Unmanned Aerial Vehicle Power Research Articles

UAV-Enabled Diverse Data Collection via Integrated Sensing and Communication Functions Based on Deep Reinforcement Learning

Unmanned aerial vehicles (UAVs) and drones are considered to represent a flexible mobile aerial platform to collect data in various applications. However, the existing data collection methods mainly consider uplink communication. The burgeoning development of integrated sensing and communication (ISAC) provides a new paradigm for data collection. A diverse data collection framework is established where the uplink communication and sensing functions are both considered, which can also be referred to as the uplink ISAC system. An optimization is formulated to minimize the data freshness indicator for communication and the detection freshness indicator for sensing by optimizing the UAV paths, the transmitted power of IoT devices and UAVs, and the transmission allocation indicators. Three state-of-the-art deep reinforcement learning (DRL) algorithms are utilized to solve this optimization. Experiments are conducted in both single-UAV and multi-UAV scenarios, and the results demonstrate the effectiveness of the proposed algorithms. In addition, the proposed algorithms outperform the benchmark in terms of accuracy and efficiency. Moreover, the effectiveness of the data collection mode with only communication or sensing functions is also verified. Also, the numerical Pareto front between communication and sensing performance is obtained by adjusting the importance parameter.

Optimization of combustion chamber geometry for a two-stroke spark-ignited direct-injection opposed-piston rod-less aviation kerosene engine

Spark-ignited direct-injection opposed-piston rod-less kerosene engines are highly suitable for small and medium-size unmanned aerial vehicles (UAVs) because of the high power-to-weight ratio, self-balancing ability, and safety of such engines. The design and optimization of combustion systems are crucial for power, fuel economy, and knock control. To meet the increasing demand for UAV power systems and fill the research void in this area, reported here for the first time are the design and optimization of a swirl-wall-guided combustion system including the spray pattern and pit structure on the top surface of the piston, with the dynamic, fuel economy, and knock intensity as the evaluation indicators. The results indicate that the spray pattern mainly affects the fuel escape rate and inhomogeneity. After it is injected into the pit, the fuel is guided to the combustion chamber by the steering role of the guide arc. If the spray drop point is located away from the pit, it leads to increased fuel escape rate and inhomogeneity. The turbulent kinetic energy in the spark-plug zone decreases as the pit diameter decreases, fluctuating by 48 %, while the average value remains relatively constant. The distribution is influenced primarily by the enveloping effect of the pit structure on the airflow. Under low-speed conditions, as the pit diameter decreases, the ignition delay period increases and then decreases, and the combustion duration decreases monotonically. The fluctuation of indicated power is relatively small (2.64 %), and the fuel escape rate decreases monotonically, resulting in the same variation pattern of indicated specific fuel consumption (ISFC). When the pit diameter is 43 mm, the fuel escape rate and ISFC reach their lowest values. Knock combustion near the wall is influenced mainly by two factors: the temperature rise rate and the flame propagation speed. The temperature rise rate increases significantly with decreasing pit diameter. When the pit diameter is less than 50 mm, knock combustion begins to occur. Under high-speed equivalent combustion conditions, as the pit diameter decreases, the indicated power changes by 5.21 %, and the ISFC decreases monotonically.

A deep reinforcement learning framework and its implementation for UAV-aided covert communication

In this work, we consider an Unmanned Aerial Vehicle (UAV)-aided covert transmission network, which adopts the uplink transmission of Communication Nodes (CNs) as a cover to facilitate covert transmission to a Primary Communication Node (PCN). Specifically, all nodes transmit to the UAV exploiting uplink non-Orthogonal Multiple Access (NOMA), while the UAV performs covert transmission to the PCN at the same frequency. To minimize the average age of covert information, we formulate a joint optimization problem of UAV trajectory and power allocation designing subject to multi-dimensional constraints including covertness demand, communication quality requirement, maximum flying speed, and the maximum available resources. To address this problem, we embed Signomial Programming (SP) into Deep Reinforcement Learning (DRL) and propose a DRL framework capable of handling the constrained Markov decision processes, named SP embedded Soft Actor-Critic (SSAC). By adopting SSAC, we achieve the joint optimization of UAV trajectory and power allocation. Our simulations show the optimized UAV trajectory and verify the superiority of SSAC compared with various existing baseline schemes. The results of this study suggest that by maintaining appropriate distances from both the PCN and CNs, one can effectively enhance the performance of covert communication by reducing the detection probability of the CNs.

