Journals >Electronics Optics & Control
Contents
2024
Volume: 31 Issue 9
19 Article(s)
Export citation format
Multi-UAV Cooperative Target Capture Based on Improved Hybrid PSO Algorithm
XU Nuo, ZHU Qian, XIE Xiaoyang, YU Tao... and LIU Sifan|Show fewer author(s)
Aiming at the problem of multi-UAV cooperative target capture,and considering that each UAV reaches the capture position at the same time,a two-layer solution architecture for task planning of multi-UAV cooperative capture is proposed on the basis of UAV kinematics constraints.In the task coordination layer,by improvinAiming at the problem of multi-UAV cooperative target capture,and considering that each UAV reaches the capture position at the same time,a two-layer solution architecture for task planning of multi-UAV cooperative capture is proposed on the basis of UAV kinematics constraints.In the task coordination layer,by improving the hybrid Particle Swarm Optimization (PSO) method and taking the minimum time for each UAV to reach the designated capture point as the goal,the multi-objective capture scheme is given by optimizing the scheduling.In the route planning layer,the initial state and kinematics constraints of UAVs are considered,and Dubins curve is adjusted to realize that each UAV reach the capture position at the same time.Simulation results demonstrate the effectiveness of the proposed method..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 1 (2024)
An Improved Particle Swarm Hybrid Path Planning Algorithm with Hill Climbing Strategy
KONG Pengfei
In order to improve the optimization ability of path planning,a hybrid path planning algorithm is proposed.Firstly,the improved Particle Swarm Optimization (PSO) algorithm is used to search the path,and then the hill climbing algorithm is used to refine the path.In the PSO algorithm,Tent chaotic mapping is used to initIn order to improve the optimization ability of path planning,a hybrid path planning algorithm is proposed.Firstly,the improved Particle Swarm Optimization (PSO) algorithm is used to search the path,and then the hill climbing algorithm is used to refine the path.In the PSO algorithm,Tent chaotic mapping is used to initialize the particle population,inertia weights are randomly updated,and the learning factors are dynamically adjusted asynchronously.After the search of the improved PSO algorithm,the hill climbing algorithm is used for further optimizing.The feasibility and effectiveness of our algorithm are verified by the simulation and comparison experiments in four kinds of terrain scenarios with different complexity levels.The algorithm combines global search with local search and improves the overall search performance..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 6 (2024)
Infrared and Visible Image Fusion Based on Multi-layer Convolution
CHEN Haixiu, FANG Weizhi, LU Kang, LU Cheng... and CHEN Ziang|Show fewer author(s)
To solve the problems of texture detail loss and poor visual perception of fused images in complex background,an infrared/visible image fusion algorithm based on multi-layer convolution is proposed.The network framework of the algorithm is divided into encoder,decoder and fusion network.An Efficient Channel Attention (To solve the problems of texture detail loss and poor visual perception of fused images in complex background,an infrared/visible image fusion algorithm based on multi-layer convolution is proposed.The network framework of the algorithm is divided into encoder,decoder and fusion network.An Efficient Channel Attention (ECA) mechanism is introduced into the encoder to encode the source image.Multi-layer Convolutional Fusion Network (MCFN) is formed by fusion of multi-layer convolution blocks,gradient convolution blocks,down sampled convolution block and the attention mechanism of convolution space channels.Feature fusion is performed through MCFN.Then the output fusion image is reconstructed through the decoder.Eight objective evaluation indicators are selected for comparison with five existing algorithms on two datasets.The results show that the fusion image of the proposed algorithm has prominent object,clear details and obvious contour,and the index is improved significantly,which accords with human visual perception..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 12 (2024)
Flight Stability Research on UAV with a Robotic Arm Based on Singular Perturbation Decomposition
MEI Ping, XIE Shuangshuang, ZHANG Hao, and LIU Yunping
