Separated transmit and receive antennas are employed to improve transmit-receive isolation in conventional short-range radars, which greatly increases the antenna size and misaligns of the transmit/receive radiation patterns. In this paper, a dual circularly polarized (CP) monostatic simultaneous transmit and receive (MSTAR) antenna with enhanced isolation is proposed to alleviate the problem. The proposed antenna consists of one sequentially rotating array (SRA), two beamforming networks (BFN), and a combined decoupling structure. The SRA is shared by the transmit and receive to reduce the size of the antenna and to obtain a consistent transmit and receive pattern. The BFN achieve right-hand CP for transmit and left-hand CP for receive. By exploring the combined decoupling structure of uniplanar compact electromagnetic band gap (UC-EBG) and ring-shaped defected ground structure (RS-DGS), good transmit-receive isolation is achieved. The proposed antenna prototype is fabricated and experimentally characterized. The simulated and measured results show good agreement. The demonstrate transmit/receive isolation is height than 33 dB, voltage standing wave ratio is lower than 2, axial ratio is lower than 3 dB, and consistent radiation for both transmit and receive is within 4.25?4.35 GHz.

Failure mode and effect analysis (FMEA) is a preventative risk evaluation method used to evaluate and eliminate failure modes within a system. However, the traditional FMEA method exhibits many deficiencies that pose challenges in practical applications. To improve the conventional FMEA, many modified FMEA models have been suggested. However, the majority of them inadequately address consensus issues and focus on achieving a complete ranking of failure modes. In this research, we propose a new FMEA approach that integrates a two-stage consensus reaching model and a density peak clustering algorithm for the assessment and clustering of failure modes. Firstly, we employ the interval 2-tuple linguistic variables (I2TLVs) to express the uncertain risk evaluations provided by FMEA experts. Then, a two-stage consensus reaching model is adopted to enable FMEA experts to reach a consensus. Next, failure modes are categorized into several risk clusters using a density peak clustering algorithm. Finally, the proposed FMEA is illustrated by a case study of load-bearing guidance devices of subway systems. The results show that the proposed FMEA model can more easily to describe the uncertain risk information of failure modes by using the I2TLVs; the introduction of an endogenous feedback mechanism and an exogenous feedback mechanism can accelerate the process of consensus reaching; and the density peak clustering of failure modes successfully improves the practical applicability of FMEA.

Low sidelobe waveform can reduce mutual masking between targets and increase the detection probability of weak targets. A low sidelobe waveform design method based on complementary amplitude coding (CAC) is proposed in this paper, which can be used to reduce the sidelobe level of multiple waveforms. First, the CAC model is constructed. Then, the waveform design problem is transformed into a nonlinear optimization problem by constructing an objective function using the two indicators of peak-to-sidelobe ratio (PSLR) and integrated sidelobe ratio (ISLR). Finally, genetic algorithm (GA) is used to solve the optimization problem to get the best CAC waveforms. Simulations and experiments are conducted to verify the effectiveness of the proposed method.

In the field of calculating the attack area of air-to-air missiles in modern air combat scenarios, the limitations of existing research, including real-time calculation, accuracy efficiency trade-off, and the absence of the three-dimensional attack area model, restrict their practical applications. To address these issues, an improved backtracking algorithm is proposed to improve calculation efficiency. A significant reduction in solution time and maintenance of accuracy in the three-dimensional attack area are achieved by using the proposed algorithm. Furthermore, the age-layered population structure genetic programming (ALPS-GP ) algorithm is introduced to determine an analytical polynomial model of the three-dimensional attack area, considering real-time requirements. The accuracy of the polynomial model is enhanced through the coefficient correction using an improved gradient descent algorithm. The study reveals a remarkable combination of high accuracy and efficient real-time computation, with a mean error of 91.89 m using the analytical polynomial model of the three-dimensional attack area solved in just 10^?4 s, thus meeting the requirements of real-time combat scenarios.

The learning algorithms of causal discovery mainly include score-based methods and genetic algorithms (GA). The score-based algorithms are prone to searching space explosion. Classical GA is slow to converge, and prone to falling into local optima. To address these issues, an improved GA with domain knowledge (IGADK) is proposed. Firstly, domain knowledge is incorporated into the learning process of causality to construct a new fitness function. Secondly, a dynamical mutation operator is introduced in the algorithm to accelerate the convergence rate. Finally, an experiment is conducted on simulation data, which compares the classical GA with IGADK with domain knowledge of varying accuracy. The IGADK can greatly reduce the number of iterations, populations, and samples required for learning, which illustrates the efficiency and effectiveness of the proposed algorithm.

