Abstract:Aiming at the problems of slower particle convergence and poor positioning accuracy when using traditional Monte Carlo positioning algorithms in the navigation and positioning process of mobile robots, as well as low relocation efficiency after artificial kidnapping, this article gives an improved Particle filter positioning method to improve the navigation and positioning efficiency of mobile robots. First of all, it is improved on the basis of the Monte Carlo positioning algorithm and integrated into the method of adaptive region division to ensure that the region contains more effective information, reduce the convergence time of particles, and complete the preliminary coarse positioning of the robot. Then, in the particle sampling and resampling stage, the normal distribution probability model is used to update the particle weights to achieve faster and more efficient global positioning. Through experimental comparison and analysis, compared with the Monte Carlo positioning algorithm, the given method has shortened the time consumption by 4 s, and the adaptive Monte Carlo positioning method in this paper can keep the positioning error at about 6 cm, thus verifying the given method Effectiveness and stability.

Abstract:A multi-layer optimization method for mobile robot path planning is proposed for the problems of map environment modeling and ant colony algorithm. In this method, firstly, the U-trap raster region is convexized to avoid the pre-search confusion, a new state transfer rule is designed to solve the problem that the path of conventional ant colony planning is too close to the obstacles, the distance heuristic function is improved to effectively improve the convergence speed of the algorithm, the smoothing heuristic function is designed to increase the chance of ants going straight when local exploration is performed to improve the initial path smoothing, the update principle is proposed to allocate pheromones according to the distance length and The update principle of pheromone assignment by distance and smoothness is proposed to further improve the convergence speed of the algorithm by using high-quality ants for global pheromone update, the maximum-minimum ant strategy is used to prevent the ant colony from falling into local optimum, the redundant points are removed by the secondary path optimization strategy to further improve the path smoothness. Simulation and experimental results show that the method can plan a safe and comprehensive path for the mobile robot, which provides a practical method for the solution of path planning.

Abstract:Aiming at the safety problem of robot force / position control application, a fast and practical robot collision detection algorithm based on a dynamic model is proposed. Firstly, design the robot force / position safety control program, establish a dynamics-based collision detection mathematical model. Then, based on the static LuGre friction model to identify and compensate the robot joint friction, design a simplified robot force / position control collision detection model. Finally, using a light industrial robotic arm as an experimental platform, the robot executes force / position control to collide with human to verify the algorithm. The experimental results show that the proposed method can detect abnormal collisions of more than 4 N·m between the connecting rods of the robot, and can effectively realize the safety control of the robot force / position, which has a certain engineering reference value.

Abstract:From the perspective of maintaining the safe and stable operation of the microgrid and prolonging the service life of the battery, an improved hybrid energy storage control strategy is proposed, which is designed by decomposing the power to be stabilized into three components in the high, middle and low frequency bands, and introducing the feedback of the energy storage system SOC. The hybrid energy storage power distribution strategy reduces the number of charging and discharging of the battery and prolongs its service life. At the same time, it prevents the energy storage system from overcharging / overdischarging, so that the hybrid energy storage can respond stably and safely to the grid energy dispatch. The actual calculation examples and the comparison simulation experiment results prove that: compared with the traditional control strategy, the improved control strategy reduces the charge and discharge times of the battery by 58. 3% under the same working conditions, and can effectively solve the problem that the system cannot operate normally and stably due to the overcharge / overdischarge of the energy storage equipment.

Abstract:The observable degree of navigation state has a significant effect on the state estimation of GNSS / INS. In order to improve the accuracy of heading of land vehicle, an improved robust cubature Kalman filter (RCKF) method is proposed. First, the resampling-free sigma-point update framework is employed to separate the cubature point update from the Gaussian information limitation, so that improve the propagation efficiency of the information contained in instantiated points in the iteratively filtering period. Secondly, in order to improve the heading of land vehicle when it travels along a straight-line, the Gaussian process (GP) is introduced into the uncertainty calibration of moment approximation of system model based on state observability analysis. Simulation results indicate that GP-RCKF improves the heading angle obviously when the state observability is weak, and compared with RCKF the heading is improved by 28. 9%.

