Abstract:The lithography machine represents the pinnacle of advanced equipment. The ultra-precision measurement is crucial to supporting the research, development, and manufacturing of these products, ensuring the precision and quality of the manufacturing process. This article provides an overview of the lithography industry′s characteristics and development trends, and explores the role of ultra-precision measurement in advancing lithography machine technology, examining its impact on components, subsystems, whole machine integration, and overall performance. The article also considers ways to enhance the precision capabilities of lithography machine products and the need to establish an ultra-precision measurement system, including an industrial measurement system to control the quality of ultra-precision lithography machine products and a metrology system to ensure measurement accuracy and reliability. Finally, the article proposes the necessity of establishing a measurement and testing center for the lithography machine industry.

Abstract:Profile of the path of contact ( PPC), as a characteristic curve on the gear tooth surface, is with specific engineering significance. It can reflect the quality of gear machining and gear transmission. The PPC is not only the generative curve in gear hobbing, worm grinding and other generative machining, but also the working curve in the involute gear transmission. However, there is no function to measure the deviation of PPC for the existing gear measuring instruments. The theoretical model of the PPC is presented, which is based on the gear geometry. By means of the existing gear measuring center, the four-coordinate method and the threecoordinate method for measuring the deviation of PPC are proposed. The measurement practice indicates that the measurement and evaluation of the PPC can be realized by using the existing gear measuring instruments. The form deviation, slope deviation and total deviation of the PPC measured by the four-coordinate method are 0. 2, 1. 3, and 0. 6 μm, which are different from those measured by the three-coordinate method. Compared with the gear involute and helix, the PPC has features of comprehensiveness, unity, and uniqueness. The optimal deviation of PPC is obtained through synthesizing the involute deviation and helix deviation of the gear, which effectively reduces accuracy requirements for the gear involute and helix.

Abstract:To quickly and intuitively detect the small offset of periodic structures, a 4f optical system-based method for detecting the small offset of periodic structures is proposed. Firstly, the VirtualLab Fusion optical simulation software is used for the theoretical study to formulate the model of periodic microstructure with preset offset, construct the optical transfer function, and obtain the image plane magnitude map corresponding to the periodic microstructure by using the 4f spatial filtering method. The analysis shows that in periodic structures of transparent substrates, regardless of size, if the offset is within 80% of the spacing between adjacent feature sizes, the fitted amplitude change is linear in relation to the microstructure offset and the location of the amplitude change is consistent with the location of the microstructure offset. The experimental system is established based on the simulated optical system parameters, and experimental results are consistent with the simulation. The system can achieve a measurement field of view of 3. 4 mm×2. 6 mm and a resolution of 5 μm, which is capable of real-time and rapid displacement or defect detection of periodic structural materials.

Abstract:It is difficult to realize high efficiency of axial scanning differential confocal measurement (ASDCM). In this article, a differential confocal profile tracking measurement method based on Kalman prediction is proposed. In this method, the linear range of hundreds of nanometers of laser differential confocal axial response curve is used for high-precision linear sensing measurement of the continuous surface profile, which improves the measurement efficiency. Meanwhile, the Kalman predictor profile tracking method is introduced to predict and track the unmeasured surface using the measured profile point data, which expands the range of linear sensing profile measurement. Compared with the ASDCM, experimental results show that the measurement efficiency of this method is improved by 8 times, the high-precision tracking measurement of the standard elliptical column with the PV value of the outer profile is greater than the linear sensing measurement range, and the repeated measurement standard deviation of the roundness of the laser inertial confinement fusion capsule is 3 nm. It provides a high quality method for high precision, fast and nondestructive measurement of continuous surface profile of rotary precision components.

Abstract:Traditional structured-light (SL) measurement methods face challenges in measuring surface with high reflectivity and complex structure. The measurement accuracy is limited by the measurement field of view. This article proposes a high-precision calibration and 3D measurement method based on structured-light projection and multi-view Scheimpflug imaging, which can make full use of the depth range of the system. A novel structured-light multi-view stereo measurement model is proposed. The projector coordinate system is used as the measurement coordinate system, and the multi-view structured-light system is calibrated in an integrated way by establishing the corresponding relationship of " 3D point-projection image point-4 camera image points" . With the use of multi-view geometric imaging constraints, the observation information of multiple perspectives is fused by calculating the least square solution to improve the calculation accuracy of 3D data. Experimental results show that the proposed method and system can accurately measure high reflectivity and occluded surfaces, with a measurement accuracy of 5 μm, which is substantially superior to the traditional SL methods.