To meet the requirements of modern air combat, an integrated fire/flight control (IFFC) system is designed to achieve automatic precision tracking and aiming for armed helicopters and release the pilot from heavy target burden. Considering the complex dynamic characteristics and the couplings of armed helicopters, an improved automatic attack system is constructed to integrate the fire control system with the flight control system into a unit. To obtain the optimal command signals, the algorithm is investigated to solve nonconvex optimization problems by the contracting Broyden Fletcher Goldfarb Shanno (C-BFGS) algorithm combined with the trust region method. To address the uncertainties in the automatic attack system, the memory nominal distribution and Wasserstein distance are introduced to accurately characterize the uncertainties, and the dual solvable problem is analyzed by using the duality theory, conjugate function, and dual norm. Simulation results verify the practicality and validity of the proposed method in solving the IFFC problem on the premise of satisfactory aiming accuracy.

An improved estimation of distribution algorithm (IEDA) is proposed in this paper for efficient design of metamaterial absorbers. This algorithm establishes a probability model through the selected dominant groups and samples from the model to obtain the next generation, avoiding the problem of building-blocks destruction caused by crossover and mutation. Neighboring search from artificial bee colony algorithm (ABCA) is introduced to enhance the local optimization ability and improved to raise the speed of convergence. The probability model is modified by boundary correction and loss correction to enhance the robustness of the algorithm. The proposed IEDA is compared with other intelligent algorithms in relevant references. The results show that the proposed IEDA has faster convergence speed and stronger optimization ability, proving the feasibility and effectiveness of the algorithm.

Most of the existing non-line-of-sight (NLOS) localization methods depend on the layout information of the scene which is difficult to be obtained in advance in the practical application scenarios. To solve the problem, an NLOS target localization method in unknown L-shaped corridor based ultra-wideband (UWB) multiple-input multiple-output (MIMO) radar is proposed in this paper. Firstly, the multipath propagation model of L-shaped corridor is established. Then, the localization process is analyzed by the propagation characteristics of diffraction and reflection. Specifically, two different back-projection imaging processes are performed on the radar echo, and the positions of focus regions in the two images are extracted to generate candidate targets. Furthermore, the distances of propagation paths corresponding to each candidate target are calculated, and then the similarity between each candidate target and the target is evaluated by employing two matching factors. The locations of the targets and the width of the corridor are determined based on the matching rules. Finally, two experiments are carried out to demonstrate that the method can effectively obtain the target positions and unknown scene information even when partial paths are lost.

To solve the problem of providing the best initial situation for terminal guidance when multiple missiles intercept multiple targets, a group cooperative midcourse guidance law (GCMGL) considering time-to-go is proposed. Firstly, a three-dimensional (3D) guidance model is established and a cooperative trajectory shaping guidance law is given. Secondly, for estimating the unknown target maneuvering acceleration, an adaptive disturbance observer (ADO) is designed, combining finite-time theory with a radial basis function (RBF) neural network, and the convergence of the estimation error is proven using Lyapunov stability theory. Then, to ensure time-to-go cooperation among missiles within the same group and across different groups, the group consensus protocols of virtual collision point mean and the inter-group cooperative consensus protocol are designed respectively. Based on the group consensus protocols, the virtual collision point cooperative guidance law is given, and the finite-time convergence is proved by Lyapunov stability theory. Simultaneously, combined with trajectory shaping guidance law, virtual collision point cooperative guidance law and the inter-group cooperative consensus protocol, the design of GCMGL considering time-to-go is given. Finally, numerical simulation results show the effectiveness and the superiority of the proposed GCMGL.

The syndrome a posteriori probability of the log-likelihood ratio of intercepted codewords is used to develop an algorithm that recognizes the polar code length and generator matrix of the underlying polar code. Based on the encoding structure, three theorems are proved, two related to the relationship between the length and rate of the polar code, and one related to the relationship between frozen-bit positions, information-bit positions, and codewords. With these three theorems, polar codes can be quickly reconstruced. In addition, to detect the dual vectors of codewords, the statistical characteristics of the log-likelihood ratio are analyzed, and then the information- and frozen-bit positions are distinguished based on the minimum-error decision criterion. The bit rate is obtained. The correctness of the theorems and effectiveness of the proposed algorithm are validated through simulations. The proposed algorithm exhibits robustness to noise and a reasonable computational complexity.

