The vibration problem of the helicopter is prominent,and vibration control technology is one of the key technologies in the development of helicopter design.The research progress of helicopter vibration control technology,especially vibration active control technology is reviewed.Firstly,the technical classification of helicopter vibration control development up to now is presented,and the principles of passive vibration control technology for helicopters are briefly introduced.Then,the principle and development of active vibration control technology are emphasized,including active control of structural response,higher harmonic control,individual blade control,active twist rotor control,active trailing flap control,and active control of blade tip and microflaps.Finally,the development direction and future trends of active control technology for helicopter vibration are discussed.

Manned Aerial Vehicle/Unmanned Aerial Vehicle(MAV/UAV) cooperative air combat is an important development direction of future air combat.If MAVs and UAVs are used in cooperation,their strengths can be complemented to form a flexible and efficient combat system,which will play an important role in improving the effectiveness of future air combat.Firstly,this paper describes the evolution of the U.S.militarys combat concept in recent years,analyzes the development of MAV/UAV cooperative combat programs,and summarizes the features of MAV/UAV cooperative air combat.Secondly,taking penetrable air combat as a scenario,the specific operational use and operational procedure of MAV/UAV cooperative air combat are then analyzed.Finally,the key technologies and research status of MAV/UAV cooperative air combat are reviewed and analyzed from three aspects: trajectory planning,air combat decision making,and MAV/UAV formation control.The development prospect of future MAV/UAV cooperative air combat application is given.

Transformer’s self-attention mechanism is computationally intensive and prone to be distracted by the background,resulting in insufficient capturing of effective information and lower tracking performance.To address the problem,a feature-enhanced Sparse Transformer target tracking algorithm is proposed.Feature extraction is performed based on Siamese network backbone.In feature enhancement module,the contextual information generated from multi-scale feature maps is utilized to enhance the local features.The most relevant features of Sparse Transformer are utilized to generate target focusing features,and position encoding is embedded to enhance the accuracy of tracking localization.The proposed tracking model is trained in an end-to-end manner,and extensive experiments are conducted on five datasets including OTB100,VOT2018, LaSOT,etc.The experimental results show that the proposed algorithm achieves better tracking performance and with a real-time tracking speed of 34 frames per second.

To solve the problems of traditional Ant Colony Optimization (ACO) such as too many nodes in three-dimensional space and difficult algorithm search,a UAV three-dimensional flight path planning algorithm based on Improved Adaptive Ant Colony Optimization (IAACO) is proposed.Firstly,the three-dimensional space is divided into grids,so that the algorithm can be applied to three-dimensional flight path planning.Then,a non-uniform initial pheromone matrix is established,and an adaptive pheromone volatile factor is added,which can improve the searching efficiency and speed up the convergence rate of the algorithm.Finally,the objective function of UAV flight path optimization is further established by defining the 3D length index function and the 3D angle index function,and the global optimization of 3D flight path planning is realized.The simulation results show that the proposed algorithm has shorter running time and faster convergence speed,and the planned flight path is also shorter and smoother.

An integrated navigation fault-tolerance algorithm based on One-Class Support Vector Machine (OCSVM) is proposed.In the integrated navigation system,the failure of a subsystem will affect the accuracy of the whole navigation system.To solve the problem,the method based on OCSVM is adopted to detect and isolate the fault,and the fault-tolerance performance is analyzed.The simulation results show that,after applying the OCSVM based fault-tolerance algorithm,the fault detection module of the system can effectively isolate fault data,thereby reducing the position error of the multi-source integrated navigation system,and its reliability and stability are also improved.

Aiming at the problem of low accuracy in small target detection faced by the YOLOv7 algorithm when applied to optical remote sensing image ship target detection tasks,an improved optical remote sensing image ship target detection algorithm based on multi-dimensional attention mechanism dynamic convolution is proposed.Firstly,an efficient aggregation network integrating multi-dimensional attention mechanism dynamic convolution is designed through parallel strategy.The multi-dimensional attention mechanism dynamic convolution adaptively adjusts the importance of features in different dimensions,and the convolution kernel learns attention distribution along four dimensions,enhancing the ability of the feature fusion network to capture fine-grained features in data;Secondly,a multi-level super large convolution kernel layer is designed based on the multi-scale differences of ship targets,enriching the global feature description and improving the perception ability of the detection network.The experimental results show that:1) The improved algorithm achieves mAP of 93.4% and 90.1% respectively on two public datasets of HRSC2016 and DOTA;and 2) Compared with existing mainstream advanced algorithms,it achieves higher detection accuracy and improves recognition ability while reducing the missed detection and false detection rate of small ship targets.