Abstract:In this article, the structure of the TM80 gas turbine flowmeter is optimized by using numerical simulation and experimental test. The numerical simulation results reveal that the sudden change of pressure gradient and boundary layer separation are mainly caused by the core support and rear deflector. Therefore, the structural optimization schemes about the slope of core support and the diameter of rear deflector are proposed. Specifically, the slope of the movement support is designed to be 15°. The step flow on the side of the movement support is transformed into a gradual flow, the diameter of the rear guide fluid is reduced to 62 mm, and the step flow on the side of the rear guide fluid is transformed into a pipe flow of equal diameter. The numerical and experimental results show that when the slope of core support is 15° and the diameter of rear deflector is 62 mm, the pressure loss of the flowmeter is significantly reduced, the instrument coefficient becomes more stable, and the linearity error becomes obviously smaller, which verify that the structural optimization schemes can greatly improve the metering performance of the flowmeter. The research results are helpful to provide theoretical guidance and technical support for the development of gas turbine flowmeter with better performance in the future.

Abstract:The PEMFC driven unmanned aerial vehicle (UAV) is developing rapidly. However, the UAV has a compact structure and limited air source capacity. It is an urgent problem to improve the utilization rate of hydrogen that will further enhance the UAV endurance. In this paper, a hydrogen recirculation system based on the ejector is designed to recycle the drained hydrogen to improve the utilization rate of fuel. For the 1. 7 kW fuel cell used in UAV, a hydrogen cycle ejector is designed by using the computational fluid dynamics method, and the performance analysis is implemented to reveal the internal flow field characteristics under different working conditions. The results show that the primary flow pressure has a good ejecting performance in the range of 300 ~ 700 kPa when the pressure difference between the secondary flow pressure and the back pressure is 10 kPa. Then, the influence of key structural parameters of area ratio (AR) and different working conditions on the ejecting performance is studied. The research shows that the optimal AR varies with the change of primary flow pressure, and the AR is selected as 16 in this paper to meet the global optimal ejecting performance under different working conditions. The maximum hydrogen utilization rate of the proposed system is increased by 30. 3% outperforming that of the traditional UAV′s system.

Abstract:To ensure the flight safety of unmanned aerial vehicles (UAV), the airborne intelligent condition monitoring has received a lot of attention. However, constrained by airborne computing resources, multi-task scenarios that monitor multiple key parameters pose a greater challenge to airborne real-time computing. To address this issue, this article proposes an airborne intelligent condition monitoring unit with multi-task scheduling optimization under limited field programmable gate array ( FPGA) resource. Firstly, the monitoring models of different scales corresponding to different monitoring parameters are established based on stacked long short-term memory, and the custom computing acceleration units for each model are constructed by using FPGA. Secondly, a multi-task scheduling optimization method with joint constraints of FPGA resources and model computing time is proposed to obtain customized computing acceleration unit deployment and computing scheduling strategies, which minimizes the completion time of all tasks. Finally, according to the above strategies, the acceleration units of specified scales are deployed in the airborne computing platform to complete multi-task scheduling and computing. Real flight data are used to verify the proposed method. The results show that the unit can efficiently perform real-time calculation for multiple condition monitoring tasks. The results show that the unit can efficiently perform real-time calculation for multiple condition monitoring tasks. Comparing with other FPGA-based computing methods, theefficiency is improved by 16. 08% .

Abstract:To accurately locate the leak in the buried gas pipes, accelerometers are mounted on the ground to detect the leak-induced vibroacoustic waves. A new method using nonuniform L-shape array is proposed to jointly estimate the 3D coordinate and wave-speed. A small-aperture L-shape subarray is embedded into the large-aperture L-shape array and is combined with the Root-MUSIC algorithm to find the relationship between DOA of far-field source and wave-speed. Then, the large-aperture L-shape array is used to perform 2D spatial spectrum estimation for wave-speed and near-field source distance. Finally, the 3D coordinate is calculated using the estimation results and the computational complexity is reduced. The results of localization experiments show a superiority performance of the used L-shaped array. The joint estimation method based on the L-shaped array can estimate the propagation speeds of leak-induced vibroacoustic waves in multiple soil conditions and the 3D coordinate of leak source is derived. Compared with the localization method using theoretical wave-speeds, the error in joint estimation method is reduced by 90. 9% , which avoids the influence of error in theoretical wave-speed on localization. Compared with the direct 3D spatial spectrum estimation, the calculation time of the joint estimation method is reduced by 88% and the localization accuracy remains well. The range of signal source, rmin ~ rmax, is calculated by using the far-and near-filed criteria. Experiments evaluate that accurate localization can be achieved within this range.