Compared with single-domain unmanned swarms, cross-domain unmanned swarms continue to face new challenges in terms of platform performance and constraints. In this paper, a joint unmanned swarm target assignment and mission trajectory planning method is proposed to meet the requirements of cross-domain unmanned swarm mission planning. Firstly, the different performances of cross-domain heterogeneous platforms and mission requirements of targets are characterised by using a collection of operational resources. Secondly, an algorithmic framework for joint target assignment and mission trajectory planning is proposed, in which the initial planning of the trajectory is performed in the target assignment phase, while the trajectory is further optimised afterwards. Next, the estimation of the distribution algorithms is combined with the genetic algorithm to solve the objective function. Finally, the algorithm is numerically simulated by specific cases. Simulation results indicate that the proposed algorithm can perform effective task assignment and trajectory planning for cross-domain unmanned swarms. Furthermore, the solution performance of the hybrid estimation of distribution algorithm (EDA)-genetic algorithm (GA) algorithm is better than that of GA and EDA.

Multi-agent systems often require good interoperability in the process of completing their assigned tasks. This paper first models the static structure and dynamic behavior of multi-agent systems based on layered weighted scale-free community network and susceptible-infected-recovered (SIR) model. To solve the problem of difficulty in describing the changes in the structure and collaboration mode of the system under external factors, a two-dimensional Monte Carlo method and an improved dynamic Bayesian network are used to simulate the impact of external environmental factors on multi-agent systems. A collaborative information flow path optimization algorithm for agents under environmental factors is designed based on the Dijkstra algorithm. A method for evaluating system interoperability is designed based on simulation experiments, providing reference for the construction planning and optimization of organizational application of the system. Finally, the feasibility of the method is verified through case studies.

Nowadays, wireless communication devices turn out to be transportable owing to the execution of the current technologies. The antenna is the most important component deployed for communication purposes. The antenna plays an imperative role in receiving and transmitting the signals for any sensor network. Among varied antennas, micro strip fractal antenna (MFA) significantly contributes to increasing antenna gain. This study employs a hybrid optimization method known as the elephant clan updated grey wolf algorithm to introduce an optimized MFA design. This method optimizes antenna characteristics, including directivity and gain. Here, the factors, including length, width, ground plane length, height, and feed offset-X and feed offset-Y, are taken into account to achieve the best performance of gain and directivity. Ultimately, the superiority of the suggested technique over state-of-the-art strategies is calculated for various metrics such as cost and gain. The adopted model converges to a minimal value of 0.2872. Further, the spider monkey optimization (SMO) model accomplishes the worst performance over all other existing models like elephant herding optimization (EHO), grey wolf optimization (GWO), lion algorithm (LA), support vector regressor (SVR), bacterial foraging–particle swarm optimization (BF-PSO) and shark smell optimization (SSO). Effective MFA design is obtained using the suggested strategy regarding various parameters.

Resource management must attach importance to effective resource deployment. Aiming at the research of resource deployment system, firstly, as an important factor of resource deployment system, corporate technological innovation social responsibility (CISR) is analyzed. Based on this, this paper constructs a system dynamics model to analyze the changes in resource deployment system affected by CISR. The simulation model is developed using Venism personal learning edition (PLE). The results show that CISR, acted as a new factor affecting the resource deployment system, has a positive effect on resource deployment system performance. Moreover, when CISR exceeds the threshold value, the resource deployment system performance increases significantly faster, reflecting that the resource deployment system becomes more efficient. The results show that the method proposed in this paper is feasible and efficient. This research provides theoretical and practical implications for resource deployment system research.

This work proposes the application of an iterative learning model predictive control (ILMPC) approach based on an adaptive fault observer (FOBILMPC) for fault-tolerant control and trajectory tracking in air-breathing hypersonic vehicles. In order to increase the control amount, this online control legislation makes use of model predictive control (MPC) that is based on the concept of iterative learning control (ILC). By using offline data to decrease the linearized model’s faults, the strategy may effectively increase the robustness of the control system and guarantee that disturbances can be suppressed. An adaptive fault observer is created based on the suggested ILMPC approach in order to enhance overall fault tolerance by estimating and compensating for actuator disturbance and fault degree. During the derivation process, a linearized model of longitudinal dynamics is established. The suggested ILMPC approach is likely to be used in the design of hypersonic vehicle control systems since numerical simulations have demonstrated that it can decrease tracking error and speed up convergence when compared to the offline controller.

