To overcome the shortcomings of the standard Grasshopper Optimization Algorithm (GOA),such as low search accuracy,prone to get local optimal solution and poor stability,an improved GOA based on mutualism and hybrid mutation strategy is proposed.Firstly,a nonlinear reconstruction method for convergence factor is introduced,which balances global search and local development.An individual hybrid mutation mechanism based on Gaussian-Cauchy distribution is designed to effectively avoid local optimal solution.Then,a mutualism strategy is introduced to enhance individual diversity and improve the global optimization ability of the algorithm.Then,the cost model of UAV path planning is established,and the path planning is transformed into a multi-dimensional function optimization problem.The improved GOA is used to solve the path planning problem.The fitness of individual positions is evaluated by the objective function that comprehensively considers the threat cost and the energy consumption cost.Iterations are conducted to obtain the optimal path,and the B-spline curve is introduced to smooth the final path connected by scattered points.The experimental results show that the improved algorithm has higher search accuracy,and the path can successfully avoid all of the threat areas.The study has high reference value for path planning in the area of Internet of Vehicles (IOV).

To solve the problem of multi-UAV cooperative strike in battlefield environments and in order to improve the survival rate of the UAVs and ensure the accurate strike on the target,a segmented path planning method with simultaneous arrival of the terminals and uniform spatial distribution is proposed.Firstly,a multi-objective trajectory optimization model that meets the requirements of battlefield is constructed.Secondly,by introducing cooperative variables,a cooperative attack strategy with multi-directional simultaneous arrival is designed.Then,Particle Swarm Optimization (PSO) algorithm is improved by adopting the principle of population splitting and merging and using the segmented inertia weight adjustment strategy and the population initialization method of Tent chaotic mapping.Finally,the improved algorithm is applied to problem solving.The simulation results show that the proposed model can realize multi-UAV cooperative strike on the target in complex and multi-threat battlefield environment,and the improved PSO algorithm has faster convergence and higher accuracy.

In order to adapt to the characteristics of cruise-type stamping extended-range guided artillery shells that the center of mass is constantly moving forward during cruise flight and the range of aerodynamic parameters is large,and in order to realize adaptive control during cruise flight,a three-loop autopilot based on fuzzy adaptive PID algorithm is designed.Firstly,the small perturbation method is used to linearize the system of shell motion equations in the longitudinal plane and deduce the transfer function of each attitude angle of the shell,and the feature points at the beginning and end of the cruise are selected for dynamic characteristic analysis.Based on this,a three-loop autopilot is designed to increase the phase angle margin of the system and improve its damping characteristics.The inner-loop control parameters at the feature points at the beginning and end of the cruise are designed by using the frequency domain analysis method and the root trajectory method,and the inner-loop control parameters of other non-feature points are calculated by interpolation.The outer loop is designed by using fuzzy adaptive PID.The simulation results show that the autopilot designed by the fuzzy adaptive algorithm has good control quality during cruise flight.

To deal with actuator fault of loss of effectiveness,drift fault and air-wake disturbance,an adaptive fixed-time fault-tolerant control method under direct lift is proposed.Firstly,a new precise-feedback linearization model of the carrier-based aircraft with flaps is established,and based on the new fixed-time stability theorem,a sliding mode controller is designed for the direct lift mode,which improves the convergence speed and carrier landing accuracy of the system.Then,an adaptive updating law in the form of nonlinear differential equations is introduced to online estimate the minimum value of the fault factor and the upper bound of the lump disturbance,and compensate for the influence of the faults and disturbance.The fixed-time convergence of the closed-loop system is proved based on Lyapunov method.Finally,a simulation is conducted in comparison with other methods,and the results prove the effectiveness and superiority of the designed method.

In order to reduce the difficulty in the ranging of monocular marine images and improve the degree of information mining of marine images,a monocular ranging method is proposed.Based on the sea-sky line detection and pose-calculation monocular ranging technology,the camera pose is calibrated by using the sea-sky line in the image,and the external parameters of the camera are obtained.In combination with the cameras internal parameters calibrated in advance,the model describing the geometric relationship between the target and the camera in the imaging space is constructed.Considering the influence of the Earths curvature,a real-time monocular marine ranging method based on sea-sky line calibration is deduced.Finally,the Monte Carlo method is used to simulate the ranging model and analyze its error characteristics.

As for Synthetic Aperture Radar (SAR) target recognition,a method based on modified sparse representation is proposed.Firstly,based on traditional Sparse Representation-based Classification (SRC),the sparse representation coefficient vector is calculated on the global dictionary.Based on this,the optimal local dictionaries are selected for different training categories.Then,the test samples are reconstructed using different local dictionaries.Finally,the target categories of the test samples are determined by comparing the reconstruction errors of different categories.In the experiments,the MSTAR dataset is taken as samples for testing and validation.The superiority of the performance of the proposed method is confirmed.

