Search Results (175)

Unmanned platforms such as unmanned aerial vehicles, unmanned ground vehicles, and autonomous underwater vehicles often face challenges of data, device, and model heterogeneity when performing collaborative data processing tasks. Existing research does not simultaneously address issues from these three aspects. To address this issue, this study designs an unmanned platform cluster architecture inspired by the cloud-edge-end model. This architecture integrates federated learning for privacy protection, leverages the advantages of distributed model training, and utilizes edge computing’s near-source data processing capabilities. Additionally, this paper proposes a federated edge intelligence method (DSIA-FEI), which comprises two key components. Based on traditional federated learning, a data sharing mechanism is introduced, in which data is extracted from edge-side platforms and placed into a data sharing platform to form a public dataset. At the beginning of model training, random sampling is conducted from the public dataset and distributed to each unmanned platform, so as to mitigate the impact of data distribution heterogeneity and class imbalance during collaborative data processing in unmanned platforms. Moreover, an intelligent model aggregation strategy based on similarity measurement and loss gradient is developed. This strategy maps heterogeneous model parameters to a unified space via hierarchical parameter alignment, and evaluates the similarity between local and global models of edge devices in real-time, along with the loss gradient, to select the optimal model for global aggregation, reducing the influence of device and model heterogeneity on cooperative learning of unmanned platform swarms. This study carried out extensive validation on multiple datasets, and the experimental results showed that the accuracy of the DSIA-FEI proposed in this paper reaches 0.91, 0.91, 0.88, and 0.87 on the FEMNIST, FEAIR, EuroSAT, and RSSCN7 datasets, respectively, which is more than 10% higher than the baseline method. In addition, the number of communication rounds is reduced by more than 40%, which is better than the existing mainstream methods, and the effectiveness of the proposed method is verified. Full article

Visual inspection of civil infrastructure for structural health assessment, as performed by structural engineers, is expensive and time-consuming. Therefore, automating this process is highly attractive, which has received significant attention in recent years. With the increasing capabilities of computers, deep neural networks have become a standard tool and can be used for structural health inspections. A key challenge, however, is the availability of reliable datasets. In this work, the U-net and DeepLab v3+ convolutional neural networks are trained on a synthetic Tokaido dataset. This dataset comprises images representative of data acquired by unmanned aerial vehicle (UAV) imagery and corresponding ground truth data. The data includes semantic segmentation masks for both categorizing structural elements (slabs, beams, and columns) and assessing structural damage (concrete spalling or exposed rebars). Data augmentation, including both image quality degradation (e.g., brightness modification, added noise) and image transformations (e.g., image flipping), is applied to the synthetic dataset. The selected neural network architectures achieve excellent performance, reaching values of 97% for accuracy and 87% for Mean Intersection over Union (mIoU) on the validation data. It also demonstrates promising results in the semantic segmentation of real-world structures captured in photographs, despite being trained solely on synthetic data. Additionally, based on the obtained results of semantic segmentation, it can be concluded that DeepLabV3+ outperforms U-net in structural component identification. However, this is not the case in the damage identification task. Full article

The translation of motion lift, as an important performance metric of a reversing six-link loader working device, is influenced by multiple factors, such as the mechanical structure, system components, and operational experience. To ensure that the loader’s motion lift performance is optimized, this paper takes the fork trajectory and the horizontal angle between the bucket cylinder and the ground as the main optimization objectives. Kinematic modeling and multi-objective optimization are conducted to reduce the influence of external factors on the motion lift process. Firstly, a parametric model of the reversing six-link mechanism is established based on its geometric and symmetric characteristics, and the expressions for the fork’s motion trajectory and the cylinder–ground angle are derived. Then, an optimization model is constructed with the aim of minimizing both the translational error during fork lifting and the horizontal angle of the bucket cylinder. An improved multi-objective genetic algorithm is employed for the global search and optimization. Inspired by the principle of symmetry, the algorithm incorporates a structured search strategy that enhances convergence efficiency and solution balance. A multi-criteria decision function is further applied to identify the optimal solution from the Pareto front. Finally, a real-vehicle experiment validates the optimization results. The findings confirm that the proposed method significantly improves the translational performance of the fork and effectively controls the horizontal angle of the cylinder while also enhancing the driver’s visibility and coordination of the entire system. These results provide a theoretical and engineering basis for the symmetry-informed multi-objective performance optimization of loader working devices. Full article