Abstract:Gesture recognition is an important means of human-computer interaction. In order to more accurately recognize gestures and eliminate the interference of environmental conditions such as lighting, and reduce the key point jitter recognition error caused by the high-dimensional space transformation of the hand at the meanwhile, a gesture key point method of extraction based on the Mask R-CNN model and Savitzky-Golay filter is proposed. This method uses the Mask R-CNN model to process RGB three-channel images, and performs object recognition and segmentation on each image, and obtains 21 bone points and background positions of the hand and performs model training. Then uses neural network features to match the video stream and mark 22 key points. Furthermore, the point data is smoothed by using Savitzky-Golay filter and then redraw the data to obtain stable gesture extraction and reconstruction results. This method is used in bone point extraction experiments. Experimental results show that the method can eliminate environmental interference to the greatest extent and accurately extract key points. Compared with traditional gesture key point extraction based on contour segmentation, the accuracy reaches 93. 48%. At the same time, the robustness of the model is greatly improved.

Abstract:Airborne oxygen concentrator based on molecular sieve bed pressure swing adsorption principle is the core component of aircraft life support system, which can provide oxygen for pilots during flight. The degradation analysis of airborne oxygen concentrator can realize fault early warning, which is of great significance for the condition based maintenance of airborne oxygen concentrator and the construction of aircraft health management system. The degradation process of airborne oxygen concentrator can be divided into two stages: steady stage and accelerated degradation stage. However, due to the uncertainty of degradation mode change point, the transformation of degradation mode is uncertain. Therefore, it is very important to correctly identify the turning point of degradation mode. Oxygen partial pressure is an important parameter to reflect the oxygen production capacity of airborne oxygen concentrator. In this paper, the Shannon entropy of the data is extracted by using the data-driven method. Then the SKF filter is used to process the real-time data samples. The current degradation mode is identified according to the posterior probability between the steady-state degradation filter and the accelerated degradation filter, the recognition results are consistent with the actual situation. Finally, compared with wavelet decomposition and K-means algorithm, the effectiveness of Entropy-SKF algorithm is proved.

Abstract:This paper addresses the tracking control problem of nonholonomic wheeled mobile robot with input saturation. An iterative learning control law with iterative leaning control method is designed, where the designed control law combines the tracking error of mobile robot system and the previous generation control law under constraint. Moreover, the convergence of the tracking error is analyzed by using the norm theory, and then the validity of the designed control law is verified. Finally, this paper gives a simulation example to prove the effectiveness of the theoretical analysis. The simulation results show that the nonholonomic wheeled mobile robot with input saturation can obtain good tracking control performance under the designed iterative learning control law, and the tracking error can converge to a small neighborhood of zero.

Abstract:In order to optimize the problems of high-frequency chattering, long response time, and large estimation errors of traditional sliding mode observers (SMO), this paper proposes an improved SMO. The sign function in the traditional SMO is replaced with a new piecewise exponential function to reduce the high-frequency chattering of the system, the arctangent algorithm with large errors is discarded, and the phase-locked loop is used to estimate the IPMSM speed and rotor position information. The Lyapunov stability criterion was used to prove the stability of the improved SMO, and the improved SMO sensorless control strategy model based on the interior permanent magnet synchronous motor was built through Matlab / Simulink, and simulation analysis was performed. The simulation results show that compared with the traditional SMO, the rotor position estimation error of the improved SMO is optimized from 0. 05 to 0. 025 rad, the system speed estimation response time is reduced by 50%, and the speed estimation error is reduced by 85%. It is proved that the improved SMO has higher dynamic performance and estimation accuracy.