UAV data link has the characteristics of multi-service fusion,complex structure and dynamic reconfiguration,and the traditional reliability modeling methods cannot accurately represent its synchronization,concurrency,distribution,conflicts,resource sharing or competition.To solve the problem,a dynamic reliability modeling method of data link service based on GSPN is studied.The concepts of resource,repository and transition in GSPN model are used to describe various resources,states,behaviors and cooperative relationships in the data link system.The dynamic running process of the system is simulated by using specific random variables to represent the states and duration of the components.PIPE software tool is used for the formal modeling and description of the three major services of UAV data link,that is,uplink remote control,downlink remote measurement and downlink task payload transmission,so as to realize the accurate reliability evaluation.The case study shows that the proposed method can provide a trustworthy reliability evaluation method for UAV data link.

To address the problems of low calculation efficiency and limited applicability of the traditional UAV trajectory planning methods,a trajectory planning method that considers the integration of global and local planning is proposed.Firstly,a geometric track collision detection method is proposed to improve the Jump Point Search (JPS) algorithm to enhance the UAV trajectory planning capability in the global environment.Secondly,a trajectory planning framework for the integration strategy is proposed to extend the evaluation sub-function of the dynamic window method by integrating the UAVs global trajectory information with local trajectory information of the dynamic window to ensure the UAVs flexible obstacle avoidance capability in dynamic environments.Finally,the advantages of the improved JPS algorithm over existing methods in terms of global search time,trajectory distance and number of trajectory turning points are demonstrated through simulation experiments,and the stable obstacle avoidance capability of the integration algorithm in dynamic environments is verified.

When using traditional Siam network tracking algorithms for target tracking,if the target position is far from the target center point on the feature graph,the tracking performance indexes will be lowered and the tracking effects will be poor.To solve this problem,an improved Siamese Box Adaptive Network (SiamBAN) algorithm is proposed.The center prediction network is introduced to determine the distance between the target position and the target center point on the feature graph.When the distance between the two positions is larger,the centrality score is lower and the algorithms ability to remove outliers in responses is better.The simulation results on datasets of OTB100,VOT2016 and VOT2018 show that the improved SiamBAN algorithm can effectively reject the outliers on the response graph and significantly enhance the tracking performance indexes in comparison with mainstream tracking algorithms.

The proportional guidance is often used in the terminal guidance section of surface-to-air missile,while the platform self-defense jammer is often used to defend surface-to-air missile.This paper discusses the control characteristics of proportional guidance for surface-to-air missile and the photoelectric interference environment.A new simplified calculation method for the flight trajectory of the missile in the terminal guidance section and a mathematical model for the miss distance of the missile after effective photoelectric interference are presented.Based on this,the hit probability of the missile under the interference condition is calculated and analyzed.This paper focuses on the boundary conditions when the missile cannot hit the target,that is,the corresponding missile attitude,range,and flight control parameters when the photoelectric interference is effective,and discusses the missile hit probability under photoelectric interference and the boundary conditions when the missile cannot hit the target.The study can be used to quantitatively evaluate the combat effectiveness of the platform self-defense interference equipment and evaluate the anti-interference performance of the missile seeker.

To solve the problem of collaborative navigation of UAV swarm,a method for collaborative navigation of UAV swarm based on Gaussian Belief Propagation (GaBP) is proposed.The sensor information of the UAV itself and the ranging information of the UAV swarm are modeled through the probabilistic graphical model,and the GaBP algorithm is introduced to complete the effective fusion of collaborative information.The simulation results show that,compared with the traditional collaborative Extended Kalman Filter(EKF)navigation method,the new method improves the positioning accuracy of the UAV without satellite signals from 4.82 m to 1.97 m.At the same time,under the condition of similar performance,the computational efficiency of the new algorithm is improved by nearly 40 times compared with that of the traditional belief propagation algorithm.

In the SINS/GPS integrated navigation system,the measurement signals provided by GPS are easily disturbed,which leads to the degradation of filtering accuracy.To solve this problem,a fault-tolerance method for the integrated navigation system based on EM algorithm is proposed.By establishing a two-dimensional Gaussian mixture model for each component of the measurement noise,an indicator variable is introduced to indicate the probability of measurement value sampling from abnormal distribution.Based on the state equation of the system,the state estimate of the last moment and the measurement value of the current moment,the EM algorithm is utilized to calculate the maximum likelihood estimate of the true measurement at the current moment.Besides,the sequential filtering algorithm is applied to finely detect and correct the measurement value of each component.Experiment aboard cars shows that the proposed method can effectively improve the fusion accuracy and reliability of the SINS/GPS integrated navigation system in the case of GPS signal anomaly.