The Progress and Trend of UAV Power Line Inspection: A Bibliometrics-based Visualization Analysis

<p> Background: The application of Unmanned Aerial Vehicle (UAV) is a major turning point in the history of power line inspection and is of increasing interest to a growing number of researchers. </p><p> Objective: This study aimed to clarify the research status, hotspots, and evolutionary trends of UAV power line inspection, to help researchers understand the dynamic evolution of research topics and provide guidance for future research directions. </p><p> Methods: 737 high-quality papers were collected from the Web of Science Core Collection (WoSCC) database from 2001 to 2023, and descriptive statistical analysis, cooperation network analysis, keyword co-occurrence analysis, keyword clustering analysis, and keyword citation burst analysis were conducted using VOSviewer and CiteSpace. </p><p> Results: The popularity of research in this field is increasing. Chinese academic institutions and scholars have made outstanding contributions. The level of cooperation between different countries, academic institutions, and scholars is low. The research on UAV power line inspection has a wide range of topics and obvious interdisciplinary characteristics, mainly focusing on fault detection and diagnosis, path planning, and the application of deep learning in intelligent inspection. The research hotspot and trend is the integration of artificial intelligence, deep learning, and modern information technology to achieve UAV autonomous intelligent inspection. Some of the challenges faced in the development of the field are also summarized and possible solutions are proposed. </p><p> Conclusion: This study provides a comprehensive and systematic review, and the results can provide a quick overview of the research status, hotspots, and evolutionary trends.

Fault diagnosis of drone motors driven by current signal data with few samples

Multi rotor unmanned aerial vehicles (UAVs) are extensively utilized across various domains, and the motor constitutes a pivotal element in the UAV power system. The majority of UAV failures and crashes stem from motor malfunctions, underscoring the imperative need for comprehensive research on fault diagnosis in UAV motors to ensure the stable and reliable execution of flight tasks. This study focuses on quadrotor UAVs as the research subject and devises targeted fault simulation experiments based on the structural features and operational characteristics of the DC brushless motor used in quadrotor UAVs, specifically examining the stator, rotor, and bearings. To address challenges related to the UAV’s own loads, limited space for redundant parts, and the high cost and difficulty associated with installing sensors for traditional fault diagnostic signals such as vibration and temperature, this study opts to use current signals as a substitute. This approach resolves the issue of challenging data collection for UAVs and investigates a current signal based fault diagnosis method for UAV motors. Lastly, in response to the limited training samples available for fault data due to the UAV’s highly sensitive characteristics regarding the health status of its components and flight stability, traditional machine learning and deep learning methods encounter difficulties in identifying representative features with a small number of training samples, leading to the risk of overfitting and reduced model accuracy in fault diagnosis. To overcome this challenge, we propose a hybrid neural network fault diagnosis model that incorporates a width learning system and a convolutional neural network (CNN). The width learning system eliminates temporal characteristics from the original current signal, capturing more comprehensive and representative sample features in the width feature space. Subsequently, the CNN is employed for feature extraction and classification tasks. In empirical small sample fault diagnosis experiments using current signal data for UAV motors, our proposed model outperforms other models used for comparison.

Task Offloading Strategy for Unmanned Aerial Vehicle Power Inspection Based on Deep Reinforcement Learning.