Abstract:This paper proposed an adaptive terminal sliding mode control strategy for disturbance rejection control of a rotary-wing flight robot equipped with a 2-DOF cable-driven manipulator. The system has been divided into quadrotor subsystem and manipulator subsystem. The dynamical model of quadrotor has been obtained by Lagrange method. And the dynamical model of manipulator has been deduced by Newton-Euler equation. Then, the dynamical model of quadrotor has been simplified in hovering. And the motion of the manipulator which is regarded as perturbed force and moment is added into the dynamical of quadrotor. The effectiveness of the proposed controller is tested through three simulation case. The results show that the proposed controller is superior to other controllers in terms of tracking accuracy, robustness and disturbance rejection. The control performances of x and y channels are sensitive to the manipulator motion. The proposed controller has a good performance for the rotary-wing flight robot in hovering, which has an engineering referenced value.

Abstract:Aiming at the problem of small amount of original fault data and unbalanced data set of bearing rollers in actual production, a data enhancement strategy was proposed to expand the original bearing image data set, and combined with the U-Net framework and lightweight deep learning model to construct an end-to-end bearing roller semantic segmentation model method. By combining the U-Net framework and lightweight deep learning models MobileNetV1 and DenseNet121, the end-to-end bearing roller semantic segmentation models LS-MobileNetV1 and LS-DenseNet121 are constructed,the proposed models are trained based on the transfer learning strategy, and compared with other models for experimental analysis. The results show that compared with the existing methods, the method in this paper achieves higher segmentation accuracy and more robust detection results with few parameters, which verifies the effectiveness of the proposed method.

Abstract:The operating status of submarine cables directly affects the safety and reliability of offshore engineering. It is of significance to timely and effectively obtain the health status of submarine cables. Due to the complex structure and operating environment of submarine cables, it is difficult to accurately obtain the health status of submarine cables. Therefore, a comprehensive health status assessment method of submarine cables with online monitoring data is proposed. Firstly, following technical guidelines of cable status evaluation, an index system suitable for submarine cable comprehensive health status evaluation is constructed by integrating static data, such as patrol inspection and testing data and dynamic data from online monitoring. Secondly, a set of reasonable and evidence-based evaluation criteria is formed based on the characteristics of online monitoring status of submarine cable and evaluation standards from cable guidelines. Then, on the basis of the above, a multi-state fusion based status evaluation model is established, which can evaluate the overall submarine cable as well as its components hierarchically to obtain its health status. Further leveraging fuzzy theory, a comprehensive health status of the submarine cable can be obtained. Finally, the feasibility and effectiveness of the proposed method is verified by applying on submarine cables in a practical offshore oil and gas project.

Abstract:In this paper, a water-target recognition algorithm based on the data acquired by the onboard visual sensor from unmanned surface vessels (USV) is reported, in order to satisfy the accuracy and speed requirements of USV intelligent sensing system. The main outcome are summarized as follows: First, images are collected based on open source datasets and experimental data, to create a watertarget recognition database which named YZ10K; second, popular deep-learning based target detection methods including Faster RCNN, SSD, YOLOv3, etc. are implemented and compared; third, based on the characteristics of water targets, an enhanced lightweight Water Target detection network WT-YOLO (water target-YOLO) is proposed. The experimental verification shows that the WT-YOLO algorithm based on improved YOLOv3 has achieved accurate and real-time target recognition with the mean average precision (mAP) of 79. 30% and frame per second of 30. 01.

Abstract:Aiming at the trajectory tracking control of the palletizing manipulator, a nonlinear sliding mode control method is proposed. Firstly, the Newton-Euler method is used to derive the dynamic model of the palletizing manipulator, which is then equivalent to a discrete form. Then, the nonlinear terminal sliding mode surface is designed to accelerate the convergence speed of the joint angles, and the equivalent reaching law is introduced to suppress the chattering effect of the system. At the same time, the problem of parameter tuning of the controller is solved with the help of the glowworm swarm optimization algorithm which is used to obtain the optimal control performance. Finally, the effectiveness of the proposed control method in this paper is verified by simulation and experiment. The results show that the control method in this paper has higher tracking accuracy than integral sliding mode control, and can ensure that the palletizing manipulator can better track the upper reference trajectory, which has certain engineering application value.