The lack of systematic and scientific top-level arrangement in the field of civil aircraft flight test leads to the problems of long duration and high cost. Based on the flight test activity, mathematical models of flight test duration and cost are established to set up the framework of flight test process. The top-level arrangement for flight test is optimized by multi-objective algorithm to reduce the duration and cost of flight test. In order to verify the necessity and validity of the mathematical models and the optimization algorithm of top-level arrangement, real flight test data is used to make an example calculation. Results show that the multi-objective optimization results of the top-level flight arrangement are better than the initial arrangement data, which can shorten the duration, reduce the cost, and improve the efficiency of flight test.

Tracking and analyzing data from research projects is critical for understanding research trends and supporting the development of science and technology strategies. However, the data from these projects is often complex and inadequate, making it challenging for researchers to conduct in-depth data mining to improve policies or management. To address this problem, this paper adopts a top-down approach to construct a knowledge graph (KG) for research projects. Firstly, we construct an integrated ontology by referring to the metamodel of various architectures, which is called the meta-model integration conceptual reference model. Subsequently, we use the dependency parsing method to extract knowledge from unstructured textual data and use the entity alignment method based on weakly supervised learning to classify the extracted entities, completing the construction of the KG for the research projects. In addition, a knowledge inference model based on representation learning is employed to achieve knowledge completion and improve the KG. Finally, experiments are conducted on the KG for research projects and the results demonstrate the effectiveness of the proposed method in enriching incomplete data within the KG.

In order to enhance the dynamic control precision of inertial stabilization platform (ISP), a disturbance sliding mode observer (DSMO) is proposed in this paper suppressing disturbance torques inherent within the system. The control accuracy of ISP is fundamentally circumscribed by various disturbance torques in rotating shaft. Therefore, a dynamic model of ISP incorporating composite perturbations is established with regard to the stabilization of axis in the inertial reference frame. Subsequently, an online estimator for control loop uncertainties based on the sliding mode control algorithm is designed to estimate the aggregate disturbances of various parameters uncertainties and other unmodeled disturbances that cannot be accurately calibrated. Finally, the proposed DSMO is integrated into a classical proportional-integral-derivative (PID) control scheme, utilizing feedforward approach to compensate the composite disturbance in the control loop online. The effectiveness of the proposed disturbance observer is validated through simulation and hardware experimentation, demonstrating a significant improvement in the dynamic control performance and robustness of the classical PID controller extensively utilized in the field of engineering.

Visual inertial odometry (VIO) problems have been extensively investigated in recent years. Existing VIO methods usually consider the localization or navigation issues of robots or autonomous vehicles in relatively small areas. This paper considers the problem of vision-aided inertial navigation (VIN) for aircrafts equipped with a strapdown inertial navigation system (SINS) and a downward-viewing camera. This is different from the traditional VIO problems in a larger working area with more precise inertial sensors. The goal is to utilize visual information to aid SINS to improve the navigation performance. In the multi-state constraint Kalman filter (MSCKF) framework, we introduce an anchor frame to construct necessary models and derive corresponding Jacobians to implement a VIN filter to directly update the position in the Earth-centered Earth-fixed (ECEF) frame and the velocity and attitude in the local level frame by feature measurements. Due to its filtering-based property, the proposed method is naturally low computational demanding and is suitable for applications with high real-time requirements. Simulation and real-world data experiments demonstrate that the proposed method can considerably improve the navigation performance relative to the SINS.

To enhance direction of arrival (DOA) estimation accuracy, this paper proposes a low-cost method for calibrating far-field steering vectors of large aperture millimeter wave radar (mmWR). To this end, we first derive the steering vectors with amplitude and phase errors, assuming that mmWR works in the time-sharing mode. Then, approximate relationship between the near-field calibration steering vector and the far-field calibration steering vector is analyzed, which is used to accomplish the mapping between the two of them. Finally, simulation results verify that the proposed method can effectively improve the angle measurement accuracy of mmWR with existing amplitude and phase errors.