In this study, we model a solution surface, with each point having nine components using artificial intelligence (AI), to optimise the effects of ground vibration during blasting operations in an open-pit diamond mine. This model has eight input parameters that can be adjusted by blasting engineers to arrive at a desired output value of ground vibration. It is built using the best performing artificial neural network architecture that best fits the blasting data from 100 blasting events provided by the Debswana diamond mine. Other AI algorithms used to compare the model’s performance were the k-nearest neighbour, support vector machine, and random forest—together with more traditional statistical approaches, i.e., multivariate and regression analyses. The input parameters were burden, spacing, stemming length, hole depth, hole diameter, distance from the blast face to the monitoring point, maximum charge per delay, and powder factor. The optimised model allows for variations in the input values, given the constraints, such that the output ground vibration will be within the minimum acceptable value. Through unconstrained optimisation, the minimum value of ground vibration is around 0.1 mm/s, which is within the vibration range caused by a passing vehicle. Full article

This paper proposes a distance estimation error reduction framework to improve ground node localization accuracy in urban environments using an unmanned aerial vehicle (UAV) and path loss measurements. The primary goal of the framework is to bound distance estimation errors arising from inherent inaccuracies in path loss measurements. A k-means clustering algorithm is first applied to identify the region in which the ground node is located. Then, an analytical approach is used to select UAV waypoints. Moreover, a mean-based exponential smoothing approach is employed to refine the path loss measurements of the selected waypoints to mitigate the effects of multipath components that introduce significant errors in distance estimation. Finally, two estimators, maximum likelihood (ML)-based and semidefinite programming (SDP)-based relaxation, are employed to estimate the ground node’s location, validating the effectiveness of the proposed framework. Evaluations using ray tracing simulation data demonstrate a notable improvement in localization accuracy. The proposed framework effectively bounds the distance estimation errors and significantly reduces overall localization errors compared to conventional unbounded methods. Moreover, both estimators with the proposed framework achieve comparable localization accuracy, highlighting the framework’s capability to address key challenges in ML-based localization. Full article

As a key component of unmanned aerial vehicles (UAVs), the vibrational characteristics of the airframe critically impact flight safety and imaging quality. These vibrations, often generated by motor-propeller systems or aerodynamic forces, can lead to structural fatigue during flight or cause image blur in payloads like cameras. To analyze the dynamic performance of the six-rotor UAV frame, this paper develops a state-space model based on linear state-space theory, structural dynamics principles, and modal information. The Direct Current (DC) gain method is employed to reduce the number of modes, followed by frequency response analysis on the reduced modes to derive the frequency–domain transfer function between the excitation input and response output points. The contribution of each mode to the overall frequency response is evaluated, and the frequency response curve is subsequently plotted. The results indicate that the model achieves a 73-fold speed improvement with an error rate of less than 13%, thereby validating the accuracy of the six-rotor UAV frame state-space model. Furthermore, the computational efficiency has been significantly enhanced, meeting the requirements for vibration simulation analysis. The dynamic analysis approach grounded in state-space theory offers a novel methodology for investigating the dynamic performance of complex structures, enabling efficient and precise analysis of frequency response characteristics in complex linear systems such as electric vehicle (EV) battery modules and motor systems. By treating EV components as dynamic systems with coupled mechanical–electrical interactions, this method contributes to the reliability and safety of sustainable transportation systems, addressing vibration challenges in both UAVs and EVs through unified modeling principles. Full article