Abstract:Computed tomography angiography(CTA), as a non-invasive detection with higher accuracy auxiliary diagnostic method, now is urgently needed to effectively eliminate the interference noise near the coronary artery target and to find a new algorithm that can fully automatic, fast and accurate tracking the target, so as to greatly reduce the pressure on doctors to read the film and assist them in reliable diagnosis and treatment. A new minimum output sum of squared error (MOSSE) algorithm was proposed to achieve automatic accurate and fast tracking of coronary targets by extracting multiple features of coronary arteries and incorporating them into the existing MOSSE tracking method. CTA data from 9 patients (5 males and 4 females, average age 65, 6 with atherosclerosis) in Affiliated Hospital of Hebei University were used to verify the algorithm, and the results were compared with existing coronary target extraction algorithms based on centerline and regional growth. Results show that the new algorithm processing track patient frame data only takes 0. 02 s, the average accuracy of multiple cases was 94. 30%, and the performance is better than the existing coronary target extraction algorithm, it realizes automatic accurate efficient tracking to form severe coronary target change, and provides more efficient assistance to the clinical diagnosis and treatment of coronary heart disease (CHD).

Abstract:Aiming at the online fault diagnosis of power devices in the switched reluctance drive, an on-line diagnosis method of transistors short-circuit and open-circuit fault is proposed based on DC component, spectral ratio eigenvalue and speed ripple coefficient ratio, taking the asymmetric half-bridge power converter as the research object. According to the characteristics that the DC component in the phase current tends to be zero after an open circuit fault occurs in the power converter, and the DC component spectrum increases rapidly after a short circuit fault occurs, the calculation method is improved on the basis of diagnosing the short-circuit fault by using the relative spectral ratio coefficient. The concept of the characteristic value of the spectral ratio is introduced, and the amplitude of the DC component and the characteristic value of the spectral ratio are used to detect the fault type, avoiding the adverse effect of a large number of calculations on the online diagnosis. According to the different characteristics of the speed pulsation change after the short circuit fault of different transistors in the same phase, the speed pulsation coefficient is applied to the fault diagnosis, and the ratio of the speed pulsation coefficient is used as speed ripple characteristic value, the change of the speed ripple characteristic value is used to locate the faulty component. This method can quickly detect the fault type, fault phase and locate the faulty component, which solves the shortcomings of the current online diagnosis method, such as long response time, large amount of calculation, and hard to find faulty transistor. Finally, simulations and experiments verify the feasibility and effectiveness of the method.

Abstract:In the actual application of the digital spectrometer, the pulse signal output by the detector can be approximately regarded as a double exponential pulse. Due to the charge collection time of the detector, the collected pulse amplitude has ballistic loss, which is not conducive to the subsequent accurate amplitude extraction and affects the energy resolution of detection system. In this paper, the trapezoidal forming of ideal double exponential pulse signal is simulated in MATLAB, and the model is built on Simulink platform, and the feasibility is tested by ideal double exponential pulse sequence. Use HDLCoder to convert Simulink logic model into VerilogHDL code and test it on ModelSim. Build the hardware platform test code and apply the hardware platform to the processing of the measured data, send single-peak and multi-peak pulse tests through the serial port, and use the embedded logic analyzer to capture the forming data and output the formed data observation through the serial port. The trapezoidal algorithm of exponential pulse signal has been well implemented in FPGA, and the recovery error is within 0. 2%. The implemented hardware platform has a good shaping effect on pulses with ballistic losses and a certain noise filtering ability, which can be used as a reference for the design and development of digital multichannel pulse amplitude analyzers.