In order to reduce the coupling effect between UAV and its robotic arm,and improve the stability of the quad-rotor UAV equipped with a robotic arm during flight,this article first establishes a system dynamics model,and then simplifies the model according to the actual situation.On this basis,the inner and outer loops In order to reduce the coupling effect between UAV and its robotic arm,and improve the stability of the quad-rotor UAV equipped with a robotic arm during flight,this article first establishes a system dynamics model,and then simplifies the model according to the actual situation.On this basis,the inner and outer loops are decomposed.The outer loop is used for controlling the position of the quad-rotor and the angle of the robotic arm,while the inner loop is for controlling the pitch angle of the quad-rotor.For the dynamic model of the outer loop,singular perturbation theory is used to decompose the outer loop variables,namely the linear velocity of the quad-rotor and the angular velocity of the robotic arm,into slow and fast subsystems,according to the velocity change characteristics,which once again reduces the coupling between the systems.For the decomposed slow and fast subsystems,non-singular terminal sliding mode control is designed and a disturbance observer is introduced to improve the stability of the armed quad-rotor UAV during flight.Finally,simulations verify the superiority of the designed algorithm..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 18 (2024)
A Weighted Response Siamese Network Tracking Algorithm Based on Adaptive Feature Fusion
FU Qiang, XIE Zhian, JI Yuanfa, and REN Fenghua
A weight response target tracking algorithm based on adaptive feature fusion is proposed to address the problem of drift or tracking loss due to the difficulty of extracting rich feature information by the tracker in complex scenes.Firstly,an improved VGG16 network is used to improve the discriminative ability.SecondlyA weight response target tracking algorithm based on adaptive feature fusion is proposed to address the problem of drift or tracking loss due to the difficulty of extracting rich feature information by the tracker in complex scenes.Firstly,an improved VGG16 network is used to improve the discriminative ability.Secondly,a residual semantic embedding module is employed to introduce deep semantic information into shallow features,and then the shallow feature response and deep feature response are weighted and fused to improve the localization accuracy and discriminative ability.The experimental results show that,compared with the benchmark algorithm,the evaluation indexes of the proposed algorithm,such as tracking success rate and accuracy,are both improved on the OTB2015 and VOT2017 datasets..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 25 (2024)
HCAC and MWFN Based Small Object Detection Algorithm
ZHU Gaofeng, WANG Zhixue, ZHU Fenghua, and XIONG Gang
UAV detection plays a key role in various fields.From the perspective of a UAV,factors such as the size of an object,different background interferences,and lighting will all affect the detection effect,leading to missed or false detection of the object.To solve this problem,a small object detection algorithm is proposeUAV detection plays a key role in various fields.From the perspective of a UAV,factors such as the size of an object,different background interferences,and lighting will all affect the detection effect,leading to missed or false detection of the object.To solve this problem,a small object detection algorithm is proposed.Firstly,a module of Hybrid Control of Attention-mechanism and Convolution (HCAC）is used to effectively extract contextual details of objects of different scales,directions,and shapes,while associating objects with position information through relative position encoding.Secondly,in view of the small size characteristics of small objects,a high-resolution detection branch is introduced,the large object detection head and its redundant network layers are pruned,and a Multi-level Weighted Feature fusion Network (MWFN) is used for multi-dimensional fusion.Finally,the WIoU loss is used as the bounding box regression loss,combined with the dynamic non-monotonic focusing mechanism,to evaluate the quality of the anchor box,so that the detector can handle anchor boxes of different qualities and improve the overall performance.Experiments were conducted on the UAV aerial photography dataset VisDrone2019.The results showed that,compared with the basic algorithm,the proposed algorithm has its accuracy and mAP increased by 9.0 and 9.8 percentage points respectively,which has better detection results..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 31 (2024)
An Infrared Ship Target Detection Algorithm Based on Improved YOLOv5s
ZHANG Lin, BO Jingdong, GONG Ruikun, and CUI Chuanjin