As the application domains of UAVs continue to expand,unauthorized UAV flights pose serious threats to public safety.To solve the problems of false positives and negatives in detecting small intruding UAVs in complex flight scenarios,an anti-UAV target detection algorithm based on YOLOv5s-AntiUAV is proposed.Firstly,the Slim-Neck paradigm incorporating deep hyperparameter convolution is introduced to enhance feature extraction capabilities while maintaining computational efficiency.Secondly,the SPD-Conv modules are integrated into both the backbone and neck networks to improve the detection performance of small targets in low-resolution images.Finally,Alpha-CIoU is used to replace CIoU in YOLOv5s to augment the algorithms versatility.Results of comparative experiments of YOLOv5s-AntiUAV with YOLOv5s,SSD,Faster R-CNN on the Anti-UAV dataset show increases in mAP@0.5 by 1.1,12.1 and 4.9 percentage points respectively,which proves its practicality.The transfer experiments on the VisDrone2019 dataset show a 4.5 percentage points improvement in mAP@0.5 in comparison with YOLOv5s,which demonstrates that the improved algorithm is more robust than the original one.

In SAR ship dataset,small objects only account for a small proportion of the images pixels,the objects cannot be clearly recognized,and the detection efficiency is low.To solve the problems,an improved YOLOv7 based SAR ship target detection algorithm STSD-YOLO is proposed.Firstly,according to the characteristics of SAR images,the network structure is redesigned,and the relationship between multi-scale feature fusion and feature extraction is changed to solve the problem of losing detailed features due to excessive down-sampling times.Secondly,a lightweight attention mechanism named Shuffle Attention is used to fuse feature grouping with channel replacement based on the spatial domain and channel domain attention mechanism to improve the feature extraction ability of the network and reduce computational complexity.Then,the convolution variant DSConv is introduced to reduce computational intensity by storing only integers in the variable quantization kernel.Finally,the NWD metric is added to improve the performance of small object detection by modeling the bounding box as a 2D Gaussian distribution to measure the similarity between the bounding boxes of small objects.The HRSID ship dataset is adopted for experimental verification.In comparison with the baseline algorithm,the STSD-YOLO algorithm has its mAP increased by 9.9%in the ship detection task,and model volume reduced by 62.55%.Through comparative experiments,it is shown that the improved algorithm has better detection effects than other mainstream algorithms.It can effectively address the difficulties of SAR image detection,which is competent to carry out the ship detection task in SAR images.

To solve the problem that the traditional 3D laser point cloud scene segmentation algorithm tends to ignore the blurring of target boundaries,a 3D laser point cloud scene segmentation network is designed using a boundary contrastive learning algorithm,so as to improve the model’s prediction performance at the boundaries through contrastive learning.Firstly,the PointNet++ is taken as the backbone network,multi-scale downsampling feature encoding and upsampling feature decoding are used to learn the semantic features of different target categories in the point cloud,and the prediction of target categories is conducted point by point,so as to achieve overall scene segmentation.Then,a contrastive learning algorithm is introduced to capture the boundaries of sub-scene point clouds through iterations and mine fuzzy boundary points.Finally,the contrastive learning loss function is used to enhance the differentiation of boundary points belonging to different categories at the network training stage,which significantly improves the accuracy of 3D laser point cloud scene segmentation.A large number of experiments are conducted on the publicly available 3D laser point cloud scene segmentation dataset,and the results show that the proposed algorithm has the bests egmentation performance in 15 out of 19 semantic categories,with overall performance indicators superior to the comparison algorithms.The ablation experiments and visualization results also verify that the proposed algorithm can effectively improve the category prediction performance of boundary points in 3D laser point cloud scene segmentation tasks,which fully demonstrates the superiority of the proposed algorithm.

