Autonomous Unmanned Aerial Vehicles Research Articles

Enhancing vehicle detection in intelligent transportation systems via autonomous UAV platform and YOLOv8 integration

This study highlights the evolving landscape of object detection methodologies, emphasizing the superiority of deep learning-based approaches over traditional methods. Particularly in intelligent transportation systems-related applications requiring robust image processing techniques, such as vehicle identification, localization, tracking, and counting within traffic scenarios, deep learning has gained substantial traction. The YOLO algorithm, in its various iterations, has emerged as a popular choice for such tasks, with YOLOv5 garnering significant attention. However, a more recent iteration, YOLOv8, was introduced in early 2023, ushering in a new phase of exploration and potential innovation in the field of object detection. Consequently, due to its recent emergence, the number of studies on YOLOv8 is extremely limited, and an application in the field of Intelligent Transportation Systems (ITS) has not yet found its place in the existing literature. In light of this gap, this study makes a noteworthy contribution by delving into vehicle detection using the YOLOv8 algorithm. Specifically, the focus is on targeting aerial images acquired through a modified autonomous UAV, representing a unique avenue for the application of this cutting-edge algorithm in a practical context. The dataset employed for training and testing the algorithm was curated from a diverse collection of traffic images captured during UAV missions. In a strategic effort to enhance the variability of vehicle images, the study systematically manipulated flight patterns, altitudes, orientations, and camera angles through a custom-designed and programmed drone. This deliberate approach aimed to bolster the algorithm's adaptability across a wide spectrum of scenarios, ultimately enhancing its generalization capabilities. To evaluate the performance of the algorithm, a comprehensive comparative analysis was conducted, focusing on the YOLOv8n and YOLOv8x submodels within the YOLOv8 series. These submodels were subjected to rigorous testing across diverse lighting and environmental conditions using the dataset. Through tests, it was observed that YOLOv8n achieved an average precision of 0.83 and a recall of 0.79, whereas YOLOv8x attained an average precision of 0.96 and a recall of 0.89. Furthermore, YOLOv8x also outperformed YOLOv8n in terms of F1 score and mAP, achieving values of 0.87 and 0.83 respectively, compared to YOLOv8n's 0.81 and 0.79. These outcomes of the evaluation illuminated the relative strengths and weaknesses of YOLOv8n and YOLOv8x, leading to the conclusion that YOLOv8n is well-suited for real-time ITS applications, while YOLOv8x exhibits superior detection capabilities.

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.

Research on experimental evaluation model of unmanned aerial vehicle autonomy based on digital twin

Abstract The current focus of unmanned aerial vehicle (UAV) research lies in the scientific evaluation of UAV autonomy and the enhancement of belief in UAVs. Researchers are required to conduct numerous tests to assess the ability of UAVs to perform missions in complex environments. However, conducting real-world tests for UAVs costs a lot. Therefore, digital twin virtual simulation serves as a viable alternative to actual flight tests. Firstly, this paper establishes a digital twin framework that enables comprehensive perception and integration within a virtual environment, proposing a UAV autonomy experimental evaluation model. Additionally, an effective method for evaluating UAV autonomy is constructed, and test tasks targeting autonomous capability assessment are designed by using simulation software and verified through system examples. Finally, based on the simulation results, the autonomous capability of the UAV twin is evaluated and utilized as a reference for assessing the autonomy of physical UAVs, thereby providing support for scientific evaluation.

Flow Field Analysis on Critical Feedback Information for UAV Autonomous High Speed Descent in Power Line Scenario

Abstract To improve the UAV efficiency in power line inspection, it must autonomously and rapidly descend to a specific height to complete the charging or landing task. This task requires the UAV to contain a stable control loop to face the specific airflow during autonomous and rapid descent: the vortex ring problem, and this specific airflow will reduce the UAV controllability. To implement the UAV autonomous high-speed descent, a control method must provide feedback information on the affected airflow and output a stable rotational speed. Hence, this paper designs a UAV autonomous high-speed descent control loop and proposes a variable rotor speed method to solve the feedback information in the control loop. This method reveals the relationship between the air movement caused by the rotor rotation and the controllability of the UAV high-speed descent. Therefore, with this method, the UAV control loop can obtain stable feedback on the air movement and achieve autonomous high-speed descent. To solve the UAV feedback information, we built the model to calculate the rotor speed in the autonomous high-speed descent UAV control loop and verified it by the CFD. The verification results showed that the model could reflect the actual fly condition. The proposed method solves the problem of UAV autonomous high-speed descent feedback information, improving the work efficiency of UAVs in power line inspection operations.

