UAV Safety Research Articles

Development and Experimental Validation of a Sense-and-Avoid System for a Mini-UAV

This paper provides an overview of the three-year effort to design and implement a prototypical sense-and-avoid (SAA) system based on a multisensory architecture leveraging data fusion between optical and radar sensors. The work was carried out within the context of the Italian research project named TERSA (electrical and radar technologies for remotely piloted aircraft systems) undertaken by the University of Pisa in collaboration with its industrial partners, aimed at the design and development of a series of innovative technologies for remotely piloted aircraft systems of small scale (MTOW < 25 Kgf). The system leverages advanced computer vision algorithms and an extended Kalman filter to enhance obstacle detection and tracking capabilities. The “Sense” module processes environmental data through a radar and an electro-optical sensor, while the “Avoid” module utilizes efficient geometric algorithms for collision prediction and evasive maneuver computation. A novel hardware-in-the-loop (HIL) simulation environment was developed and used for validation, enabling the evaluation of closed-loop real-time interaction between the “Sense” and “Avoid” subsystems. Extensive numerical simulations and a flight test campaign demonstrate the system’s effectiveness in real-time detection and the avoidance of non-cooperative obstacles, ensuring compliance with UAV aero mechanical and safety constraints in terms of minimum separation requirements. The novelty of this research lies in (1) the design of an innovative and efficient visual processing pipeline tailored for SWaP-constrained mini-UAVs, (2) the formulation an EKF-based data fusion strategy integrating optical data with a custom-built Doppler radar, and (3) the development of a unique HIL simulation environment with realistic scenery generation for comprehensive system evaluation. The findings underscore the potential for deploying such advanced SAA systems in tactical UAV operations, significantly contributing to the safety of flight in non-segregated airspaces

Diagnosability-optimization design of fixed-wing unmanned aerial vehicle based on causal matching and tree seed algorithm

With mission intensity and difficulty escalating, the future market demand for UAV reliability and safety grows dramatically. So, this paper proposes an optimal design strategy based on causal matching and the Tree Seed Algorithm (TSA) based on structural analysis. First, a novel algorithm is designed for finding the minimum set of consistency relations for diagnosability analysis. The complexity of this algorithm is at a polynomial level, which is a significant improvement over previous algorithms with exponential complexity. A causal consistency search algorithm is also innovatively proposed for causal diagnosability analysis considering the causal constraints of the dynamic variables. Secondly, a diagnosability optimization strategy based on TSA is designed to balance the diagnosability requirements and the design cost of consistency relations. This strategy can satisfy the system’s diagnosability demand under different causal constraints with minimum consistency relations. Finally, a fixed-wing UAV model is established to analyze the diagnosability under different causal constraints qualitatively. Based on the TSA, consistency relations with the minimum integrated diagnosis cost and the best diagnosis performance are preferred.

Autonomous UAV Safety Oriented Situation Monitoring and Evaluation System

In this paper, a LabVIEW-based online monitoring and safety evaluation system for UAVs is designed to address the deficiencies in UAV flight state parameter monitoring and safety evaluation. The system consists of a lower unit for UAV recording and an upper unit on the ground. The lower unit collects and detects flight data and connects to the upper unit through a wireless digital transmission module via a serial port. The upper unit receives the data and carries out the monitoring and safety situation evaluation of the UAV. The lower unit of the system adopts multi-sensors to collect UAV navigation information in real time to achieve flight detection, while the upper unit adopts LabVIEW to design the UAV online monitoring and safety situation prediction system, enabling monitoring and safety situation prediction during UAV navigation. The test results show that the system can detect and comprehensively display the navigation information of the UAV in real time, and realize the safety evaluation and warning function of the UAV.