In order to get rid of the dependence on high-precision centrifuges in accelerometer nonlinear coefficients calibration, this paper proposes a system-level calibration method for field condition. Firstly, a 42-dimension Kalman filter is constructed to reduce impact brought by turntable. Then, a biaxial rotation path is designed based on the accelerometer output model, including orthogonal 22 positions and tilt 12 positions, which enhances gravity excitation on nonlinear coefficients of accelerometer. Finally, sampling is carried out for calibration and further experiments. The results of static inertial navigation experiments lasting 4000 s show that compared with the traditional method, the proposed method reduces the position error by about 390 m.

In this paper, a comprehensive overview of radar detection methods for low-altitude targets in maritime environments is presented, focusing on the challenges posed by sea clutter and multipath scattering. The performance of the radar detection methods under sea clutter, multipath, and combined conditions is categorized and summarized, and future research directions are outlined to enhance radar detection performance for low–altitude targets in maritime environments.

The influence of ocean environment on navigation of autonomous underwater vehicle (AUV) cannot be ignored. In the marine environment, ocean currents, internal waves, and obstacles are usually considered in AUV path planning. In this paper, an improved particle swarm optimization (PSO) is proposed to solve three problems, traditional PSO algorithm is prone to fall into local optimization, path smoothing is always carried out after all the path planning steps, and the path fitness function is so simple that it cannot adapt to complex marine environment. The adaptive inertia weight and the “active” particle of the fish swarm algorithm are established to improve the global search and local search ability of the algorithm. The cubic spline interpolation method is combined with PSO to smooth the path in real time. The fitness function of the algorithm is optimized. Five evaluation indexes are comprehensively considered to solve the three-demensional (3D) path planning problem of AUV in the ocean currents and internal wave environment. The proposed method improves the safety of the path planning and saves energy.

Long-time coherent integration (LTCI) is an effective way for radar maneuvering target detection, but it faces the problem of a large number of search parameters and large amount of calculation. Realizing the simultaneous compensation of the range and Doppler migrations in complex clutter background, and at the same time improving the calculation efficiency has become an urgent problem to be solved. The sparse transformation theory is introduced to LTCI in this paper, and a non-parametric searching sparse LTCI (SLTCI) based maneuvering target detection method is proposed. This method performs time reversal (TR) and second-order Keystone transform (SKT) in the range frequency & slow-time data to complete high-order range walk compensation, and achieves the coherent integration of maneuvering target across range and Doppler units via the robust sparse fractional Fourier transform (RSFRFT). It can compensate for the nonlinear range migration caused by high-order motion. S-band and X-band radar data measured in sea clutter background are used to verify the detection performance of the proposed method, which can achieve better detection performance of maneuvering targets with less computational burden compared with several popular integration methods.

Nonperiodic interrupted sampling repeater jamming (ISRJ) against inverse synthetic aperture radar (ISAR) can obtain two-dimensional blanket jamming performance by joint fast and slow time domain interrupted modulation, which is obviously different from the conventional multi-false-target deception jamming. In this paper, a suppression method against this kind of novel jamming is proposed based on inter-pulse energy function and compressed sensing theory. By utilizing the discontinuous property of the jamming in slow time domain, the unjammed pulse is separated using the intra-pulse energy function difference. Based on this, the two-dimensional orthogonal matching pursuit (2D-OMP) algorithm is proposed. Further, it is proposed to reconstruct the ISAR image with the obtained unjammed pulse sequence. The validity of the proposed method is demonstrated via the Yake-42 plane data simulations.

This survey presents a comprehensive review of various methods and algorithms related to passing-through control of multi-robot systems in cluttered environments. Numerous studies have investigated this area, and we identify several avenues for enhancing existing methods. This survey describes some models of robots and commonly considered control objectives, followed by an in-depth analysis of four types of algorithms that can be employed for passing-through control: leader-follower formation control, multi-robot trajectory planning, control-based methods, and virtual tube planning and control. Furthermore, we conduct a comparative analysis of these techniques and provide some subjective and general evaluations.