A lightweight infrared ship detection algorithm CBYOLOv5 combined with Knowledge Distillation (KD) is proposed to solve the problem of large amount of parameters and computation in traditional ship detection algorithms. Lightweight network Ghost module is introduced in YOLOv5s backbone network to realize lightweight dA lightweight infrared ship detection algorithm CBYOLOv5 combined with Knowledge Distillation (KD) is proposed to solve the problem of large amount of parameters and computation in traditional ship detection algorithms. Lightweight network Ghost module is introduced in YOLOv5s backbone network to realize lightweight detection network. A new neck structure of Asymptotic Feature Pyramid Network (AFPN) is introducedwhich can avoid the large semantic gap of nonadjacent level by fusing two adjacent lowlevel features and gradually fusing to higherlevel features. The VFL function is used to improve the imbalance of positive and negative samples in infrared ship target detection tasksso as to improve the overall performance of the model. FinallyKD is adopted to transfer the “knowledge” in the network of teachers with strong learning ability into the improved network model to improve the accuracy of classification and localization. Experimental results show that in infrared ship datasetin comparison with original algorithm YOLOv5sparameter amount is reduced by 38%and mAP is increased by 3.9 percentage pointswhile the model weight file is only 8.96×106which proves the proposed algorithm is effective and has certain practical value..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 38 (2024)
3D Path Planning of UAVs Based on Lens Imaging OppositeLearning TSO
SUN Xi, LIU Feng, and XUE Xiao
The Tuna Swarm Optimization (TSO) algorithm is easy to have poor search accuracy and high path cost when used for threedimensional path planning of UAVs.Thereforean improved TSO algorithm is designed based on lens imaging oppositelearning mechanism.Firstlynonlinear weight coefficient updatingthe best and worst oppThe Tuna Swarm Optimization (TSO) algorithm is easy to have poor search accuracy and high path cost when used for threedimensional path planning of UAVs.Thereforean improved TSO algorithm is designed based on lens imaging oppositelearning mechanism.Firstlynonlinear weight coefficient updatingthe best and worst opposite learningand lens imaging opposite learning strategies are introduced to optimize the performance of the TSO algorithm and to improve its accuracy when approaching the optimal solution.Thena threat model with geomorphic constraints and performance constraints is establishedand the 3D path planning problem of UAV is converted into an optimization problem of cost function.The improved algorithm is used to iteratively solve the path planningand search for the optimalsafe and obstacleavoiding path with the minimum cost.Simulation results of path planning show thatthe improved algorithm effectively solves the premature convergence problemand its planned path is shorter with lower costand the planning efficiency is improved..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 45 (2024)
Motion of Multi-robot Formation Based on ACO and APF
YANG Liwei, LI Ping, QUAN He, QIAN Song... and XUE Yan|Show fewer author(s)
An integrated formation path planning method based on improved ant colony algorithm and artificial potential field method is proposed for multi-robot path planning,which constructs the formation of multiple robots based on the leader-follower method,uses the improved Ant Colony Optimization (ACO) to provide global pathAn integrated formation path planning method based on improved ant colony algorithm and artificial potential field method is proposed for multi-robot path planning,which constructs the formation of multiple robots based on the leader-follower method,uses the improved Ant Colony Optimization (ACO) to provide global paths for the leader under complex obstacle environments,and uses the improved Artificial Potential Field (APF) to make decisions for obstacle avoidance.In the global path planning stage,the ACO improves the optimality and safety of the path.In the local path planning stage,the gravitational and repulsive coefficient of APF and the obstacle avoidance function are optimized to solve the problems of the standard APF of being prone to falling into local optimums and being susceptible to the interference of complex environments.In order to improve the formation movement capability of multiple robots,the leader adopts an adaptive navigation strategy to improve the path tracking capability,and the follower adaptively corrects the motion speed to better maintain the formation.The experimental results show that the proposed method successfully coordinates the multi-robot path planning and formation problems..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 52 (2024)
A Fast Dehazing Algorithm Based on Morphological Operators and Boundary Constraints
YUAN Zeqi, and HUANG Fuzhen