Trajectory Interception Classification for Prediction of Collision Scope between Moving Objects

In the fields of autonomous navigation and vehicle safety, accurately predicting potential collision field points between moving objects is a significant challenge. A novel computing technique to enhance trajectory interception analysis is presented in this paper. Our objective is to develop a field model that can accurately forecast collision zones, improving road transportation safety and the use of autonomous cars. Our main contribution is a binary classification model called PCSMO (Prediction of Collision Scope between Moving Objects), which is based on zero-shot learning. Gann angles, which are typically 45 degrees, are used to analyze the trajectories of moving objects. This method is inspired by GANN (Gann Angle Numeric Nomenclature). Compared to earlier techniques, this model more accurately identifies potential collision collision interception zones. The technique computes Gann angles for trajectory analysis and extracts GPS coordinates of moving objects from video data using OpenCV. It offers a more sophisticated comprehension of object movement patterns and points of interception. To assess the precision, recall, F1-score, and prediction accuracy of our model, we employ 10-fold cross-validation. Comparing the PCSMO model to existing models, these metrics demonstrate how well the PCSMO model predicts potential collision zones. Our approach, we discovered, enhances trajectory analysis—a critical component of safer autonomous navigation systems. With potential applications in autonomous vehicle and UAV safety, the PCSMO model improves field interception classification.

Stable flight control of UAV under atmospheric turbulence

Unmanned aerial vehicles will be subjected to various environmental disturbances such as atmospheric turbulence during the execution of tasks, which will seriously affect the flight safety of UAV. By establishing UAV flight dynamics model and atmospheric turbulence model, and using PID algorithm to design controller, this paper studies the attitude and altitude control effect of UAV under atmospheric turbulence disturbance. The simulation results show that PID algorithm can effectively control the UAV's stable flight under atmospheric turbulence and ensure the safety of UAV.

A New Trajectory in UAV Safety: Leveraging Reinforcement Learning for Distance Maintenance Under Wind Variations

A New Trajectory in UAV Safety: Leveraging Reinforcement Learning for Distance Maintenance Under Wind Variations

Enhancing UAV Safety: Accurate Distance Measurement with YOLOV8-based Measuring Application

This article introduces a lightweight and efficient model for measuring applications, aimed at enhancing the current UAV monitoring system. The primary objective of this project is to develop a measuring application capable of determining and displaying the distance between the camera on the UAV and the facial model. The YOLOV8 framework is employed as a detection model to identify and interpret objects within the region of interest. Additionally, the algorithm incorporates the concept of focal length in lenses to calculate the distance between the facial expressions of a human face and the camera. To assess the algorithm’s accuracy, facial models were placed at various distances from the camera during testing. The predicted distance values obtained through the algorithm were then compared to the actual measured distances using a measuring tape. The results demonstrated a maximum tolerance of ±0.9 cm, indicating the algorithm’s reliable performance in predicting distance measurements.

A Real-Time Semantic Segmentation Method Based on STDC-CT for Recognizing UAV Emergency Landing Zones.

With the accelerated growth of the UAV industry, researchers are paying close attention to the flight safety of UAVs. When a UAV loses its GPS signal or encounters unusual conditions, it must perform an emergency landing. Therefore, real-time recognition of emergency landing zones on the ground is an important research topic. This paper employs a semantic segmentation approach for recognizing emergency landing zones. First, we created a dataset of UAV aerial images, denoted as UAV-City. A total of 600 UAV aerial images were densely annotated with 12 semantic categories. Given the complex backgrounds, diverse categories, and small UAV aerial image targets, we propose the STDC-CT real-time semantic segmentation network for UAV recognition of emergency landing zones. The STDC-CT network is composed of three branches: detail guidance, small object attention extractor, and multi-scale contextual information. The fusion of detailed and contextual information branches is guided by small object attention. We conducted extensive experiments on the UAV-City, Cityscapes, and UAVid datasets to demonstrate that the STDC-CT method is superior for attaining a balance between segmentation accuracy and inference speed. Our method improves the segmentation accuracy of small objects and achieves 76.5% mIoU on the Cityscapes test set at 122.6 FPS, 68.4% mIoU on the UAVid test set, and 67.3% mIoU on the UAV-City dataset at 196.8 FPS on an NVIDIA RTX 2080Ti GPU. Finally, we deployed the STDC-CT model on Jetson TX2 for testing in a real-world environment, attaining real-time semantic segmentation with an average inference speed of 58.32 ms per image.

Unmanned Aerial Vehicle Perspective Small Target Recognition Algorithm Based on Improved YOLOv5

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.