Abstract:Nowadays, the mainstream method of LED defects detection is low-efficiency manual visual inspection. And the traditional machine visual detection cannot meet its application’s standard with its low precision. To deal with these issues, a LED defects detection method is proposed based on improved faster R-CNN network framework. In order to improve the robustness and generalization capability of the network, the dataset is expanded by adding noise and changing the brightness. Resnet50 and FPN network are the backbone network to extract the characteristics, and the anchor scale is adjusted according to the characteristics of its different feature prediction layers of the feature pyramid, to construct and train the network. Eventually, the quantitative analysis of the test results on the dataset testing shows that the method of LED bubble-like defects detection can achieve an overall accuracy of 95. 6%, and the recall rate has a 20. 8% increase. Single-picture detection time is about 100 ms. It can be affirmed that this method can meet the needs of the automatic detection in manufacture.

Abstract:By combining the advantages of eddy current pulse excitation and infrared thermography, eddy current pulsed thermography (ECPT) has been used for testing and evaluating defects of key components in many areas. This paper focuses on using ECPT to test rolling contact fatigue (RCF) cracks and quantify their surface lengths. Due to the complex surface shapes of RCF cracks, a crack shape extraction procedure is designed which mainly includes crack response extraction and crack identification. Fitting parameters are utilized to evaluate the performances of four methods used in the crack response extraction step. Experimental results show that PCA-based thermal pattern in the early heating stage has the best performance for surface length quantification with its R 2 , sensitivity, and 2-norm of residuals of 92. 8%, 5. 91, and 2. 24, respectively.

Abstract:In view of the abundant vegetation and mountainous hills in the plateau and mountainous areas, which cause serious path loss to communication users, increase the energy consumption of communication systems. This paper proposes an optimization algorithm based on energy efficiency (EE). First, the multi-peak diffraction model is used to simulate the transmission loss of radio waves in the plateau and mountainous areas. Under the premise of ensuring the quality of server (QoS) for communication users, the total energy efficiency of the D2D user equipment in the system is maximized, and the optimization problem is decomposed into power. Two subproblems of control and channel selection are solved. Secondly, the Lagrangian duality and Dinkbach algorithm are used to jointly solve the power control. Finally, on the basis of power control, the Gale-Shapley matching algorithm is used to achieve channel matching. The simulation results show that compared with the existing algorithms, the proposed algorithm can effectively restrain co-channel interference, increase the utilization of channel resources in the system, improve the energy efficiency of the system, reduce system energy consumption, and improve system performance.

Abstract:Because the optimization algorithm of support vector machine ( SVM) falls into local optimum easily and has many control parameters, a SVM optimized by surface-simplex swarm evolution (SSSE) algorithm is proposed and the classification of Motor imagery EEG signals is studied. The fuzzy entropy and AR model parameters of MI EEG signals were extracted as input features, and then SSSE is applied to parameters optimization of SVM to classify MI EEG signals. In the test experiment, which classified the 2003 international brain-computer interface (BCI) competition Data sets Ⅲ and the 2008 BCI competition Data sets 2b by left-hand and right-hand. The results showed that the average classification accuracy and Kappa value of the proposed method were 82. 47% and 0. 88 respectively. SSSE reduced the control parameters and effectively avoided the particles falling into the local optimum. The validity of this method in the classification of MI EEG signals was verified.

Abstract:With increase of the dynamic force test range and the improvement of test accuracy requirements, the key role of piezoelectric test technology becomes gradually prominent, which has been widely used in heavy-duty equipment and aerospace vector force measurement. In the process of assembling and applying preload of the piezoelectric three-way force measuring unit, problems such as the deflection of the upper plate and lower plate and the different axis incurred. Based on the load sharing principle of force, this paper quantifies the error and establishes a theoretical model of the self-preload error of the force measurement unit. A new type of assembly fixture is designed, and the angle deflection test experiment and static calibration experiment of the force measuring unit assembled with two types of fixtures are carried out. The experimental results show that the test accuracy of the force measurement unit assembled with the new fixture has been significantly improved, and the maximum value of interphase interference is reduced from 6. 06% to 2. 97%, which provides theoretical support for the improvement of the accuracy of the force measurement unit and the development of related structures.