Aiming at the shortcomings of the dark channel prior dehazing algorithm,such as inaccurate transmission rate estimation,long running time,and color distortion in the sky area,a new dark channel prior fast dehazing algorithm is proposed.First,the foggy image is gray-scale corroded,and the improved dark channel is used tAiming at the shortcomings of the dark channel prior dehazing algorithm,such as inaccurate transmission rate estimation,long running time,and color distortion in the sky area,a new dark channel prior fast dehazing algorithm is proposed.First,the foggy image is gray-scale corroded,and the improved dark channel is used to determine the global atmospheric light value.Secondly,the minimum channel map is normalized by the atmospheric light to calculate the initial transmittance.Then,the morphological opening and closing operations are used to smooth the transmittance,and obtain the optimal transmittance according to the minimized objective function under the boundary constraints through iteration.Finally,the image after dehazing is obtained by using the atmospheric scattering model.The experimental results show that the proposed method achieves good results in visual effects,shows good dehazing ability in various image quality evaluation indicators,and is far superior to the traditional dark channel prior dehazing algorithm in terms of speed..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 58 (2024)
Vision/Gyro Combination Attitude Determination Technique for System with Large Time Delay
MA Haozhe, YU Feng, ZHANG Jie, and ZHEN Zihan
To address the issues of poor real-time performance and unsatisfied accuracy in attitude determination systems with large time delay,an algorithm is proposed for compensating the time delay in monocular vision.Taking the air-floating platform as the application background,a combined attitude determination system of “moTo address the issues of poor real-time performance and unsatisfied accuracy in attitude determination systems with large time delay,an algorithm is proposed for compensating the time delay in monocular vision.Taking the air-floating platform as the application background,a combined attitude determination system of “monocular vision+MEMS gyroscope” is designed to acquire the attitude information.The time-delay issue in monocular vision is compensated by integrating the gyroscope to correct the attitude quaternion,and a compensation model is established to analyze the main components of the time-delay error in monocular vision.The Extended Kalman Filter (EKF) is employed to process the angular velocity and the compensated attitude quaternion in real time,thus to obtain the attitude information of the lightweight air floating platform.Finally,the rationality and advantages of the attitude determination system design are validated through physical simulations.The experimental results demonstrate that the system design is reasonable and capable of providing high-precision attitude information,which can meet the requirements of the air-floating platform..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 64 (2024)
An Improved HighPerformance Object Detector Based on YOLOv7-tiny
ZHU Wenxu, SHI Tao, ZHOU Jiarun, LIU Zulin, and LIU Haixin
Aiming at the problems of large amount of network parameters and low detection accuracy of the existing YOLOseries object detectorsa highperformance universal object detector named YOLOv7TT is proposed based on YOLOv7tiny model.FirstlyGeneralized and Friendly ELAN (GFELAN) module is introduced into Backbone and Aiming at the problems of large amount of network parameters and low detection accuracy of the existing YOLOseries object detectorsa highperformance universal object detector named YOLOv7TT is proposed based on YOLOv7tiny model.FirstlyGeneralized and Friendly ELAN (GFELAN) module is introduced into Backbone and Neck networks to expand the width and depth of the network and eliminate the redundant features generated by the networkso as to reduce the parameter quantity and computation cost.Thenthe improved SimOTA sample allocation method is used to optimize the allocation of positive samples in the training process and accelerate the convergence speed of the network.Finallythe knowledge distillation method is used to distill and train the model to improve its detection accuracy while ensuring lightweight.The experimental results show that:1) Compared with YOLOv7tinyYOLOv7TT reduces the quantity of network parameters by 11% and 9.7%and improves the AP by 4.2 and 3.0 percentage points respectively on the VOC2007 and COCO2017 datasets;and 2) The model detection accuracy is further improved by using knowledge distillationthe AP reaches 59.4% (with 5.3 percentage points improved) on the VOC2007 datasetwhich effectively solves the problems of large quantity parameters and low detection accuracy..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 70 (2024)