Design of Quadrotor UAV Based on MSP432

In order to solve the problems of UAV flexibility, stability and safety, this paper designs a quadrotor UAV based on MSP432. The UAV includes five parts: main control part, flight height measurement module, gyroscope, fixed point correction module and visual recognition module. In order to protect the main control and other parts, we designed the outer protective carbon plate, so that its security is greatly improved. Based on the in-depth study of the existing remote control flight function, an improved active disturbance rejection controller is added, and the dual control mode of autonomous control and manual control is adopted to prevent unnecessary losses due to operation errors. The combination of GPS and laser rangefinder allows for more precise fixed-point positioning and measurement of the altitude above the ground, allowing the operator to determine the position of the UAV so as to observe its status. The integration function and key technology of each subsystem of UAV are analyzed in detail by simulation software. The experimental results show that each part of the quadrotor UAV and the surrounding protective carbon plate can provide a systematic solution for the intelligent and systematic development of the UAV, further improve the stability, flexibility, safety and the ability of swarm intelligence of the UAV, and make it have a better performance in the flight process.

Fatigue detection method for UAV remote pilot based on multi feature fusion

<abstract> <p>In recent years, UAV industry is developing rapidly and vigorously. However, so far, there is no relevant research on the fatigue detection method for UAV remote pilot, which is the core technology to ensure the flight safety of UAV. Aiming at this problem, a fatigue detection method for UAV remote pilot is proposed in this paper. Specifically, we first build a UAV operator fatigue detection database (OFDD). By analyzing the fatigue features in the database, we find that multiple facial features are highly correlated to the fatigue state, especially the head posture, and the temporal information is essential for distinguish between yawn and speaking in the study of UAV remote pilot fatigue detection. Based on these findings, a fatigue detection method for UAV remote pilots was proposed by efficiently locating the related facial regions, a multiple features extraction module to extract the eye, mouth and head posture features, and an efficient temporal fatigue decision module based on SVM. The experimental results show that this method not only performs well on the traditional driver dataset, but also achieves an accuracy rate of 97.05%; and it achieves the highest detection accuracy rate of 97.32% on the UAV remote pilots fatigue detection dataset OFDD.</p> </abstract>

GENERALIZED MODEL OF THE ADS-B UNMANNED AERIAL VEHICLE DATA TRANSMISSION PROCESS IN A STEGANOGRAPHIC SYSTEM

The subject of the article is a model of the ADS-B data transmission process of an unmanned aerial vehicle in a steganographic system using direct spectrum expansion technology. The aim of the publication is to improve the security of unmanned aerial vehicles with an integrated ADS-B system. Particular scientific tasks: analysis of basic methods of ADS-B format data protection; development of the scheme of ADS-B drone data transfer in steganographic system with the use of direct spectrum expansion technology; improvement of the model of ADS-B drone data transfer in steganographic system with the use of direct spectrum expansion technology; determination of qualitative and quantitative characteristics as well as security properties of ADS-B format data. The following research results were obtained: as the result of the scientific works analysis the hypothesis about the perspectives of the ADS-B format steganographic data protection usage was put forward; the scheme of an unmanned aerial vehicle ADS-B data transmission in the steganographic system with the usage of the direct spectrum spreading technology was developed; the main safety properties as well as the safety parameters and characteristics of the ADS-B format information signal were formulated; the generalized model of an unmanned aerial vehicle ADS-B data transmission was further developed. This will improve the safety of UAVs. The advantages and disadvantages of the model were revealed, which allowed to determine the priority of further research and possible promising ways of solving the assigned tasks.