Abstract:To achieve the online monitoring of the buckling deformation of the beam structure, a shape sensing method for monitoring the buckling deformation of FRP composite beam structure is proposed. Firstly, according to the higher order shear deformation theory, a method for describing the deformation field of laminate composite beam structure is proposed. Based on the von Karman strain gradient theory, the section strain expressions are derived. Then, the least square function is employed to formulate the displacement reconstruction model. In this model, the displacement interpolation functions are constructed by using the quartic non-uniform rational Bsplines (NURBS) basis function as the shape function, and the theoretical section strain field formulation is derived. Based on a small number of discrete point strain measurement values, a nonlinear strain decoupling method is proposed, and the transformation relationship between surface strains and measured section strains is established. Finally, to evaluate the accuracy and effectiveness of the proposed method, a simply supported beam test platform is established with 25 layer carbon fiber composite beams as samples, and its numerical calculation and experimental demonstration are carried out. Results show that the proposed model and the reconstruction displacement error is less than 8% under different axial loads.

Abstract:The accuracy improvement of ultra-short-term wind power forecast with deep learning methods is of great significance for intraday unit commitment, ultra-short-term economic dispatch, and reserve scheduling of the power systems, which can further enhance the safety and efficiency. To address the problem in the existing feature extraction models that the similarity of implicit features and changing trends in time series curves have not been adequately extracted, this article proposes an ultra-short-term wind power forecast model based on contrastive learning-assisted training, which mainly consists of an input encoding module, a feature extraction module, a contrastive learning module, and a regression module. The self-supervised contrastive learning module autonomously generates positive and negative samples and enlarges the distance between the positive and negative samples in the projection space, which help to extract the implicit features of the similarity of the input information. In this way, the redundant information is reduced, the sample correlation is enhanced, and the accuracy of wind power forecast is ultimately improved. Compared with LSTM and Lightgbm methods, experimental results show that the mean absolute error of the proposed method is decreased by 19. 9% and 6. 5% , which effectively increase the wind power prediction accuracy.

Abstract:In view of the problems of complex background, small target and inconspicuous fine features of aero-engine bolts, an automated detection algorithm of aero-engine bolt installation detection based on key points detection is proposed. First, a cascaded convolutional neural network based on the Faster RCNN and the improved CPN (AD-CPN) is proposed to achieve the detection of bolts and 2D key points which can determine whether the bolt has fallen off or missed. To further detect the 3D installation detection of the bolt, the Euclidean distance selection strategy is introduced to match and screen the key points to obtain the detected point pairs. Finally, the 3D coordinates of the key points are calculated by using binocular stereo vision technology. In this way, it can judge whether the bolt is wrongly installed. Compared with CPN, the mAP,AP50 , and AP75 of AD-CPN are improved by 2. 9% , 3. 3% , and 4% , respectively. In addition, the relative average error of bolt measurement length is approximately 3. 0% . It can be seen that the algorithm could enhance the accuracy of detection, and ensure the safe operation of aero-engines, which has great practical significance.

Abstract:The composite defect existing together with the stress is one of the important factors which affects the safe operation of pipeline. The dual magnetic field method can be used to judge the degree of stress damage at composite defects. But the stress signal extraction method needs to be solved urgently. In this article, the magnetomechanical relationship in the J-A theory is introduced into the magnetic charge model. The magnetic signals of composite defects under different stress and external magnetic field are analyzed and calculated. The stress signal extraction model of composite defects based on the ratio of dual magnetic field signals is formulated. The ratio factor is proposed to evaluate the stress level at the defect, and the systematic experimental evaluation is carried out. The results show that the strong magnetic signal is not sensitive to the stress at the defect, and the signal mainly includes the defect size information. The weak magnetic signal is sensitive to the stress at the defect, and the signal includes defect size information and stress information at the defect. The ratio factor can reflect the stress at the defect. When the weak magnetic field intensity is low, the average change rate of the ratio factor with the stress at the defect is larger than 9% . With the increase of weak magnetic field intensity, the variation range of ratio factor with stress decreases.