The structure of image background clutters is irregular and complex,while the structure of the target is regular and simple.The structural differences between them provide an effective way for clutter modeling.This paper proposes a new method to characterize background clutters based on the images structural complexity.Firstly,the directional gradient histogram can effectively characterize the structural information of the image,and the entropy of directional gradient distribution in the local region of the image is used to characterize the structural complexity of the corresponding region.Secondly,considering that the visual system of human eyes is sensitive to the contrast of the image,the differences between background clutters and the target are measured in image direction gradient distribution entropy space and contrast space respectively,and the background clutter model is constructed by weighted summation of standard variance.Finally,the relationship between background clutters and target acquisition performance is deduced,and the target acquisition performance evaluation model of the photoelectric imaging system is established.The model is verified by using open datasets.The experimental results show that the target acquisition performance evaluated by the proposed method is consistent with the actual target acquisition performance in the field,and the proposed method is superior to the existing clutter modeling methods in terms of Root Mean Square Error (RMSE) and correlation.

To solve the problems of the existing CNN-based methods,such as feature information loss,serious clutter information interference,ignoring the correlation between different scale features,and requiring a large number of training samples,a tank detection method based on Siamese Feature-Guided Multi-Scale Network (SFGMSN) is proposed.In SFGMSN,an improved Inception module is designed to extract the multi-scale features of the tank target image and carry out feature fusion to better recover the fine segmentation information of the tank target.To improve the feature perception ability of the target region and suppress background interference,a Local Channel Attention Mechanism (LCA-M) is designed to obtain more accurate detection results.Finally,ameta-learner is used to detect tank targets.SFGMSN makes full use of the advantages of multi-scale convolution,dilated convolution,Siamese network,local channel attention mechanism and the meta-learner.It can solve the problems that the traditional CNN model over-relies on a large number of training samples and may have low accuracy and poor generalization under the condition of small sample size.The experimental results on the tank image dataset show that the proposed method is effective in tank detection,with an average detection accuracy of 90.12%.It can realize tank detection in complex scenes and has good robustness to low-resolution tank images.

To solve the problem of trajectory tracking of fully-actuated Unmanned Surface Vessels (USV),a dual-layer control method combining Model Predictive Control(MPC) with Integral Sliding Mode Control(ISMC) is proposed.Firstly,as for the kinematics model of the USV system,a model predictive controller is designed to obtain the desired velocity signal satisfying the constraints according to the desired trajectory.As for the dynamics model,an integral sliding mode controller is designed to track the desired signal in the presence of external disturbances,which improves the robustness of the system.A nonlinear disturbance observer is designed to estimate the external disturbance and compensate for the external disturbance during the design of the control law.Finally,the stability of the system is proved by the Lyapunov method.Numerical simulation proves that the combination of the two can effectively achieve the trajectory tracking of the fully-actuated USV.

The high complexity of modern battlefield and the incompleteness of various situational information make it difficult for the traditional air target threat assessment methods to be applicable to the case of lack of data.To solve this problem,firstly,the Improved Quantitative Biclustering Algorithm (IQBA) that introduces Isometric Logratio Transformation (ILT) and spherical coordinate transformation is used to complete the missed data of target attributes.Then,the improved entropy weight method based on the gray target theory and index correlation is used to analyze the influence degree of each attribute,and the weight of the target attributes is objectively calculated.Finally,the improved gray correlation TOPSIS method is used to assess the threat of the target in a combat mission and classify threat levels.The simulation experiment results show that the proposed method can effectively realize air target threat assessment under the condition of information shortage,and the assessment results are more objective and reliable.

At present,there are difficulties in the constructing the software testing process and guaranteeing the quality and efficiency of software testing.To solve the problems,based on the relevant practices of software testing,the relevant ideas,strategies and practices of software testing are put forward.In terms of testing process,the testing process and testing technology are integrated and reconstructed to make the arrangement of the testing more reasonable.In terms of testing methods,the testing is conducted by comparing the functions of similar products to reduce demand dependency,and the software security is analyzed from the perspective of source codes to reduce demand dependency.The software testing tasks are classified and graded based on software importance.A targeted testing on high-risk requirements is conducted based on historical problem data.The combined testing method is adopted to improve the automation level of test case design,and a simulation test environment oriented to use scenarios is constructed and the test process is recorded to improve the automation level of test case generation.

With the increasingly complex low-altitude airspace environment,the probability of conflict among UAVs performing missions is increasing.Traditional reinforcement learning algorithms of SAC and DDPG suffer from slow convergence rate and unstable convergence in solving the problem of collision avoidance among multiple UAVs in limited airspace.To solve the problems,a Multi-Agent Reinforcement Learning (MARL) method based on PPO2 algorithm is proposed.Firstly,the multi-UAV flight decision-making problem is described as a Markov decision-making process.Secondly,the state space and reward function are designed to optimize the strategy by maximizing the cumulative reward,so that the overall training is more stable and converges faster.Finally,a flight simulation scene is built based on the deep learning framework TensorFlow and the reinforcement learning environment Gym,and simulation experiments are carried out.The experimental results show that the proposed method improves the success rate of collision avoidance by about 37.74 and 49.15 percent points respectively compared with that of the SAC and DDPG algorithms,which can better solve the problem of collision avoidance among multiple UAVs,and is better in terms of convergence rate and convergence stability.