With the ongoing advancement of electric power Internet of Things (IoT), traditional power inspection methods face challenges such as low efficiency and high risk. Unmanned aerial vehicles (UAVs) have emerged as a more efficient solution for inspecting power facilities due to their high maneuverability, excellent line-of-sight communication capabilities, and strong adaptability. However, UAVs typically grapple with limited computational power and energy resources, which constrain their effectiveness in handling computationally intensive and latency-sensitive inspection tasks. In response to this issue, we propose a UAV task offloading strategy based on deep reinforcement learning (DRL), which is designed for power inspection scenarios consisting of mobile edge computing (MEC) servers and multiple UAVs. Firstly, we propose an innovative UAV-Edge server collaborative computing architecture to fully exploit the mobility of UAVs and the high-performance computing capabilities of MEC servers. Secondly, we established a computational model concerning energy consumption and task processing latency in the UAV power inspection system, enhancing our understanding of the trade-offs involved in UAV offloading strategies. Finally, we formalize the task offloading problem as a multi-objective optimization issue and simultaneously model it as a Markov Decision Process (MDP). Subsequently, we proposed a task offloading algorithm based on a Deep Deterministic Policy Gradient (OTDDPG) to obtain the optimal task offloading strategy for UAVs. The simulation results demonstrated that this approach outperforms baseline methods with significant improvements in task processing latency and energy consumption.

Decoupled Association With Rate Splitting Multiple Access in UAV-Assisted Cellular Networks Using Multi-Agent Deep Reinforcement Learning

In unmanned aerial vehicles (UAVs) assisted cellular networks, user association plays an important role in interference control and spectrum efficiency. In this paper, we study the performance of uplink-downlink decoupled (UDDe) user association in a multi-UAV assisted network in which each user can associate with different UAVs or the macro base station (MBS) for uplink (UL) and downlink (DL) transmissions. Since some popular data may be requested by multiple users, grouping these users and applying multicasting can significantly improve spectral efficiency. Unlike traditional linear precoding that treats interference entirely as noise, we propose a rate-splitting multiple access (RSMA) policy that employs rate splitting at the transmitter and successive interference cancellation (SIC) at the receiver. To be specific, the transmitted signal is split into a common part and a private part, and the interference is partially decoded and partially treated as noise. In this context, we formulate a joint optimization problem of UL-DL association and beamforming for maximizing the sum-rate of users in UL and that of multicast groups in DL under the constraints of UAV backhaul capacity and power budget. Since the formulated problem is non-convex with intricate states and an individual UAV may not know the rewards of other UAVs, we convert it into a robust partially observable Markov decision process (POMDP). Then we resort to multi-agent deep reinforcement learning (MADRL) that enables each UAV to learn and optimize its policy in a distributed manner. To achieve an optimal policy, we further propose an improved clip and count-based proximal policy optimization (PPO) algorithm to train actor and critic networks. Simulation results demonstrate the superiority of the proposed decoupled association strategy with RSMA and the MADRL learning algorithm.

3D Path Planning of the Solar Powered UAV in the Urban-Mountainous Environment with Multi-Objective and Multi-Constraint Based on the Enhanced Sparrow Search Algorithm Incorporating the Levy Flight Strategy

In response to practical application challenges in utilizing solar-powered unmanned aerial vehicle (UAV) for remote sensing, this study presents a three-dimensional path planning method tailored for urban-mountainous environment. Taking into account constraints related to the solar-powered UAV, terrain, and mission objectives, a multi-objective trajectory optimization model is transferred into a single-objective optimization problem with weight factors and multi-constraint and is developed with a focus on three key indicators: minimizing trajectory length, maximizing energy flow efficiency, and minimizing regional risk levels. Additionally, an enhanced sparrow search algorithm incorporating the Levy flight strategy (SSA-Levy) is introduced to address trajectory planning challenges in such complex environments. Through simulation, the proposed algorithm is compared with particle swarm optimization (PSO) and the regular sparrow search algorithm (SSA) across 17 standard test functions and a simplified simulation of urban-mountainous environments. The results of the simulation demonstrate the superior effectiveness of the designed improved SSA based on the Levy flight strategy for solving the established single-objective trajectory optimization model.