In ship detection of SAR images,the existing detection methods have the problems of low accuracy and low recall rate for ship detection because the ship targets are small and numerous in SAR images.To address the above problems,this paper proposes a ship target detection algorithm,Vessel-YOLO model for SAR images.Firstly,YOLOv8n is taken as the benchmark network,and a CASPP context space pyramid pooling structure is proposed to improve the capability of the model to extract features of different scales.Secondly,by improving the loss function of this model to Wise-IoU bounding box loss based on dynamic non-monotonic focusing mechanism,the models adaptability to different quality images is improved.The robustness and reliability of the model are verified by extensive experiments on the standard datasets of SAR-Ship-Dataset and SSDD.The experimental results show that,compared with YOLOv8n,Vessel-YOLO improves mAP0.5∶0.95 by 1.8 and 2.2 percentage points on the two datasets respectively,and the proposed model with higher accuracy outperforms existing SAR image ship detection models.

As a common weather condition,rainy weather can have a certain impact on computer vision,causing rain streaks and blurred details in images.Therefore,an efficient single image rain removal algorithm is needed to improve image quality.Most existing rain removal algorithms only focus on removing rain streaks,while neglecting the restoration of detailed information in the image afterremoval.In order to better detect rain streaks,a shallow feature extraction module and a deep feature extraction module is proposed.In the shallow extraction module,the residual dense block is selected,while in the deep extraction module,two dual-attention modules and two convolution layers are selected as residual groups composed of residual blocks.In order to restore image detail information,a multi-scale detail restoration module containing global and local branches is proposed.Numerous experiments on both synthetic and real datasets have shown that the proposed algorithm achieves PSNR and SSIM of 40.41 dB and 0.989 respectively,while preserving image details.

In the field of target detection,in complex environments such as nighttime,heavy fog,occlusion and battlefield camouflage,using a single image sensor cannot obtain all the scene information,which makes it difficult to improve the accuracy of target detection in complex environments.In view of this,a BF-YOLOv5 algorithm based on YOLOv5 is proposed.The algorithm adopts a dual-branch structure.Visible images and infrared images are read through two Backbones,and CBAM is fused into each Backbone.The importance of each feature channel is automatically obtained by learning,and the obtained importance is used to enhance features and suppress those that are not important to the current task.BiFormer attention mechanism is integrated into the Neck section to improve the detection ability on small targets.The experiments show that the detection accuracy of BF-YOLOv5 algorithm on infrared and visible image dataset FLIR and LLVIP is higher than that of the original algorithm,and the mean Average Precision (mAP) on FLIR dataset is as high as 86.6%,which is 2.2 percentage points higher than that of the original dual-branch algorithm,and the detection performance on the fused infrared and visible images is significantly improved.

In order to solve the obstacle avoidance problem of UAV in complex environments,a UAV dynamic path planning algorithm based on improved Bi-RRT and DWA is proposed.The Bi-RRT algorithm is improved by setting heuristic function,dynamic step size and safe distance to improve the search efficiency and safety of the global path.Then,the redundant road sections in the path are trimmed,and the optimal global path is obtained by interpolation and smoothing operations on the trimmed path.The obstacle distance evaluation function is modified and the target distance evaluation function is introduced to improve the accuracy of the score of the predicted local trajectory.Then,the velocity command is output in real time to control the UAV to track the optimal global path and realize local dynamic obstacle avoidance.The simulation results show that the path generated by the improved Bi-RRT algorithm is shorter and smoother with higher safety and less planning time.In the complex environments with both dynamic and static obstacles,the proposed fusion algorithm can control the UAV to accurately track the optimal global path and efficiently complete local dynamic obstacle avoidance.