Abstract:The metallic components are normally subject to such structural anomalies as corrosion and cracks, etc. during fabrication and practical service. These flaws have posed a severe threat to structural integrity and safety of engineering apparatus. In consequence, it is imperative to non-invasively inspect and evaluate the metallic components via efficient non-destructive evaluation techniques. Based on the conventional structure of the eddy current probe, in this paper a ferrite-cored funnel probe is proposed in an effort to further enhance the testing sensitivity in pulse-modulation eddy current testing of surface and back surface corrosion. The analytical model and closedform expressions of testing signals from the proposed probe have been established for investigation of the testing performance. Following the identification of the higher sensitivity of the proposed probe to surface and back surface corrosion in metallic components, a series of experiments are conducted. Through the experiments, the validity of the conclusion of simulation is affirmed. Further experiments demonstrate the enhancement of testing sensitivity in respect to localized defect on surface and back surface.

Abstract:Aiming at the problem that the temporal resolution of correlation method in ultrasonic water meter can only reach the sampling period, a time delay estimation method based on correlation method with low sampling rate is proposed. Firstly, the maximum time delay in ultrasonic water meter is studied. Secondly, the specific application of correlation method in ultrasonic water meter and the function of interpolation algorithm to improve the accuracy of flow measurement are discussed. Finally, MATLAB software is used to simulate the algorithm and evaluate the estimation accuracy of the algorithm under the condition of actual noise, and the flow measurement system is built to prove the effectiveness of the algorithm. The simulation results show that the average absolute error of time delay estimation is within 100 picoseconds when the frequency of ultrasonic transducer is 1 MHz and the sampling frequency is 4 MHz. It is shown that the proposed algorithm can achieve high precision delay estimation at low sampling frequency and is superior to the traditional correlation method in the accuracy of delay estimation.

Abstract:In order to solve the misdiagnosis of encephalitis and glioma in clinical diagnosis while using MRI images, we proposed a classification method of convolutional layer feature transfer combined with active sample labeling. The method firstly adopts the convolutional layer features parameter transfer and uses the multi-modal MRI image data for the fine-tuning of models to verify the distinguishing ability of different MRI modal features. Secondly, in view of the difficulty of sample labeling, an entropy uncertainty based active labeling algorithm is designed to extract the uncertainty information of samples to further improve the convergence speed and generalization ability of the model. Experiments were carried out on a dataset of 175 cases (118 cases of encephalitis and 57 cases of glioma) collected by the radiology department of the First Affiliated Hospital of Chongqing Medical University. The results show that the classification accuracy under cross-validation reached 95. 08% and area under the curve reached 0. 98. The accuracy of the model was superior to the method mainly relying on the experience of doctors at present; and the accuracy and area under the curve was 17. 51% and 0. 15 higher than that of doctors, respectively. At the same time, only 30% of the data samples need to be annotated, so the model can achieve optimal performance, reduce a lot of data annotation work, and provide meaningful guidance for the initial diagnosis.

Abstract:Ph-ADC uses quadrature IQ channels to extract phase information, but IQ channel offset errors will cause the system’ s bit error rate (BER) to increase. Aiming at the above problems, a method of extracting I/ Q offset error based on frame sampling is proposed, and the trapezoidal integration method is used to compensate. Relative to the traditional method, this method can save time wasted due to memory access, data delay, and the system’s interrupt response to each sample. The proposed phase domain ADC offset error detection and compensation technology is verified by establishing a π/ 4 DQPSK demodulation, 6-bit Ph-ADC, Eb / N0 of 12 dB digital modulation system. The simulation results show that when the input signal frequency is 450 kHz, the I/ Q offset error is 10%, the signal-to-noise distortion ratio (SNDR) of the system is increased from 7. 02 to 37. 22 dB, the spurious-free dynamic range (SFDR) of the system is increased from 17. 37 to 38. 74 dB, and the ENOB is increased from 1. 03. After calibration, this method can reduce the BER of the system to the order of 10 -5 and make the error vector magnitude (EVM) less than 15 dB.