INCEPTION SH: A NEW CNN MODEL BASED ON INCEPTION MODULE FOR CLASSIFYING SCENE IMAGES

In this study, a light-weight model with an optimum block structure that can be used in autonomous unmanned aerial vehicles (UAVs) was designed. The Inception SH model, which was developed based on the Inception V3 model, was compared on "Intel Image Dataset", a publicly available dataset in the literature. As a result of the comparison, values of 0.882, 0.883, 0.882 and 0.882 were obtained for the accuracy, precision, recall, and F1 score metrics for the Inception V3 model, respectively. In the Inception SH model, values of 0.958, 0.957, 0.974 and 0.967 were obtained for accuracy, precision, recall and F1 score metrics, respectively. As can be seen from these values, the proposed Inception SH model offers higher performance values than the underlying Inception V3 model. The Inception SH model was compared with different models in the literature using the same data set and was superior in accuracy, precision, recall and F1 score metrics compared to the compared models. According to the results obtained, it is predicted that the Inception SH model can be used as a lightweight model in various IoT devices, considering the popularity of autonomous UAVs.

Israeli Military Artificial Intelligence, Its Possible Use in the War in Gaza.

This paper is focused on examining selected Israeli technologies operating based on artificial intelligence and the possible use of these technologies by the Israeli Defense Forces in the war in Gaza. These technologies include the Besorah system, an AI technology for identifying and recommending targets suitable for aerial bombardment, as well as other forms of attacks carried out from a distance. The SMASH system is an automatic targeting and firing system for small arms. The Goshawk system is a fully autonomous UAV used for aerial protection. The IRIS robot can be used for the initial investigation of tunnels ahead of assaults. The alleged use of these technologies in the war in Gaza cannot be verified by independent sources, mainly due to the ongoing conflict. The paper strives to present a general theoretical assessment of these technologies and weapons from the perspective of international law and armed conflicts.

Path Planning for Unmanned Aerial Vehicles in Complex Environments

This paper introduces a comprehensive framework for generating obstacle-free flight paths for unmanned aerial vehicles (UAVs) in intricate 3D environments. The system leverages the Rapidly Exploring Random Tree (RRT) algorithm to design trajectories that effectively avoid collisions with structures of diverse shapes and sizes. Discussion revolves around the challenges encountered during development and the successful achievement of generating collision-free routes. While the system represents an initial iteration, it serves as a foundation for future projects aiming to refine and expand upon its capabilities. Future work includes simulation testing and integration into UAV missions for image acquisition and structure scanning. Additionally, considerations for swarm deployment and 3D reconstruction using various sensor combinations are outlined. This research contributes to the advancement of autonomous UAV navigation in real-world scenarios.

Autonomous UAV Navigation with Adaptive Control Based on Deep Reinforcement Learning

Unmanned aerial vehicle (UAV) navigation plays a crucial role in its ability to perform autonomous missions in complex environments. Most of the existing reinforcement learning methods to solve the UAV navigation problem fix the flight altitude and velocity, which largely reduces the difficulty of the algorithm. But the methods without adaptive control are not suitable in low-altitude environments with complex situations, generally suffering from weak obstacle avoidance. Some UAV navigation studies with adaptive flight only have weak obstacle avoidance capabilities. To address the problem of UAV navigation in low-altitude environments, we construct autonomous UAV navigation in 3D environments with adaptive control as a Markov decision process and propose a deep reinforcement learning algorithm. To solve the problem of weak obstacle avoidance, we creatively propose the guide attention method to make a UAV’s decision focus shift between the navigation task and obstacle avoidance task according to changes in the obstacle. We raise a novel velocity-constrained loss function and add it to the original actor loss to improve the UAV’s velocity control capability. Simulation experiment results demonstrate that our algorithm outperforms some of the state-of-the-art deep reinforcement learning algorithms performing UAV navigation tasks in a 3D environment and has outstanding performance in algorithm effectiveness, with the average reward increasing by 9.35, the success rate of navigation tasks increasing by 14%, and the collision rate decreasing by 14%.