Data Generation for Modeling Attacks on UAVs for the Purpose of Testing Intrusion Detection Systems

Today, issues related to ensuring the safety of UAVs are very relevant. Researchers need to develop new protection methods to detect attacks in a timely manner and implement mitigation measures. The authors propose a new concept of attack detection "from inside" the UAV. The idea is to analyze the cyber-physical parameters of the UAV, which may indicate an attack, and its possible consequences. It was determined that to detect an attack and determine the consequences to which it can lead, it is necessary to control not only the initial parameters, but also the internal cyber-physical parameters of the UAV. This will allow predicting the possible consequences of an attack and taking emergency measures. A scheme of the impact of an attack on UAVs and the relationship with security incidents, built using an ontological approach, has been worked out. Two main essences of the UAV are considered - the physical and digital aspects of the UAV. Examples of chains of attacks leading to various consequences are also shown. In the review part, the analysis of methods and algorithms for detecting spoofing attacks using data generators is carried out, based on which conclusions are drawn about their advantages and disadvantages. Further, based on the experiments performed, the authors propose a method for assessing the quality of data and a method for generating anomalous data sets similar to real attack data, which can be used to develop and test methods for detecting and blocking attacks. The architecture of the experimental stand, which was used in the framework of full-scale simulation, is described. At this stand, designed to parse GPS spoofing attacks (GPS spoofing), several scenarios of a normal flight, and then several attack scenarios, were tested. Based on the results of the experiments, a method has been proposed that allows simulating the data corresponding to the attack with the required accuracy. A method for assessing the quality of fake data has also been proposed.

Research on UAV Safety Distance Warning Technology Based on Electric Field Ring Rate of Change

With the increasing scale of transmission lines, drone inspection has become a major method of transmission line inspection. In the process of UAV inspection, the safety distance warning of UAV and transmission lines is crucial, and the existing UAV safety distance warning methods have difficulties, such as high cost, low environmental adaptability, and data processing. In this paper, we propose a safety distance warning method based on the rate of change of electric field ring ratio, which can be applied to the inspection process of transmission lines of different voltage levels. In order to accurately measure the electric field strength around the transmission lines, a three-dimensional electric field measurement system for measuring the electric field in space is designed in this paper. Finally, a 10 kV three-phase transmission line experimental platform was built to verify the feasibility of the method proposed in this paper.

Research on Multi Machine Cooperative Autonomous Inspection Strategy for UHV Dense Transmission Channel Based on 5G Technology

With the continuous progress of UAV technology and 5G technology, the safety and reliability of UAV are gradually improved. In addition, UAV has many advantages, such as light weight, small size, low cost, and fast response in view of the low economy, high labor intensity, low efficiency, and other problems of the traditional inspection of UHV intensive transmission channels. This paper proposes a multi machine collaborative autonomous inspection scheme for UHV dense transmission channels based on 5G technology. This method is to realize the intelligent detection service for UHV dense transmission channels by building a self-service inspection model under the joint cooperation of 5G technology and UAV collaborative operation technology. The research of this paper shows that the multi machine coordinated autonomous inspection scheme of UHV dense transmission channel based on 5G technology constructed in this paper is not only more efficient but also safer than the traditional inspection scheme and single UAV inspection scheme. The related technologies can not only improve the maintenance efficiency of power grid dense transmission channel but also provide theoretical and technical reference for the improvement of multi machine coordinated technology.

Measuring the Tree Height of Picea crassifolia in Alpine Mountain Forests in Northwest China Based on UAV-LiDAR

Forests in alpine mountainous regions are sensitive to global climate change. Accurate measurement of tree height is essential for forest aboveground biomass estimation. Unmanned aerial vehicle light detection and ranging (UAV-LiDAR) in tree height estimation has been extensively used in forestry inventories. This study investigated the influence of varying flight heights and point cloud densities on the extraction of tree height, and four flight heights (i.e., 85, 115, 145, and 175 m) were set in three Picea crassifolia plots in the Qilian Mountains. After point cloud data were classified, tree height was extracted from a canopy height model (CHM) on the basis of the individual tree segmentation. Through comparison with ground measurements, the tree height estimations of different flight heights and point cloud densities were analyzed. The results indicated that (1) with a flight height of 85 m, the tree height estimation achieved the highest accuracy (R2 = 0.75, RMSE = 2.65), and the lowest accuracy occurred at a height of 175 m (R2 = 0.65, RMSE = 3.00). (2) The accuracy of the tree height estimation decreased as the point cloud density decreased. The accuracies of tree height estimation from low-point cloud density (R2 = 0.70, RMSE = 2.75) and medium density (R2 = 0.69, RMSE = 2.80) were comparable. (3) Tree height was slightly underestimated in most cases when CHM-based segmentation methods were used. Consequently, a flight height of 145 m was more applicable for maintaining tree height estimation accuracy and assuring the safety of UAVs flying in alpine mountain regions. A point cloud density of 125–185 pts/m2 can guarantee tree height estimation accuracy. The results of this study could potentially improve tree height estimation and provide available UAV-LiDAR flight parameters in alpine mountainous regions in Northwest China.