This paper addresses inherent limitations in unmanned aerial vehicle (UAV) undercarriages hindering vertical takeoff and landing (VTOL) capabilities on uneven slopes and obstacles. Robotic landing gear (RLG) designs have been proposed to address these limitations; however, existing designs are typically limited to ground slopes of 6–15°, beyond which rollover would happen. Moreover, articulated RLG concepts come with added complexity and weight penalties due to multiple drivetrain components. Previous research has highlighted that even a minor 3-degree slope change can increase the dynamic rollover risks by 40%. Therefore, the design optimization of robotic landing gear for enhanced VTOL capabilities requires a multidisciplinary framework that integrates static analysis, dynamic simulation, and control strategies for operations on complex terrain. This paper presents a novel, hybrid, compliant, belt-driven, three-legged RLG system, supported by a multidisciplinary design optimization (MDO) methodology, aimed at achieving enhanced VTOL capabilities on uneven surfaces and moving platforms like ship decks. The proposed system design utilizes compliant mechanisms featuring a series of three-flexure hinges (3SFH), to reduce the number of articulated drivetrain components and actuators. This results in a lower system weight, improved energy efficiency, and enhanced durability, compared to earlier fully actuated, articulated, four-legged, two-jointed designs. Additionally, the compliant belt-driven actuation mitigates issues such as backlash, wear, and high maintenance, while enabling smoother torque transfer and improved vibration damping relative to earlier three-legged cable-driven four-bar link RLG systems. The use of lightweight yet strong materials—aluminum and titanium—enables the legs to bend 19 and 26.57°, respectively, without failure. An animated simulation of full-contact landing tests, performed using a proportional-derivative (PD) controller and ship deck motion input, validate the performance of the design. Simulations are performed for a VTOL UAV, with two flexible legs made of aluminum, incorporating circular flexure hinges, and a passive third one positioned at the tail. The simulation results confirm stable landings with a 2 s settling time and only 2.29° of overshoot, well within the FAA-recommended maximum roll angle of 2.9°. Compared to the single-revolute (1R) model, the implementation of the optimal 3R Pseudo-Rigid-Body Model (PRBM) further improves accuracy by achieving a maximum tip deflection error of only 1.2%. It is anticipated that the proposed hybrid design would also offer improved durability and ease of maintenance, thereby enhancing functionality and safety in comparison with existing robotic landing gear systems. Full article

Tethered drones—defined in this work as multirotor aerial platforms physically connected to a ground station via a cable—have emerged as a transformative subclass of Tethered Unmanned Aerial Vehicles (TUAVs), offering enhanced power autonomy, communication robustness, and safety through a physical ground connection. This review provides a comprehensive analysis of the current state of tethered drone systems technology, focusing on critical system components such as power delivery, data transmission, tether management, and modeling frameworks. Emphasis is placed on the tether multifunctional role—not only as a physical link but also as a sensor, actuator, and communication channel—impacting both hardware design and control strategies. By consolidating fragmented research across disciplines, this work offers a unified reference for the design, implementation, and advancement of TUAV systems, with tethered drones as their principal application. Full article

The correlation between aerodynamic data obtained from ground and flight tests is crucial in developing aerospace vehicles. This paper proposes methods for modelling this correlation that combine feature extraction and symbolic regression. The neighborhood component analysis (NCA) method is utilized to extract features from the high-dimensional state space and then symbolic regression (SR) is applied to find the concise optimal expression. First, a simulation example of the NASA Twin Otter aircraft is used to validate the NCA and the SR tool developed by the research team in modeling the aerodynamic coefficient deviation between ground and flight due to an unpredictable inflight icing failure. Then, the method and tool are applied to real flight tests of two types of aerospace vehicles with different configurations. The final optimized mathematical models show that the two vehicles’ pitching moment coefficient deviations are related to the angle of attack (AOA) only. The mathematical model built using NCA and the SR tool demonstrates higher fitting accuracy and better generalization performance for flight test data than other typical data-driven methods. The mathematical model delivers a multi-fold enhancement in fitting accuracy over data-driven methods for all fight cases. For UAV flight test data, the average root mean square error (RMSE) of the mathematical model demonstrates a maximum improvement of 37% in accuracy compared to three data-driven methods. For XRLV flight test data, the prediction accuracy of the mathematical model shows an enhancement exceeding 80% relative to Gaussian kernel SVM and Gaussian process data-driven models. The research verifies the feasibility and effectiveness of the data feature extraction combined with the symbolic regression method in mining the correlation law between ground and flight deviations of aerodynamic characteristics. This study provides valuable insight for modeling problems with finite data samples and explicit physical meanings. Full article