Abstract:The tilt of the tower will cause serious damage to the entire power grid and threaten the lives of surrounding residents. The power inspection performed by the computer vision of the UAVs not only saves labour, but also significantly improves the inspection efficiency of the power grid. In order to get early warning before the tower falls for State Grid inspectors. In this paper, the algorithm of computer vision-based tower tilt detection in electric patrol unmanned aerial vehicles is researched. And the tilt of tower is detected using YOLOv3’s deep neural network combined with LSD line segment extraction method. Using the pole pictures of the actual inspection of the UAVs in Shanxi power grid to make the VOC2007 dataset of the pole tower and use the YOLOv3 neural network to detect the pole tower. The Bounding box obtained after the detection is fine-tuned according to the mIOU parameters after network training and used as LSD detection ROI, the detected line segment is filtered and fused, and the secondary identification of the tower is performed according to the characteristics of the tower. Finally, the outer line of the tower is used to make the center line of the tower in this direction and the inclination of the tower in this direction is calculated. The experiment uses the data provided by Shanxi State Grid Electric Power Company for verification. The tilt detection effect of the tower is more accurate under various backgrounds, and the accuracy and environmental adaptability are significantly improved compared with other algorithms. The correct recognition rate of the tower target reaches 97%, and the average error of the inclination detection is less than 0. 85°.

Abstract:The establishment of flying ad hoc networks (FANETs) can effectively meet the networking communication requirements of UAV cluster operation, but compared with the traditional mobile ad hoc network, flying ad hoc network has the characteristics of faster node movement and dynamic network topology. To solve the problem of network performance degradation caused by frequent link disconnection, this paper proposes a high stability routing protocol based on load sensing and pseudo gossip mechanism. Network congestion and node mobility are two main factors affecting path stability. In order to solve the problem of network congestion, a pseudo rumor limited flooding mechanism based on node load prediction is proposed. In the process of initiating routing requests, the node forwarding probability is dynamically adjusted to balance the network load and reduce the control overhead. Aiming at the high mobility characteristics of UAV nodes, a joint measurement index of link stability is proposed based on the received packet signal power strength. Simulation results show that, compared with AODV protocol and other improved protocols, this protocol can effectively reduce the network control overhead, improve the packet delivery rate, reduce the average end-to-end delay, significantly improve the network performance, and enhance the real-time and reliability of data transmission.

Abstract:In the process of preparation and long-term service, CFRP / Al honeycomb structure is prone to debonding, delamination, ponding and other defects, so it is very important to detect its state by infrared thermal wave nondestructive testing technology. And In the process of collecting the infrared thermal image sequence of CFRP / Al honeycomb structure defects, there is a large background noise, which easily leads to low detection efficiency and poor contrast. In order to improve the defect detection effect, principal component analysis (PCA) algorithm is used to reduce the dimension of defect feature information of the infrared image sequence after background removal, which can effectively filter out the uneven background noise in the infrared image sequence. Combined with multistructure morphology and PCNN hybrid algorithm, the defect area is extracted by image enhancement and image segmentation. The experimental results show that the proposed method can further filter out the uneven background noise of the infrared image, improve the defect area extraction effect, and effectively improve the defect detection rate.

Abstract:In order to solve the problem as non-closed contour and non-uniform edge of reconstruction results in convex total variational regularization model, a mode of image deblurring based on hybrid non-convex second-order total variation and overlapping group sparse is proposed. Overlapping group sparse regularization item well considering the cross relationship between adjacent elements, non convexity second-order l p norm regularization item better keep the edge of the image shape information, and the two regular constraint into total variation method at the same time, which can accurately restore edge structure characteristics and eliminate the staircase effect and smooth ringing effect. Finally, in order to achieve the optimal solution of the non-convex higher-order model, the variable splitting method is proposed to separate the model into four sub-problems, and then the method of a re-weighted l 1 alternating direction method is used to complete the calculation of image deblurring. The test data show that compared with the current image restoration technology, the proposed algorithm has better deblurring effect, the restored image shows higher peak signal-to-noise ratio and structural similarity, which can recover edge shape information and texture details more effectively.