DOMA: Deep Smooth Trajectory Generation Learning for Real-Time UAV Motion Planning

In this paper, we present a Deep Reinforcement Learning (DRL) based real-time smooth UAV motion planning method for solving catastrophic flight trajectory oscillation issues. By formalizing the original problem as a linear mixture of dual-objective optimization, a novel Deep smOoth Motion plAnning (DOMA) algorithm is proposed, which adopts an alternative layer-by-layer gradient descending optimization approach with the major gradient and the DOMA gradient applied separately. Afterward, the mix weight coefficient between the two objectives is also optimized adaptively. Experimental result reveals that the proposed DOMA algorithm outperforms baseline DRL-based UAV motion planning algorithms in terms of both learning efficiency and flight motion smoothness. Furthermore, the UAV safety issue induced by trajectory oscillation is also addressed.

Automated Path Planning for unmanned aerial vehicle in Urban Dynamic Area

AbstractThis paper addresses the path planning and autonomous obstacle avoidance problem of UAVs in urban dynamic area. A flight path planning strategy for UAVs in complex urban environments is proposed.First, the A* algorithm is used to construct a desired global path in a 3D static environment, which is used as the static reference path for dynamic obstacle avoidance below.The environmental and the key points of algorithm are also elaborated. In this paper, the dynamic obstacles are divided into three categories, then, in order to avoid the collision between dynamic obstacles and static optimal paths, two strategies to achieve local online path adjustment are proposed. Finally, the seven order minimum snap trajectories generation based on piecewise polynomials is utilized to smooth the flight path.We can obtain a smooth trajectory based on UAV dynamics and safety. The simulation results verify the effectiveness of the proposed UAV path planning strategy in the dynamic and complex urban area.

An approach of mapping quarries in Vietnam using low-cost Unmanned Aerial Vehicles

In Vietnam, there are a huge number of quarries that areexploited and mainly provide materials to the construction sector of thecountry. However, most of the quarries are operating without topographicplans due to a lack of surveying activities. This paper introduces anapproach of using low-cost UAVs to produce digital surface models whichin turn are used to draw topographic maps of quarries in Vietnam. Forassessments of accuracy, safety, and working efficiency, four quarriesdifferent in terrain conditions, namely Luong Son, Long Son, Nui Nho, andNui Dai were selected as the study areas. Ground control points wereestablished in each area by using GNSS/RTK for camera calibration andaccuracy assessment. The accuracy of DSM was assessed using the rootmean-square error (RMSE) in X, Y, Z, XY, and XYZ components.Capturing images from each site were processed by usingAgisoft®PhotoScan Professional 1.5.2. The results showed that all theDSM models of the four areas have high accuracy, RMSE on the checkedGCPs ranges from 1.0 to 9.0 cm, from 1.2 to 5.0 cm, from 4.4 to 13.4 cm,from 1.6 to 10.3 cm, and from 4.9 to 16.9 cm for X, Y, Z, XY, XYZcomponents, respectively. We concluded that the low-cost UAV basedmapping technology can guarantee the accuracy of DSMs, the safety ofUAV flying, and the efficiency of surveying working simultaneously whenusing in quarries.

Sensors and Measurements for UAV Safety: An Overview.

Unmanned aerial vehicles’ (UAVs) safety has gained great research interest due to the increase in the number of UAVs in circulation and their applications, which has inevitably also led to an increase in the number of accidents in which these vehicles are involved. The paper presents a classification of UAV safety solutions that can be found in the scientific literature, putting in evidence the fundamental and critical role of sensors and measurements in the field. Proposals from research on each proposed class concerning flight test procedures, in-flight solutions including soft propeller use, fault and damage detection, collision avoidance and safe landing, as well as ground solution including testing and injury and damage quantification measurements are discussed.