The navigation of Unmanned Ground Vehicles (UGVs) in underground mining environments is critical for enhancing operational safety, efficiency, and automation in hazardous and constrained conditions. This paper presents a thorough review of path-planning algorithms employed for the navigation of UGVs in underground mines. It outlines the key components and requirements that are essential for an effective path planning framework, including sensors and the Robot Operating System (ROS). This review examines both global and local path-planning techniques, encompassing traditional graph-based methods, sampling-based approaches, nature-inspired algorithms, and reinforcement learning strategies. Through the analysis of the extant literature on the subject, this study highlights the strengths of the employed techniques, the application scenarios, the testing environments, and the optimization strategies. The most favorable and relevant algorithms, including A*, Rapidly-exploring Random Tree (RRT*), Dijkstra’s, Ant Colony Optimization (ACO), were identified. This paper acknowledges a significant limitation: the over-reliance on simulation testing for path-planning algorithms and the computational difficulties in implementing some of them in real mining conditions. It concludes by emphasizing the necessity for full-scale research on path planning in real mining conditions. Full article

Despite the growing availability of very-high-resolution (VHR) remote sensing imagery, extracting fine-grained urban features and materials remains a complex task. Land use/land cover (LULC) maps generated from satellite imagery often fall short in providing the resolution needed for detailed urban studies. While hyperspectral imagery offers rich spectral information ideal for material classification, its complex acquisition process limits its use on aerial platforms such as manned aircraft and unmanned aerial vehicles (UAVs), reducing its feasibility for large-scale urban mapping. This study explores the potential of using only RGB and LiDAR data from VHR aerial imagery as an alternative for urban material classification. We introduce an end-to-end workflow that leverages a multi-head segmentation network to jointly classify roof and ground materials while also segmenting individual roof components. The workflow includes a multi-offset self-ensemble inference strategy optimized for aerial data and a post-processing step based on digital elevation models (DEMs). In addition, we present a systematic method for extracting roof parts as polygons enriched with material attributes. The study is conducted on six cities in Flanders, Belgium, covering 18 material classes—including rare categories such as green roofs, wood, and glass. The results show a 9.88% improvement in mean intersection over union (mIOU) for building and ground segmentation, and a 3.66% increase in mIOU for material segmentation compared to a baseline pyramid attention network (PAN). These findings demonstrate the potential of RGB and LiDAR data for high-resolution material segmentation in urban analysis. Full article

Accurate evapotranspiration (ET) monitoring is important for making scientific irrigation decisions. Unmanned aerial vehicle (UAV) remote sensing platforms allow for the flexible and efficient acquisition of field data, providing a valuable approach for large-scale ET monitoring. This study aims to enhance the accuracy and reliability of ET estimation in rice paddies through two synergistic approaches: (1) integrating the energy flux diurnal variations into the Two-Source Energy Balance (TSEB) model, which considers the canopy and soil temperature components separately, for physical estimation and (2) optimizing the flight altitudes and observation times for thermal infrared (TIR) data acquisition to enhance the data quality. The results indicated that the energy flux in rice paddies followed a single-peak diurnal pattern dominated by net radiation (Rn). The diurnal variation in the ratio of soil heat flux (G) to Rn could be well fitted by the cosine function with a max value and peak time (R² > 0.90). The optimal flight altitude and time (50 m and 11:00 am) for improved identification of temperature differentiation between treatments were further obtained through cross-comparison. These adaptations enabled the TSEB model to achieve a satisfactory accuracy in estimating energy flux compared to the single-source SEBAL model, with R² values of 0.8501 for Rn − G and 0.7503 for latent heat (LE), as well as reduced rRMSE values. In conclusion, this study presents a reliable method for paddy field scale ET estimation based on a calibrated TSEB model. Moreover, the integration of ground and UAV multimodal data highlights its potential for precise irrigation practices and sustainable water resource management. Full article