Abstract:To ensure the accuracy of the dynamic angle measurement results for the turntable angle measurement system, a dynamic angle measurement method based on Kalman filtering is studied in this article. The principle of dynamic angle measurement method is described, and the parameter optimization model and dynamic prediction model based on Kalman filtering are explained in detail. The dynamic angle measurement circuit is designed according to the requirements of measurement accuracy, and the real-time prediction of the turntable angle is realized based on the field programmable gate array platform. The simulation experiment of constant acceleration motion is implemented with acceleration of 10°/ s 2 , and the validity of the method is verified by taking the theoretical value of turntable angle as a reference. A turntable experimental platform is established to evaluate the application effect of the dynamic angle measurement method with the ring laser gyro as the reference value. Experimental results show that the proposed dynamic angle measurement method can effectively reduce the system error introduced by the measurement delay of the turntable angle measurement system. The error is reduced from -82. 62″ to -0. 01″ at the speed of 30°/ s, the consistency of measurement results can be maintained at various speeds, and effectively improve the dynamic measurement accuracy of the turntable angle measurement system.

Abstract:Analog circuit is an essential part of the integrated circuit. One of the current research hotspots in integrated circuit testing is the detection of faults occurring in analog circuits and the accurate identification of fault types based on deep learning techniques. To address the difficulties in fault detection of analog integrated circuits, the advanced achievements of artificial intelligence in the field of image recognition and speech classification is referenced and an analog circuit fault detection idea based on a deep learning algorithm of self-attention mechanism is proposed, which can be used to detect faults in Sallen-Key low-pass filter circuits. The output signal is sampled into an audio signal and fed into an audio classification model based on a self-attentive transform network for training, testing, and optimization. The results show that fault detection based on the self-attentive mechanism audio classification has an average accuracy of 93. 1% and a maximum accuracy of 98. 1% . Nine different fault types can be detected. The model converges fast and can detect faults in analog circuits, which thoroughly verifies the feasibility of the proposed idea.

Abstract:For the application of capacitance-coupled non-contact voltage measurement, the coupling capacitance between probe and wire is affected by wire diameter, wire insulation material and relative position deviation. It is difficult to determine the voltage division relation and the voltage cannot be reconstructed. In this article, a self-calibration method of non-contact voltage measurement based on impedance transformation is proposed to realize the self-calibration of sensor gain in practical measurement. Firstly, the basic principle of capacitive coupling voltage measurement is introduced, from which the problems exist and the self-calibration method based on impedance transformation is proposed. Then, the influence of calibration accuracy on system parameters is analyzed by simulation and the principle of parameter selection is given. On this basis, the sensor probe and the circuit topology are developed. Finally, the accuracy test, anti-interference ability test and scene adaptability test of the sensor prototype are carried out. The accuracy test shows that the maximum relative error of voltage amplitude is 0. 59% , and the phase relative error is 0. 76% . The anti-jamming ability test shows that the coaxial probe has good shielding effect on the surrounding coupling electric field. The scenario adaptability test shows that the maximum relative error is 1. 24% when different types of circuits are tested.

Abstract:To analyze the vibration characteristic and output characteristic of the arc spiral piezoelectric energy harvester, a finite element approximation (FEA) model is proposed in this article. It is found that the FEA model can simplify the model while maintaining the accuracy of results. The modeling, simulation and measurement of the arc spiral piezoelectric energy harvester are implemented before and after the approximation. The resonant frequency, output voltage, and output power of the spiral piezoelectric energy harvester are obtained. The circular arc cantilever beam is divided into 6 ~ 16 rectangular cantilever beams with the finite element method. The resonant frequency and output performance of the arc spiral energy harvester under different degrees of FEA are analyzed. Results show that when the number of FEA sides is larger than 10, the performance of the spiral piezoelectric energy harvester has little error. A 2π spiral piezoelectric energy harvester is fabricated and measured. The test results show that the resonant frequency is 158 Hz, the theoretical resonant frequency is 153 Hz with the 10 FEA model, and the error is 3. 5% . The maximum output power is 53. 5 μW, the theoretical output power is 55. 2 μW, and the relative error is 3. 18% . The test results is in good agreement with the theoretical results.

Abstract:The complexity of current IC test is increasing, which leads to rising test cost. To address this issue, a reliable and effective test set optimization method is proposed in this article. The features in the original test set are filtered by clustering with K-means clustering. Then, a modified mRMR algorithm is used to introduce redundancy weighting factors among features by using a segmented formula to weigh the feature relevance and redundancy metrics. Meanwhile, the SVM cross-validation is inserted to reinforce the accuracy of the test pattern selection. The reduction in the number of dimensions of the original test set is achieved while ensuring that the fault coverage remains largely unchanged. Experiments on the ISCAS89 circuit show that the proposed method takes the test patterns of the original test set and streamlines them considerably. This method ensures test quality and also optimizes the test set. The test flow after redundancy elimination and sequencing reduces test time by 40. 43% , improves test efficiency and reduces test cost.