With the rapid development of informatization, autonomy and intelligence, unmanned swarm formation intelligent operations will become the main combat mode of future wars. Typical unmanned swarm formations such as ground-based directed energy weapon formations, space-based kinetic energy weapon formations, and sea-based carrier-based formations have become the trump card for winning future wars. In a complex confrontation environment, these sophisticated weapon formation systems can precisely strike mobile threat group targets, making them extreme deterrents in joint combat applications. Based on this, first, this paper provides a comprehensive summary of the outstanding advantages, strategic position and combat style of unmanned clusters in joint warfare to highlight their important position in future warfare. Second, a detailed analysis of the technological breakthroughs in four key areas, situational awareness, heterogeneous coordination, mixed combat, and intelligent assessment of typical unmanned aerial vehicle (UAV) swarms in joint warfare, is presented. An in-depth analysis of the UAV swarm communication networking operating mechanism during joint warfare is provided to lay the theoretical foundation for subsequent cooperative tracking and control. Then, an in-depth analysis of the shut-in technology requirements of UAV clusters in joint warfare is provided to lay a theoretical foundation for subsequent cooperative tracking control. Finally, the technical requirements of UAV clusters in joint warfare are analysed in depth so the key technologies can form a closed-loop kill chain system and provide theoretical references for the study of intelligent command operations.

To address the confrontation decision-making issues in multi-round air combat, a dynamic game decision method is proposed based on decision tree for the confrontation of unmanned aerial vehicle (UAV) air combat. Based on game theory and the confrontation characteristics of air combat, a dynamic game process is constructed including the strategy sets, the situation information, and the maneuver decisions for both sides of air combat. By analyzing the UAV’s flight dynamics and the both sides’ information, a payment matrix is established through the situation advantage function, performance advantage function, and profit function. Furthermore, the dynamic game decision problem is solved based on the linear induction method to obtain the Nash equilibrium solution, where the decision tree method is introduced to obtain the optimal maneuver decision, thereby improving the situation advantage in the next round of confrontation. According to the analysis, the simulation results for the confrontation scenarios of multi-round air combat are presented to verify the effectiveness and advantages of the proposed method.

International freedom of the air (traffic rights) is a key resource for airlines to carry out international air transport business. An efficient and reasonable traffic right resource allocation within a country between airlines can affect the quality of a country’s participation in international air transport. In this paper, a multi-objective mixed-integer programming model for traffic rights resource allocation is developed to minimize passenger travel mileages and maximize the number of traffic rights resources allocated to hub airports and competitive carriers. A hybrid heuristic algorithm combining the genetic algorithm and the variable neighborhood search is devised to solve the model. The results show that the optimal allocation scheme aligns with the principle of fairness, indicating that the proposed model can play a certain guiding role in and provide an innovative perspective on traffic rights resource allocation in various countries.

As commercial drone delivery becomes increasingly popular, the extension of the vehicle routing problem with drones (VRPD) is emerging as an optimization problem of interests. This paper studies a variant of VRPD in multi-trip and multi-drop (VRP-mmD). The problem aims at making schedules for the trucks and drones such that the total travel time is minimized. This paper formulate the problem with a mixed integer programming model and propose a two-phase algorithm, i.e., a parallel route construction heuristic (PRCH) for the first phase and an adaptive neighbor searching heuristic (ANSH) for the second phase. The PRCH generates an initial solution by concurrently assigning as many nodes as possible to the truck–drone pair to progressively reduce the waiting time at the rendezvous node in the first phase. Then the ANSH improves the initial solution by adaptively exploring the neighborhoods in the second phase. Numerical tests on some benchmark data are conducted to verify the performance of the algorithm. The results show that the proposed algorithm can found better solutions than some state-of-the-art methods for all instances. Moreover, an extensive analysis highlights the stability of the proposed algorithm.

Deep learning has achieved excellent results in various tasks in the field of computer vision, especially in fine-grained visual categorization. It aims to distinguish the subordinate categories of the label-level categories. Due to high intra-class variances and high inter-class similarity, the fine-grained visual categorization is extremely challenging. This paper first briefly introduces and analyzes the related public datasets. After that, some of the latest methods are reviewed. Based on the feature types, the feature processing methods, and the overall structure used in the model, we divide them into three types of methods: methods based on general convolutional neural network (CNN) and strong supervision of parts, methods based on single feature processing, and methods based on multiple feature processing. Most methods of the first type have a relatively simple structure, which is the result of the initial research. The methods of the other two types include models that have special structures and training processes, which are helpful to obtain discriminative features. We conduct a specific analysis on several methods with high accuracy on public datasets. In addition, we support that the focus of the future research is to solve the demand of existing methods for the large amount of the data and the computing power. In terms of technology, the extraction of the subtle feature information with the burgeoning vision transformer (ViT) network is also an important research direction.