The increasing number and complexity of cyber–physical systems in vehicles necessitate a rigorous approach to identifying functional safety and cybersecurity hazards during the concept phase of product development. This study establishes a systematic method for identifying safety and security requirements for E/E components in the automotive sector, utilizing the SysML language within the CAMEO environment. The method’s activities and work products are grounded in the ISO 26262:2018 and ISO/SAE 21434:2021 standards. Comprehensive requirements were defined for the method’s application environment and activities, including generic methods detailing the creation of work products. The method’s metamodel was developed using the MagicGrid framework and validated through an application example. Synergies between the two foundational standards were identified and integrated into the method. The solution generation was systematically described by detailing activities for result generation and the production of standard-compliant work products. To facilitate practical implementation, a method template in SysML was created, incorporating predefined stereotypes, relationships, and tables to streamline the application and enhance consistency. Full article

The tire grounding parameters are a crucial component of the vehicle dynamics control system; accurate acquisition of grounding parameters is important for improving traction, braking force, and handling stability during vehicle operation. This paper studies strain-based intelligent tire contact patch length and vertical force estimation; first, a 205/55R16 radial tire was established, and static grounding experiments were carried out to verify the validity of the finite element model. Second, the sensitivity of the circumferential strain signal of the inner liner in the contact area of a tire with complex tread patterns was discussed. Methods for estimating the contact angle and contact patch length of rolling tires were established, and the estimation accuracy under different tire parameters and operating conditions were analyzed. Finally, the vertical force-sensitive response characteristics were analyzed and extracted, and the vertical force prediction model of a radial tire based on particle swarm optimization BP neural network was established. Full article

Probabilistic random vibration can speed up wear and tear on several components of the tracked vehicle, including the track system, drivetrain, and suspension. Extended exposure to high levels of vibration can cause structural damage to the vehicle frame and other critical components. Assessing random vibration in track vehicles requires a comprehensive approach that considers both the root causes and potential consequences of the vibrations. This random vibration significantly influences the structural performance of suspension arm which is key component of tracked vehicle. Damage due to fatigue is conventionally computed using time domain loaded signals with stress or strain data. This approach generally holds good when loading is periodic in nature but not be a good choice when dynamic resonance is in process. In this case an alternative frequency domain fatigue life analysis is used where the random loads and responses are characterized using a concept called Power spectral density (PSD). The current research article investigates the fatigue damage characteristics of a tracked vehicle suspension arm considering the dynamic loads induced by traversing on smooth and rough terrain. The analysis focusses on assessing the damage and stress response Power spectral density (PSD) ground-based excitation which is termed PSD-G acceleration. Quasi Static Finite Element Method based approach is used to simulate the operational conditions experienced by the suspension arm. Through comprehensive numerical simulations, the fatigue damage accumulation patterns are examined, providing insights into the structure integrity and performance durability of the suspension arm under varying operational scenarios. The obtained stress response PSD data and fatigue damage showed that the rough terrain response exhibits higher stresses in suspension arm. The accumulated stresses in case of rough terrain may prompt to brittle failure at specific critical locations. This research contributes to the advancement to the design and optimization strategies for tracked vehicle components enhancing their reliability and longevity in demanding operational environments. Full article