A dual-scale hybrid prediction model for UAV demand power: Based on VMD and SSA optimization algorithm

The prediction of power demand for unmanned aerial vehicles (UAV) is an essential basis to ensure the rational distribution of the energy system and stable economic flight. In order to accurately predict the demand power of oil-electric hybrid UAV, a method based on variational mode decomposition (VMD) and Sparrow Search Algorithm (SSA) is proposed to optimize the hybrid prediction model composed of long-short term memory (LSTM) and Least Squares Support Vector Machine (LSSVM). Firstly, perform VMD decomposition on the raw demand power data and use the sample entropy method to classify the feature-distinct mode components into high-frequency and low-frequency categories. Then, each modality component was separately input into the mixed model for rolling prediction. The LSSVM model and LSTM model were used to process low-frequency and high-frequency components, respectively. Finally, the predicted values for each modal component are linearly combined to obtain the final predicted value for power demand. Compared with the current models, the prediction model constructed in this paper stands out for its superior ability to track the changing trends of power demand and achieve the highest level of prediction accuracy.

Technical and economic feasibility of applying fuel cells as the power source of unmanned aerial vehicles

This study aims to quantitatively compare the weight, cost, and flight endurance of small unmanned aerial vehicles (UAVs) powered by batteries and fuel cells. We compare the use of lithium-polymer (LiPo) batteries, proton exchange membrane fuel cells (PEMFCs), and direct methanol fuel cells (DMFCs) in powering two representative small UAVs: KHawk-Thermal (fixed-wing, 1.4 m wingspan) and KHawk-CarbonQuad (quadcopter, 0.7 m diagonal length). This study uses experimentally measured polarization curves of fuel cells and power consumption data of LiPo-powered UAVs during flight tests as input to obtain the minimum mass, power consumption, and life cycle costs (LCC) of these power systems. The analyses consider the voltage, efficiency, and power at each current density to find the optimal operating current density to obtain the minimum weight and cost of fuel cell systems to obtain the desired flight endurance. The LiPo battery is the most cost-effective option to power UAVs with an endurance of 0.7 hrs (42 min) or less. At medium flight endurance (e.g., 0.8 – 2.1 hrs), the PEMFC system has the lowest mass and LCC. While hydrogen is the most energy-dense fuel, hydrogen storage significantly increases the mass and cost of the system. For UAVs undergoing long endurance missions (>2.2 hrs for fixed-wing UAVs, > 4.3 hrs for quadcopter UAVs), a DMFC system has the lowest mass and power consumption due to the high energy density of methanol and its simple storage requirements. This study also analyzes the impact of key parameters, such as the specific energy and power of battery and fuel cells, the mass ratio of hydrogen storage, and the lift-to-drag ratio, on UAVs' mass and power consumption.

SEMANTIC SEGMENTATION OF UAV LIDAR DATA FOR TREE PLANTATIONS

Abstract. Tree plantations, characterized by large-scale cultivation of trees with high commercial values, often rely on accurate inventory data to improve their capacity. However, understanding tree plantations with different components on a large scale for growth prediction is still a tricky problem. In this paper, we harness the power of Unmanned Aerial Vehicle (UAV) Light Detection and Ranging (LiDAR) systems to acquire 3D point clouds of tree plantations and investigate the potential of deep learning segmentation for enhanced understanding of plantation UAV LiDAR point clouds, thereby promoting precision forest management. Two datasets from the same plantation without debris on the ground and with harvested debris were tested. Experimental results showed that we were able to process a plantation consisting of 300 trees in 2 min and achieve an overall accuracy of 95% segmentation for this plantation. This research demonstrates the feasibility of the deep learning method in segmenting large-scale tree plantation point clouds, which is able to speed up the inventory of tree plantations.