To solve the problem of low computational efficiency of existing configuration optimization methods for collaborative navigation of clustered UAVs,this paper proposes a configuration optimization method for collaborative navigation of clustered UAVs based on the improved particle swarm optimization algorithm.Firstly,the collaborative accuracy factor is taken as the configuration evaluation standard to improve the evaluation accuracy.Based on this,the adaptive inertia factor and the asymmetric learning factor are used to improve the particle swarm optimization algorithm,which improves the optimization performance and convergence rate of the algorithm.Meanwhile,the optimal configuration is selected to improve the positioning accuracy of the clustered UAV collaborative navigation.The simulation results show that,compared with the traditional particle swarm optimization algorithm,the improved particle swarm optimization algorithm reduces the value of the collaborative accuracy factor by 33%.Based on this,the proposed method improves the positioning accuracy of user UAVs in the cluster and reduces the computational amount of configuration optimization,which is conducive to its wide application to large-scale UAV cluster.

In view of the large number of small targets with wide scale scope in aerial images,the YOLOv5s structure is improved from four aspects,and the LL-YOLOv5s network is proposed.The improvement measures mainly include deleting the feature layer of 32 times undersampling in the backbone network and using only two detection heads,so that the model can focus on the detection of small targets.The LL-YOLOv5s improves the mAP by 2.9 percentage points on the DOTA-v1.5 aerial dataset.Then,a simple and efficient self-attention module called qv self-attention module is proposed,which is added to the position before the first detection head of LL-YOLOv5s,and the mAP is further improved by 0.9 percentage points at the cost of adding a small amount of calculation.It is found that the combination of qv self-attention module with convolution layer further improves the mAP by 0.4 percentage points.Compared with YOLOv5s,the improved model greatly reduces the number of parameters and improves the detection accuracy significantly at the cost of adding a small amount of calculation.

To mitigate the adverse impacts of atmospheric turbulence-induced wavefront distortion of optical signals during communication in coherent free-space optical communication systems, an enhanced Stochastic Parallel Gradient Descent (SPGD) algorithm based on the Aggregated Momentum (AM) and the Adaptive and Momental Bound (AdaMod) optimizer is proposed. Simulation results show that the optimization algorithm can significantly improve the convergence speed and robustness, and effectively reduce the peak-to-valley and root-mean-square of wavefront distortion, thus more effectively suppressing the negative impact of atmospheric turbulence on the mixing efficiency and BER of coherent FSOC systems.

In the existing color image deblurring process,there are such phenomena as color imbalance,step effect and artifacts.To solve the problems,an image deblurring optimization method based on fractional total variation and low-rank regularization is proposed.Firstly,the color image in traditional RGB space is converted to YCbCr color space,and its luminance channel characteristics are used to solve the problem of color imbalance.Secondly,fractional total variation characteristics are used to eliminate step effect in image recovery tasks.Moreover,the weighted kernel norm low-rank regularization is introduced to further suppress artifacts and noise.Finally,the Alternating Direction Method of Multipliers (ADMM) is used to design an efficient solving method,and the optimal estimation of the clear image is obtained through iterative optimization.The experimental results of color image testing show that the proposed method exhibits fine performance in image deblurring tasks in terms of both visual recovery effects and objective evaluation indexes.

Based on the design analysis and calculation of multi-layer diffractive optical elements in medium and long wavelength infrared bands,an infrared optical system with a focal length of 200 mm is designed.The transmittance in medium and long wavelength infrared bands is greater than 80%,the imaging effect is good and meets the requirements of non-thermal design.Double-layer diffraction technology is adopted,and materials such as germanium,zinc selenide and zinc sulfide are used to achieve the diffraction goal with high efficiency and improve the optical transmittance of the two bands.The optical system structure of secondary imaging is adopted to realize optical small-diameter design.The optical transfer function can achieve the requirements of non-thermal imaging in the environment of -55 ℃ to +71 ℃,thus achieving the design purpose.

In order to apply CAN bus to the Remote Power Distribution Unit (RPDU) system of domestically-produced large aircraft,an improved application scheme of CAN bus is designed.According to the particularity of the RPDU system of domestically-produced large aircraft,the appropriate network topology and communication rate are selected.The structure of CAN identifiers and data domain is reasonably divided,and the format of converting CAN message into ARINC664 message is defined.Then,the communication module of satellite RPDU and gateway RPDU is designed,in which the availability and integrity of CAN communication are improved based on ARINC825 protocol,and the communication scheduling mechanism of ARINC825 is improved based on TTCAN protocol to ensure the certainty of CAN communication.The experimental results verify that the proposed scheme meets the design requirements and is feasible.