A Pose Measurement Method Based on Time-Varying Color Fiducial Markers
WU You, and TAO Hongfeng
large-scale position and pose measurement in a low illumination environment,a method is proposed based on time-varying color fiducial markers.Firstly,LED lamps with programmable color are used to construct a multi-layer nested cooperative marker with time-varying color.Subsequently,the LEDs are detected,recognized,and large-scale position and pose measurement in a low illumination environment,a method is proposed based on time-varying color fiducial markers.Firstly,LED lamps with programmable color are used to construct a multi-layer nested cooperative marker with time-varying color.Subsequently,the LEDs are detected,recognized,and positioned.The EPnP algorithm is utilized to obtain single-layer pose measurement results.Based on an information fusion method weighted by size confidence,the fusion of multi-layer marker pose measurement results is achieved.Experimental results indicate that:1) Under normal lighting conditions,the proposed method demonstrates excellent performance with time-varying color fiducial markers and a large effective measurement range;and 2) Under low illumination conditions,the accuracy of position and attitude measurement using time-varying color fiducial markers surpasses those achieved under normal lighting conditions..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 77 (2024)
Detection of Object in UAV Aerial Photography Based on Reparameterized Attention
PENG Yanfei, CHEN Yankang, ZHAO Tao, YUAN Xiaolong, and CHEN Kun
Aiming at the problems of false detection and missed detection of objects in UAV aerial images caused by large changes in the object scale or mutual interference between the object and background,a object detection method based on reparameterized attention is proposed,which is applied to UAV aerial object detection.FirAiming at the problems of false detection and missed detection of objects in UAV aerial images caused by large changes in the object scale or mutual interference between the object and background,a object detection method based on reparameterized attention is proposed,which is applied to UAV aerial object detection.Firstly,the reparameterized coordinate attention module is proposed to enhance the relevant features and improve network ability to capture context information.Secondly,a multi-scale receptive field enhancement module is designed to reconstruct the backbone network,thereby enhancing the acceptance domain of feature map and improving the feature extraction ability of network.Then,a four-scale feature fusion detection network is proposed to improve detection ability of the network for small objects.Finally,a decoupling detection head is introduced to resolve the conflict between classification and regression tasks.In experiments on the VisDrone2021 dataset,mAP0.5 and racall rate of our algorithm is improved respectively by 7.6 and 5.5 percentage points in comparison with the original algorithm,and the algorithm also shows obvious advantages over other methods.The experimental result shows that,the improved method can better solve the above-mentioned problems of false detection and missed detection,and has good detection effect..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 81 (2024)
Design and Analysis of Composite Damping Vibration Reduction for Airborne Optoelectronic Pod
XU Ruirui, WANG Qianjin, HE Lei, LI Ya’nan, and LIU Xiansheng
Composite damping technology can reduce the vibration response through energy consumption of internal damping material.A composite damping vibration reduction design and an analysis is carried out for an airborne optoelectronic pod to attenuate its large vibration.The damping material,laying area,thickness, restrainingComposite damping technology can reduce the vibration response through energy consumption of internal damping material.A composite damping vibration reduction design and an analysis is carried out for an airborne optoelectronic pod to attenuate its large vibration.The damping material,laying area,thickness, restraining stiffness and so on are determined with overall consideration of the weight,size,service environment and attenuation effect.Simulation and test results show that the composite damping technology can effectively reduce the response to the optoelectronic pod,improve the imaging effect,and has a strong engineering application value..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 87 (2024)
A YOLOv7 Based Lightweight Underwater Target Detection Algorithm
TANG Luting, and HUANG Hongqiong