Design of a UAV Trajectory Prediction System Based on Multi-Flight Modes

With the burgeoning impact of artificial intelligence on the traditional UAV industry, the pursuit of autonomous UAV flight has emerged as a focal point of contemporary research. Addressing the imperative for advancing critical technologies in autonomous flight, this paper delves into the realm of UAV flight state recognition and trajectory prediction. Presenting an innovative approach focused on improving the precision of unmanned aerial vehicle (UAV) path forecasting via the identification of flight states, this study demonstrates its efficacy through the implementation of two prediction models. Firstly, UAV flight data acquisition was realized in this paper by the use of multi-sensors. Finally, two models for UAV trajectory prediction were designed based on machine learning methods and classical mathematical prediction methods, respectively, and the results before and after flight pattern recognition are compared. The experimental results show that the prediction error of the UAV trajectory prediction method based on multiple flight modes is smaller than the traditional trajectory prediction method in different flight stages.

Fully Autonomous Trustworthy Unmanned Aerial Vehicle Teamwork: A Research Guideline Using Level 2 Blockchain

The vast range of possible fully autonomous multiunmanned aerial vehicle (multi-UAV) operations is creating a new and expanding market where technological advances are happening at a breakneck pace. The integration of UAVs in airspaces (not just for military purposes but also for civil, commercial, and leisure use) is essential in realizing the potential of this growing industry. Furthermore, with the advent of 6G, such integration will be cost-effective and more flexible. However, to reach widespread adoption, new models focusing on the safety, efficiency, reliability, and privacy of fully autonomous multi-UAV operations, ensuring that the operation history is trustworthy and can be audited by the relevant stakeholders, need to be developed. Accordingly, this work presents a research guideline for fully autonomous trustworthy UAV teamwork through layer 2 blockchains that provide efficient, privacy-preserving, reliable, and secure multi-UAV service delivery. We show the implications of this approach for an aerial surveillance use case.

Autonomous UAV Maneuvering Decisions by Refining Opponent Strategies

Autonomous UAV Maneuvering Decisions by Refining Opponent Strategies

Vertical Takeoff and Landing for Distribution of Parcels to Hospitals: A Case Study about Industry 5.0 Application in Israel’s Healthcare Arena

To gain a sustained competitive advantage, organizations such as UPS, Fedex, Amazon, etc., began to seek for industry 5.0 innovative autonomous delivery options for the last mile. Autonomous unmanned aerial vehicles are a promising alternative for the logistics industry. The fact that drones are propelled by green renewable energy source fits the companies’ need to become sustainable, replacing their fuel truck fleets, especially for traveling to remote rural locations to deliver small packages, but a major obstacle is the necessity for charging stations which is well documented in the literature. Therefore, the current research embarks on devising a novel yet practical piece of technology adopting the simplicity approach of direct flights to destinations. The analysis showcases the application for a network of warehouses and hospitals in Israel while controlling costs. Given the products in the case study are medical, direct flight has the potential to save lives when every moment counts. Hydrogen cell technology allows long-range flying without refueling, and it is both vibration-free which is essential for sensitive medical equipment and environmentally friendly in terms of air pollution and silence in urban areas. Importantly, hydrogen cells are lighter, with higher energy density than batteries, which makes them ideal for drone usage to reduce weight, maintain a longer life, and enable faster charging, all of which minimize downtime. Also, hydrogen sourcing is low-cost and unlimited compared to lithium-ion material which needs to be mined. The case study investigates an Israeli entrepreneurial company, Gadfin, which builds a vertical takeoff-and-landing-type of drone with folded wings that enable higher speed for the delivery of refrigerated medical cargo, blood, organs for transplant, and more to hospitals in partnership with the Israeli medical logistic conglomerate, SAREL. An analysis of shipping optimization (concerning the number and type of drone) is conducted using a mixed-integer linear programming technique based on various types of constraints such as traveling distance, parcel weight, the amount of flight controllers and daily number of flights allowed in order to not overcrowd the airspace. Importantly, the discussion assesses the ecosystem’s variety of risks and commensurate safety mechanisms for advancing a newly shaped landscape of drones in an Israeli tight airspace to establish a network of national routes for drone traffic. The conclusion of this research cautions limitations to overcome as the utilization of drones expand and offers future research avenues.