Exploring the power of heterogeneous UAV swarms through reinforcement learning

The object of research is heterogeneous and homogeneous swarms of unmanned aerial vehicles (UAVs). The primary focus of this study is the comparison between heterogeneous and homogeneous UAV swarms, examining their performance in a simulated environment designed using the Python Gym library. The research involves implementing reinforcement learning algorithms, specifically the Proximal Policy Optimization (PPO), to train and evaluate the swarms. The central issue addressed by this research is to determine which type of UAV swarm – heterogeneous or homogeneous – exhibits better performance in a defined task. The chosen task involves searching for groups of objects in an unknown area, emphasizing the ability of the swarm to adapt and efficiently locate objects in dynamic environments. The obtained results reveal an advantage for heterogeneous UAV swarms over their homogeneous counterparts. The heterogeneous swarm has a steeper learning curve and achieves higher rewards in fewer episodes during the training phase. The key finding indicates that the varied skill set within the heterogeneous swarm allows for quicker adaptation to changing environmental conditions. The superior performance of the heterogeneous swarm is attributed to the diversity of capabilities among its UAV agents, enabling them to leverage their individual strengths to achieve better overall performance in the given task. The practical application of these results is contingent upon the task requirements and environmental conditions. In scenarios where tasks demand diverse skills and adaptability to changing conditions, heterogeneous UAV swarms are recommended. The results suggest their efficacy in applications such as search and rescue operations, environmental monitoring, and other dynamic tasks. In conclusion, this research provides valuable insights into optimizing UAV swarm composition for specific tasks. The results contribute both theoretically and practically by highlighting the advantages of heterogeneity in swarm capabilities.

Comparison of Aerodynamics and Flight Performance of Fixed Wing Solar Powered Unmanned Aerial Vehicle When Separate Lift and Thrust Vertical Take-Off and Landing System is Installed

The objective of this research paper is to analyze the effects of installing a separate Lift and Thrust vertical takeoff system for fixed-wing solar powered UAVs in terms of flight performance changes. This research article uses a prototype aircraft in this study, which is a fixed-wing aircraft powered by an electrical system and equipped with a solar system. There are 2 types of designs, namely the type with installation and without installation of the Vertical Take-Off and Landing system which the vertical take-off and landing aircraft compared to the normal configuration will weigh 42.48% more and have a zero-lift drag coefficient of 67.1% more. The initial step involves creating an aircraft model in the XFLR5 software to assess the lift and drag coefficients and compare them with the wind tunnel test results. Subsequently, the software analysis findings are used to evaluate the flight performance of both aircraft configuration, as well as to identify the performance changes that occur when the VTOL system is installed. The outcomes align with expectations, installing a VTOL system leads to a significant reduce in flight performance, In particular, the Endurance and maximum range of vertical take-off aircraft are only 41.55% and 43.43% of normal aircraft and for solar-powered systems. The vertical takeoff has the Endurance and range is 31.46% and 32.06% of normal aircraft equipped with solar systems. which can be distinctly contrasted with the performance of aircraft in various aspects. However, these compromises are made in exchange for the capability to take off and land vertically.

UAV-assisted finite block-length backscatter: Performance analysis and optimization

This paper introduces and investigates the performance of a system where an unmanned aerial vehicle (UAV) is utilized to assist terrestrial backscattering devices (BDs) in wireless energy charging and data transmission. The UAV task is data collection through the backscatter communication with finite block length. To assess performance, the study not only considers energy efficiency (EE) but also derives closed-form expressions for age-of-information (AoI). The block error rate (BLER) expression is approximated using Gaussian-Chebyshev quadrature and the first-order Riemann integral. Furthermore, the optimization problem is solved to maximize EE and minimize transmit power through a search algorithm. Both EE and spectral efficiency (SE) are investigate via the number of transmission bits. The analytical and simulation results show that the optimized EE and minimum transmit power of UAV can be obtained by selecting an appropriate packet length and UAV altitude. Besides, the Rician factor K also has a significant impact on BLER of the considered system. Finally, the study points out that the system performance is strongly influenced by factors such as the utilized bandwidth, the specific operating environments, and the elevation angle of UAV.