Underwater target detection is of great significance in marine scienceenvironmental protectionresource developmentmilitary defensecultural heritage protection and other fields.Howeverthe complex underwater environmentpoor underwater image quality and small biological aggregation may lead to missed detection and fUnderwater target detection is of great significance in marine scienceenvironmental protectionresource developmentmilitary defensecultural heritage protection and other fields.Howeverthe complex underwater environmentpoor underwater image quality and small biological aggregation may lead to missed detection and false detection in underwater target detectionso it is necessary to improve the detection accuracy.The realtime detection needs to design a faster network structure.Underwater devices have limited storage and computing power and need to maintain low computational overhead while ensuring accuracy.In view of these difficultiesan improved network YOLOv7PSS is proposed based on YOLOv7.FirstlyPConv convolution is used to replace some ordinary convolutions in the backbone network to reduce parameter quantity of the model and speed up training and prediction of the model.Thenthe SE attention mechanism is added to enhance the feature extraction abilityand SIoU loss function is adopted to accelerate network convergence and optimize model training process.Experimental results show that on the URPC2021 underwater target detection datasetthe proposed algorithm has a mAP of 87.3%which is 7.5% higher than that of the original modeland the parameter quantity is reduced by 11.9%which lays a foundation for the deployment of underwater equipment..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 92 (2024)
Binocular SLAM Based on Projection Transformation and Optical Flow
XIE Ying, and SHI Yongkang
Aiming at the decrease of positioning accuracy caused by key-point matching error of left and right images of binocular SLAM,an optical flow algorithm based on projection transformation is proposed.Firstly,the corner points on the left image are extracted,and the positions of the corner points on the right image are caAiming at the decrease of positioning accuracy caused by key-point matching error of left and right images of binocular SLAM,an optical flow algorithm based on projection transformation is proposed.Firstly,the corner points on the left image are extracted,and the positions of the corner points on the right image are calculated by the forward and inverse LK optical flow.Then,the homography matrix from right image to left image is calculated by using the successful matching corner points by RANSAC algorithm.Finally,the projection transformation image of the right image is calculated,and the corresponding position of the corner point of the left image is calculated by the forward and reverse LK optical flow,and the corner point is mapped to the original image of the right image.The experiments use partial sequences of the public datasets Euroc and KITTI,and the localization error of SLAM is evaluated by EVO tool.The results show that the improved algorithm can effectively reduce the corner point matching error with a certain increase in time complexity,and the root mean square error of binocular positioning is reduced respectively by 24.7% and 28.1% on average on partial sequences of Euroc and KITTI..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 98 (2024)
SQP-GPMP2 Algorithm Based Path Planning of Mobile Robot
GUO Xiwen, FU Shimo, WEI Yuanyuan, CHANG Qing, and WANG Yaoli
Aiming at the problem that the Gaussian Process Motion Planner 2 (GPMP2) has limited ability to deal with the constraints of nonlinear inequalities and is prone to fall into local minimum in maps with complex obstaclesand thus may produce collisionsan improved SQPGPMP2 algorithm is proposed by combining with SequentAiming at the problem that the Gaussian Process Motion Planner 2 (GPMP2) has limited ability to deal with the constraints of nonlinear inequalities and is prone to fall into local minimum in maps with complex obstaclesand thus may produce collisionsan improved SQPGPMP2 algorithm is proposed by combining with Sequential Quadratic Programming (SQP).Firstthe motion plan is regarded as trajectory optimizationand the initial trajectory state is obtained.Seconda collision cost function is introduced to represent the collision cost relationship between the robot and the obstacle.Finallythe SQP algorithm is used for iterative correction to ensure that the trajectory is collisionfree and dynamically reasonable.Simulation experiments show thatcompared with algorithms such as GPMP2the planning success rate of the algorithm on mazes of different sizes is improved by at least 20 percentage pointswhich proves that the algorithm is superior to them in dealing with the constraints and ensuring the planning efficiency..
Electronics Optics & Control
- Publication Date: Oct. 22, 2024
- Vol. 31, Issue 9, 104 (2024)
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20