To solve the problem of UAV formation control,an aircraft formation control algorithm based on the combination of virtual lead aircraft and nonlinear sight guidance law is designed.The formation flying process is divided into two phases,that is,route flight and route switching.In the phase of route flight,based on the virtual lead aircraft,the reference points are generated by using the nonlinear sight guidance law,the desired reference points for the formation aircraft are generated based on the geometric position of the formation,and the roll angle command needed for each aircraft to maintain the formation is obtained.In the phase of route switching,the virtual lead aircraft adopts circular arc flight,and the formation aircraft takes the virtual lead aircraft as the desired reference points to generate the roll angle command needed for formation keeping.The velocity command is obtained from the distance error between the virtual aircraft and the formation aircraft.Based on the nonlinear simulation model of a fixed-wing aircraft,a typical “V” formation is designed.The reliability and control accuracy of the multi-aircraft formation control algorithm in the phases of route flight and route switching are verified by numerical simulations.

In order to meet the requirements of multi-target reconnaissance tasks undertaken by UAVs,a UAV mission assignment and path planning optimization method based on the improved A* algorithm is proposed.Firstly,in the Geographic Information System (GIS),the threat area is marked,and the planning area is confirmed.Secondly,the route range matrix is estimated based on the improved A* algorithm,and then the mission assignment result is obtained by using the model of Travelling Salesman Problem (TSP) and the genetic algorithm.Thirdly,according to the multi-target sequential mission assignment result,the path optimization is realized based on the improved A* algorithm,and the path smoothing is implemented by using the method of cubic B-spline curve.Finally,according to the variation of threat areas and mission objectives,the local dynamic programming space is determined,and the dynamic assignment of local missions and the dynamic planning of local path are implemented.A UAV mission planning software is developed,and the experiments have verified the effectiveness of the proposed method.

Aiming at the cooperative search problem of UAV swarm under limited communication bandwidth and considering the existence of threat areas in the battlefield environment,an optimized search algorithm for distributed multi-UAV intention interaction is proposed.Firstly,based on the designed environmental situation map,the environment state model and the update method are built.Secondly,the search reward function under multiple step sizes is defined to obtain the decision-making intention of each UAV.Then,the fusion update operator and the dynamic communication topology under limited bandwidth are designed to realize the intention interaction among multiple UAVs.Finally,decision-making is conducted on the basis of obtaining the intentions of other UAVs.The simulation results show that the proposed algorithm can complete the full-coverage cooperative search task of multiple UAVs under the condition of limited communication bandwidth,and realize the avoidance of the threat areas.

To deal with the problem that it is difficult to predict acceleration information in the process of intercepting highly maneuvering target,a novel 3D intelligence guidance law is proposed by using adaptive RBF neural network and backstepping sliding mode control technique.Firstly,a 3D guidance model for maneuvering target interception is established based on the principle of zeroing the rate of Line-of-Sight (LOS),and a finite-time guidance law is designed by using backstepping sliding mode control algorithm.Then,the target acceleration is regarded as the uncertainties of the control system,which is online estimated and compensated for by using RBF neural network.Meanwhile,an adaptive switching gain is designed to restrain the chattering of the control system.Finally,the stability of the closed-loop control system is proved based on Lyapunovs direct method,and the effectiveness and superiority of the proposed 3D guidance law are verified via comparative simulations.

To solve the problem of nonlinear multi-agent cooperative output regulation under topology switching and time delays,a distributed state observer control algorithm and a fuzzy feedback control strategy are proposed.Firstly,the T-S fuzzy model is used to transform the nonlinear system into a linear system.Secondly,for the cooperative output regulation problem that the follower agents cannot obtain the state of the external leader system,a distributed state observer and a fuzzy output feedback controller are designed.The state observer is used to estimate the state information of the external leader system,and the state information is then fed back to the controller of the system,which solve the problem that the followers cannot obtain the state of the external leader.Thirdly,Lyapunov stability theory is applied to prove the stability of the nonlinear multi-agent system under topology switching and time delays.Finally,the correctness and effectiveness of the control law are verified by Matlab calculation examples.The simulation results show that the designed state observer can solve the problem of system state unmeasurability,the followers can track the leader signal,and the tracking error is zero.

The UAV formation control has such problems as insufficient intelligence level and lack of self-learning ability.Aiming at the problems,a UAV formation controller based on DDQN algorithm in deep reinforcement learning is designed.The controller can control the speed and heading channels simultaneously, so that the followers can track the leader and maintain the formation through self-learning,and the intelligence level of the UAV is improved.To verify the effectiveness of the designed controller,the traditional PID controller is used for comparison.The simulation results show that the DDQN-based controller can effectively create UAV formation and meet the requirements of UAV formation.The study is an effective exploration for intelligent control of UAV formation.

In order to solve the task allocation problem of UAV cluster searching cooperatively,based on beetle antennae search-particle swarm optimization(BSO)algorithm,a hybrid BSO Beetle Antennae Search(BSO-BAS) algorithm is designed.The new algorithm overcomes the shortcomings of Particle Swarm Optimization (PSO)algorithm,which is easy to fall into the local optimal solution and has unstable searching.Based on the model of Multi-Traveling Salesman Problem (MTSP),a UAV cluster task allocation model with multiple optimization objectives and constraints is established.Compared with the original optimization algorithm,the feasibility and stability of the proposed algorithm for UAV cooperative search task allocation are verified by experimental simulations.

In recent years,high-resolution digital displays have been widely used in airborne cockpit.Limited by the performance of the image sensor in a specific band and the video transmission link,the resolution of the airborne sensor image arriving at the display is often lower than that of the display.When these images are directly displayed,the effective display area is small,which is inconvenient for observation,and simply using the interpolation algorithm for image magnification will cause image blurring.Super Resolution (SR) reconstruction technology can predict image details while magnifying the image at the same time,which effectively improves image resolution.Currently,the CNN-based SR algorithm has become the most advanced SR algorithm owing to its excellent reconstruction effects (PSNR & SSIM).The existing SR algorithm has the problems of complex network structure,a large amount of parameters and huge computational resource consumption,which is inapplicable for real-time implementation in airborne embedded environment.To solve the above problems,a lightweight super-resolution algorithm named C-EDSR is proposed,which greatly reduces the amount of computation without harming reconstruction effects.The new algorithm provides a basis for further real-time implementation of SR algorithm in airborne embedded environment.The experimental results show that the proposed C-EDSR reduces 61.1% computation while decreasing PSNR by 0.026 dB and SSIM by 0.000 176 on average compared with EDSR.

Aiming at the problems of poor effect and low processing efficiency of image defogging algorithms,a fast defogging algorithm based on image fusion is proposed.Firstly,the original image is down-sampled to reduce the image size.The original image is copied to produce two images,and then the two images are converted to HSV color space respectively.For the first converted image,the v component is reduced and the accuracy of transmittance is improved.For the second converted image,the v component is adaptively processed and the s component is linearly stretched to improve the image color saturation.Then,the processed images are re-converted to RGB color space.The first re-converted image is defogged by using the improved dark channel algorithm and then histogram stretched.Finally,the two fused images are resampled.The experimental results show that the image after defogging has good subjective visual effect. Objectively,the complexity of the algorithm is reduced,and the Peak Signal-to-Noise Ratio (PSNR), Structural Similarity (SSIM),Information Entropy (IE) and Standard Deviation (SD) are significantly improved,which verify the effectiveness of the algorithm.