Queue-aware computation offloading for UAV-assisted edge computing in wind farm routine inspection

Integration of unmanned aerial vehicles (UAVs) and edge computing into the wind farm routine inspection provides a promising approach to enhancing inspection effectiveness and decreasing operation maintenance costs. In light of the finite battery power and computational capacity of UAVs, a dynamic queue-aware UAV-assisted edge computing inspection wind farm framework is investigated with the goal of minimizing the long-term energy consumption of UAVs. The Lyapunov optimization theory is utilized to decouple the long-term stochastic optimization problem into four short-term deterministic subproblems, including the task splitting, the UAV-side computing resource allocation, the task offloading, and the edge server-side computing resource allocation. Furthermore, a Lyapunov optimization-based dynamic queue-aware computation offloading algorithm (LODQCO) is presented to optimize task offloading and resource allocation jointly. The optimal UAV-side computing resource is determined by a closed form formula, and then the optimal task offloading decision is tackled by applying the classical interior point method. Finally, the edge server-side computing resource is addressed via a linear optimization CPLEX solver. Based on simulation results, LODQCO is superior to the benchmark algorithms with respect to the energy consumption, queue backlogs, and queuing delays.

Intelligent Resource Allocation Using an Artificial Ecosystem Optimizer with Deep Learning on UAV Networks

An Unmanned Aerial Vehicle (UAV)-based cellular network over a millimeter wave (mmWave) frequency band addresses the necessities of flexible coverage and high data rate in the next-generation network. But, the use of a wide range of antennas and higher propagation loss in mmWave networks results in high power utilization and UAVs are limited by low-capacity onboard batteries. To cut down the energy cost of UAV-aided mmWave networks, Energy Harvesting (EH) is a promising solution. But, it is a challenge to sustain strong connectivity in UAV-based terrestrial cellular networks due to the random nature of renewable energy. With this motivation, this article introduces an intelligent resource allocation using an artificial ecosystem optimizer with a deep learning (IRA-AEODL) technique on UAV networks. The presented IRA-AEODL technique aims to effectually allot the resources in wireless UAV networks. In this case, the IRA-AEODL technique focuses on the maximization of system utility over all users, combined user association, energy scheduling, and trajectory design. To optimally allocate the UAV policies, the stacked sparse autoencoder (SSAE) model is used in the UAV networks. For the hyperparameter tuning process, the AEO algorithm is used for enhancing the performance of the SSAE model. The experimental results of the IRA-AEODL technique are examined under different aspects and the outcomes stated the improved performance of the IRA-AEODL approach over recent state of art approaches.

UAV-Assisted Maritime Legitimate Surveillance: Joint Trajectory Design and Power Allocation

This paper investigates a novel maritime wireless surveillance scenario, where a legitimate monitor vessel moves around to eavesdrop the suspicious communication from a suspicious unmanned aerial vehicle (UAV) to a suspicious vessel with the help of a cooperative UAV. Specifically, the cooperative UAV can adjust its jamming power and trajectory to exactly control the transmission rate of the suspicious link, thus improving the monitor vessel's surveillance performance. Furthermore, since the cooperative UAV cannot land or replenish energy on the sea surface, its jamming power allocation on the ocean should be carefully designed by the energy thresholds. Under such setup, we formulate a sum eavesdropping rate maximization problem, which jointly optimizes the jamming power and three-dimensional (3D) trajectory of the cooperative UAV, as well as the two-dimensional (2D) trajectory of the monitor vessel. To address this non-convex optimization problem, we decompose the design problem into three subproblems and propose an iterative algorithm to find its suboptimal solution. Numerical results show that the proposed jamming-assisted 3D joint design can significantly improve the eavesdropping rate and save the jamming power compared to the benchmark schemes.

