafauna, including dolphins, whales, sharks, rays and turtles. To characterise variation in megafaunal assemblages in this challenging habitat, we used drone-based remote sensing to survey marine megafauna off 23 beaches along 1050 km of the New South Wales (NSW, Australia) coast from 2017 to 2020. The surveys occurred from September to May and included 17,085 drone flights, with megafaunal abundances standardised by flight hours. In total, we identified 3838 individual animals from 16 taxa, although no megafauna was observed off 5 of the 23 beaches surveyed. Bottlenose dolphins were the most commonly sighted taxa and accounted for 82.3% of total megafaunal abundance. Cownose (6.7%) and eagle (3.4%) rays were the next most abundant taxa, with potentially dangerous sharks being rarely sighted (<1% of total megafauna). The megafaunal assemblages off wave-exposed beaches in northern NSW significantly differed from those in the central region, whereas the assemblages off the central region and southern NSW did not differ significantly. Wave exposure and water temperature were the best predictors of megafaunal assemblage structure. The richness of marine megafauna off ocean beaches was significantly greater in northern than southern NSW, and turtles were only observed off beaches in the northern region. However, variation in megafaunal richness, as well as the abundances of total megafauna, dolphins, rays, sharks and turtles were not significantly explained by water temperature, wave height, distance to estuary, or proximity to the nearest reef. Overall, drone-based surveys determined that megafaunal assemblages off wave-exposed beaches are characterised by sparse individuals or small groups of sharks, turtles and rays, punctuated by occasional large aggregations of dolphins, cownose rays and schooling sharks. The exception to this pattern was bottlenose dolphins, which routinely patrolled some beaches in northern NSW. Full article

Drones are becoming a greater threat to modern electrical grids with the capability to cause expensive and time-consuming damage repairs to substations and transmission lines. Consumer drones have the potential to cause harm at a low cost, and finding methods to counter these threats is becoming more crucial to keep grids secure. In 2021, there was an attempted attack on a substation with a consumer drone which highlighted the need for research in this area. Previously, there has been a large focus on counter drones around places such as airports; however, more focus is warranted to analyze drone impact on the grid infrastructure. Methods to counter drones’ harmful impacts vary from physical methods to using electromagnetic waves. This article looks to identify and propose potential applications for existing technologies, as well as developing anti-drone technologies. These methods have not been adopted yet; thus, there is a great opportunity to utilize these existing technologies to defend the grid. The methods investigated were surveillance cameras, patrolling drones, nets, signal jammers, and energy weapons. The existing technology is currently lacking in the area of drone defense and can be improved with existing studies. However, there is a need to identify those methods and find ways to apply them to the power grid. Different defending technologies vary concerning their potential implementation. This paper also identifies and categorizes different results these methods produce to counter drones and their associated costs. Full article

Small target detection has been widely used in applications that are relevant to everyday life and have many real-time requirements, such as road patrols and security surveillance. Although object detection methods based on deep learning have achieved great success in recent years, they are not effective in small target detection. In order to solve the problem of low recognition rate caused by factors such as low resolution of UAV viewpoint images and little valid information, this paper proposes an improved algorithm based on the YOLOv5s model, called YOLOv5s-pp. First, to better suppress interference from complex backgrounds and negative samples in images, we add a CA attention module, which can better focus on task-specific important channels while weakening the influence of irrelevant channels. Secondly, we improve the forward propagation and generalisation of the network using the Meta-ACON activation function, which adaptively learns to adjust the degree of linearity or nonlinearity of the activation function based on the input data. Again, the SPD Conv module is incorporated into the network model to address the problems of reduced learning efficiency and loss of fine-grained information due to cross-layer convolution in the model. Finally, the detection head is improved by using smaller, smaller-target detection heads to reduce missed detections. We evaluated the algorithm on the VisDrone2019-DET and UAVDT datasets and compared it with other state-of-the-art algorithms. Compared to YOLOv5s, mAP@.5 improved by 7.4% and 6.5% on the VisDrone2019-DET and UAVDT datasets, respectively, and compared to YOLOv8s, mAP@.5 improved by 0.8% and 2.1%, respectively. For improving the performance of the UAV-side small target detection algorithm, it will help to enhance the reliability and safety of UAVs in critical missions such as military reconnaissance, road patrol and security surveillance. Full article