Robust digital-twin airspace discretization and trajectory optimization for autonomous unmanned aerial vehicles

The infiltration of heterogenous fleets of autonomous Unmanned Aerial Vehicles (UAVs) in smart cities is leading to the consumerization of city air space which includes infrastructure creation of roads, traffic design, capacity estimation, and trajectory optimization. This study proposes a novel autonomous Advanced Aerial Mobility (AAM) logistical system for high density city centers. First, we propose a real-time 3D geospatial mining framework for LiDAR data to create a dynamically updated digital twin model. This enables the identification of viable airspace volumes in densely populated 3D environments based on the airspace policy/regulations. Second, we propose a robust city airspace dynamic 4D discretization method (Skyroutes) for autonomous UAVs to incorporate the underlying real-time constraints coupled with externalities, legal, and optimal UAV operation based on kinematics. An hourly trip generation model was applied to create 1138 trips in two scenarios comparing the cartesian discretization to our proposed algorithm. The results show that the AAM enables a precise airspace capacity/cost estimation, due to its detailed 3D generation capabilities. The AAM increased the airspace capacity by up to 10%, the generated UAV trajectories are 50% more energy efficient, and significantly safer.

Autonomous UAV forced landing site prediction using machine learning

Autonomous UAV forced landing site prediction using machine learning

A Robust Strategy for UAV Autonomous Landing on a Moving Platform under Partial Observability

A Robust Strategy for UAV Autonomous Landing on a Moving Platform under Partial Observability

Suboptimal Trajectory Planning Technique in Real UAV Scenarios with Partial Knowledge of the Environment

In recent years, the issue of trajectory planning for autonomous unmanned aerial vehicles (UAVs) has received significant attention due to the rising demand for these vehicles across various applications. Despite advancements, real-time trajectory planning remains computationally demanding, particularly with the inclusion of 3D localization using computer vision or advanced sensors. Consequently, much of the existing research focuses on semi-autonomous systems, which rely on ground assistance through the use of external sensors (motion capture systems) and remote computing power. This study addresses the challenge by proposing a fully autonomous trajectory planning solution. By introducing a real-time path planning algorithm based on the minimization of the snap, the optimal trajectory is dynamically recalculated as needed. Evaluation of the algorithm’s performance is conducted in an unknown real-world scenario, utilizing both simulations and experimental data. The algorithm was implemented in MATLAB and subsequently translated to C++ for onboard execution on the drone.

Improving local path planning for UAV flight in challenging environments by refining cost function weights

AbstractUnmanned Aerial Vehicles (UAV) constantly gain in versatility. However, more reliable path planning algorithms are required until full autonomous UAV operation is possible. This work investigates the algorithm 3DVFH* and analyses its dependency on its cost function weights in 2400 environments. The analysis shows that the 3DVFH* can find a suitable path in every environment. However, a particular type of environment requires a specific choice of cost function weights. For minimal failure, probability interdependencies between the weights of the cost function have to be considered. This dependency reduces the number of control parameters and simplifies the usage of the 3DVFH*. Weights for costs associated with vertical evasion (pitch cost) and vicinity to obstacles (obstacle cost) have the highest influence on the failure probability of the local path planner. Environments with mainly very tall buildings (like large American city centres) require a preference for horizontal avoidance manoeuvres (achieved with high pitch cost weights). In contrast, environments with medium-to-low buildings (like European city centres) benefit from vertical avoidance manoeuvres (achieved with low pitch cost weights). The cost of the vicinity to obstacles also plays an essential role and must be chosen adequately for the environment. Choosing these two weights ideal is sufficient to reduce the failure probability below 10%.