Channel Assignment and Power Allocation Utilizing NOMA in Long-Distance UAV Wireless Communication

Non-orthogonal multiple access (NOMA) is indeed an important technique of 5 G and beyond 5 G (B5G) communication systems, as it can increase the spectrum efficiency and system capacity under the same bandwidth. Long-distance unmanned aerial vehicle (UAV) networks, comprising a base station and multiple UAVs, are commonly used for reconnaissance purposes due to the extended flight range of the UAVs, which can provide broad coverage and satisfy the requirements for real-time communication. To alleviate spectrum underutilization of single channel orthogonal multiple access, we investigate resource allocation in UAV networks by utilizing NOMA and then formulate an optimization problem that maximizes the network capacity. The formulated problem can be decomposed into two sub-problems, namely UAV grouping and power allocation optimization. Accordingly, we present a UAV grouping method with low complexity according to the difference of channel gain between UAVs and base station, in which the UAVs with larger difference will be in the same cluster such that they can transmit signals over the same channel. Then, a power allocation optimization algorithm based on Lagrange multiplier method is then used to optimize the transmit power of UAVs utilizing the same channel. This algorithm considers the quality of service requirements for each UAV, and assigns power levels that maximize the overall network capacity. Simulation results have shown that the presented algorithms improve network capacity by approximately 19% compared with that of orthogonal multiple access, which demonstrates the effectiveness of utilizing NOMA in long-distance UAV networks.

Joint Trajectory Design and Power Allocation for UAV Assisted Network With User Mobility

This paper investigates the unmanned aerial vehicle (UAV) assisted wireless communications where the UAV acts as the aerial base station to provide downlink data service to the ground users. We consider a typical hot-spot scenario in which the ground users move towards the target points of interests (PoIs) in groups. The UAV can also fly in air to cope with the movement of users. To maximize the system throughput, an optimization problem is formulated to jointly determine the subchannel assignment, the trajectory of UAV and its power allocation per time slot, where the navigation power of UAV, the co-channel interference as well as the quality of service (QoS) requirements of users are taken into account. As the problem is non-convex and hard to solve, we propose a cascade approach. First, we design a user clustering and pairing algorithm to accomplish the subchannel assignment. Then, we apply the block coordinate descent (BCD) based framework to decompose the remaining problem into the power allocation sub-problem and the trajectory design one. The sub-optimal solution can be gradually achieved through alternatively solving these two sub-problems. As for the power allocation sub-problem, we opt to the difference of convex (DC) functions form and the successive convex approximation (SCA) method to solve it. While for the trajectory design sub-problem, the auxiliary variables and SCA method are utilized. Vivid simulation results reveal that the proposed iterative algorithm can well satisfy the QoS of users and achieve significant performance gain compared with other benchmark schemes.

Automated deep learning system for power line inspection image analysis and processing: Architecture and design issues

The continuous growth in the scale of unmanned aerial vehicle (UAV) applications in transmission line inspection has resulted in a corresponding increase in the demand for UAV inspection image processing. Owing to its excellent performance in computer vision, deep learning has been applied to UAV inspection image processing tasks such as power line identification and insulator defect detection. Despite their excellent performance, electric power UAV inspection image processing models based on deep learning face several problems such as a small application scope, the need for constant retraining and optimization, and high R&D monetary and time costs due to the black-box and scene data-driven characteristics of deep learning. In this study, an automated deep learning system for electric power UAV inspection image analysis and processing is proposed as a solution to the aforementioned problems. This system design is based on the three critical design principles of generalizability, extensibility, and automation. Pre-trained models, fine-tuning (downstream task adaptation), and automated machine learning, which are closely related to these design principles, are reviewed. In addition, an automated deep learning system architecture for electric power UAV inspection image analysis and processing is presented. A prototype system was constructed and experiments were conducted on the two electric power UAV inspection image analysis and processing tasks of insulator self-detonation and bird nest recognition. The models constructed using the prototype system achieved 91.36% and 86.13% mAP for insulator self-detonation and bird nest recognition, respectively. This demonstrates that the system design concept is reasonable and the system architecture feasible