Deep-learning based autonomous-exploration for UAV navigation

Unmanned Aerial Vehicles (UAVs), functioning as aerial robots, undertake multifarious missions within intricate and uncharted environments, necessitating autonomous exploration for efficient UAV navigation. This paper introduces a Deep Learning-based Autonomous UAV Exploration method (DLAE), which distinguishes itself from traditional autonomous ground robot exploration by employing cameras instead of radar sensors. This difference propels us toward proposing a hybrid action space that integrates both positional and yaw actions, aimed at circumventing the limitations imposed by the Field of View (FOV). Furthermore, to enhance learning efficiency in autonomous UAV exploration, we design an invalid action masking scheme. Significantly, our contribution includes the development of an autoregressive network model alongside a distinct training procedure specifically tailored for autonomous UAV exploration. Comprehensive experiments conducted across a variety of test maps reveal that our proposed method surpasses existing learning-based approaches, delivering superior exploration efficiency and reduced decision-making time in comparison to conventional strategies. Moreover, physical simulations have corroborated the effectiveness of our method in practical scenarios, showcasing its potential for real-world applications.

Enhancing air pollution mapping with autonomous UAV networks for extended coverage and consistency

In the context of today's pressing air pollution challenges, accurate and consistent air pollution mapping plays a pivotal role in understanding pollutant distribution and identifying pollution sources. While current technologies, including sensors and unmanned aerial vehicles (UAVs), have begun to address this issue, the full potential of UAV-based solutions remains largely untapped. This pioneering numerical study demonstrates the effective feasibility of precise and consistent air pollution mapping in local areas that exceed the coverage capacity of a single UAV. The approach involves employing multiple UAVs, which requires rigorous mission planning encompassing various complex stages. These stages include subdividing the mapping area into manageable sub-areas, evaluating the technical capabilities of each UAV, assigning specific tasks to UAVs, and conducting individual mapping operations. By endowing UAVs with full autonomy, horizontal air pollution maps are generated across different layers within the designated area. This method's distinct advantage is its simultaneous acquisition of vertical profiles at all points within the study region, eliminating the need for additional efforts. Through strategic technical analysis, it was revealed that each UAV's mission coverage area could be expanded by over 30%, leading to more consistent air pollution mapping. Furthermore, this finding suggests a reduction of up to 25% in the total number of UAVs required for studies covering significantly larger areas.

Visual SLAM for Unmanned Aerial Vehicles: Localization and Perception.

Localization and perception play an important role as the basis of autonomous Unmanned Aerial Vehicle (UAV) applications, providing the internal state of movements and the external understanding of environments. Simultaneous Localization And Mapping (SLAM), one of the critical techniques for localization and perception, is facing technical upgrading, due to the development of embedded hardware, multi-sensor technology, and artificial intelligence. This survey aims at the development of visual SLAM and the basis of UAV applications. The solutions to critical problems for visual SLAM are shown by reviewing state-of-the-art and newly presented algorithms, providing the research progression and direction in three essential aspects: real-time performance, texture-less environments, and dynamic environments. Visual-inertial fusion and learning-based enhancement are discussed for UAV localization and perception to illustrate their role in UAV applications. Subsequently, the trend of UAV localization and perception is shown. The algorithm components, camera configuration, and data processing methods are also introduced to give comprehensive preliminaries. In this paper, we provide coverage of visual SLAM and its related technologies over the past decade, with a specific focus on their applications in autonomous UAV applications. We summarize the current research, reveal potential problems, and outline future trends from academic and engineering perspectives.