Abstract:The traditional active equalization topologies often require a large number of switches, inductors, transformers or bidirectional switches to transfer energy, which undoubtedly increase the system size and cost significantly for electric vehicles. In view of this, this article proposes a voltage equalization topology based on voltage multiplier for a hybrid energy storage system consisted of lithium-ion batteries and supercapacitors. The battery equalization topology and the supercapacitor charger are integrated into a circuit containing only two switches, three inductors, and several capacitors and diodes by sharing a half-bridge inverter circuit, which can significantly reduce the size and cost of the system. The battery equalization topology has a self-modularization feature, leading to faster voltage equalization, while the supercapacitor charger has a constant current charging capability. In this article, the operation modes and waveforms are described in detail and the equalization characteristics of the proposed topology is verified by using PSIM simulation. Finally, a hybrid array consisted of 4 batteries and 4 supercapacitors was designed for experimental verification. The experimental results show that the voltage gap is reduced to less than 10 mV after equalization, proving the effectiveness of the proposed method.

Abstract:To solve the problem that it is difficult to balance the therapeutic effect and the pleasant degree of audio in the traditional tinnitus treatment, an innovative audio processing algorithm is proposed, which considers the pleasant degree of audio and the therapeutic effect of tinnitus. The logarithmic distribution characteristics of cochlear sensing are followed. The original natural sound based on the digital filter set with fixed upper and lower limit cutoff frequencies is divided and filtered, and the original natural sound along the frequency logarithmic axis is equalized. The therapeutic sound can achieve energy balance in the full frequency range while taking into account the therapeutic effect. The results show that the audio after logarithmic energy balance processing not only conforms to the distribution of pink noise at whole frequency bands, but also significantly improves the energy at the high frequency bands of 4 001 ~ 8 000 Hz and 8 001~ 10 000 Hz. The audio energy after processing increased to 15 times and 100 times more than the original audio energy, respectively. The proposed innovative algorithm not only satisfies the pleasant habit of tinnitus-affected ears to audio sensibility, but also provides a treatment scheme for the international problem that high-frequency tinnitus has no effective therapy.

Abstract:According to the principle of foot biomechanics, a three-division method of measuring distributed force of the sole was proposed, and the corresponding distributed force measuring platform (DFP) system was developed. Twenty-seven young adults and forty elder adults were randomly selected for stability measurements in the continuous process from double-leg stance with eyes closed to double-leg stance with eyes opened and to single-leg stance with eyes opened. The foot support principle was analyzed by solving the CoP trajectories (95% confidence circle area) of three zones of the right foot and the CoP trajectories offset in three standing stages. The maximum values of the CoP trajectory range in the three partitions of the youth and elderly groups were respectively 27. 00~227. 46 mm 2 and 116. 35~387. 22 mm 2 , the maximum offsets of the anterior-posterior direction were 6. 07±3. 13 mm and 5. 88±3. 21 mm respectively, and the maximum offsets of the medial-lateral direction were 6. 51±2. 29 mm and 6. 77±2. 34 mm respectively. The results show that the CoP trajectories are concentrated on the bottom end of the calcaneus and the medial and lateral end points of the transverse arch of the metatarsophalangeal joint in both doubleleg stance and single-leg stance, and the position is less affected by the difference in postural stability.

Abstract:Micro-electro-mechanical system (MEMS) has great potential application in biomedical field. The piezoelectric drive system based on piezoelectric ceramics plays an important role in the field of high-precision control of micro-nano operation. In the practical application of various fields, the interaction between micro system and environment is always a hot research direction. Since the tracking force error at the micro scale is more non-negligible than that at the macro scale, the precise control of the contact force between the micromanipulator and the environment is the key to improve the accuracy of the micro operation. Therefore, a force tracking impedance control strategy is proposed in this article, which can accurately track the contact force during operation. Because the impedance model requires high environmental parameters, an extended Kalman filter is proposed to estimate the external environmental parameters online. Experimental results show that the proposed control method can successfully realize the force tracking control of the micro-manipulator environment interaction model and has good tracking accuracy. The average absolute error of force tracking is 7. 82 mN, and the root mean square error is 10. 16 mN. Therefore, the method is feasible to accurately track the contact force in uncertain environment.

Abstract:To solve the problem of interaction between robot and human in the human-robot integration environment, this article proposes an estimation method of gesture pointing for human-robot interaction scenes by pointing gesture to achieve the information interaction between the robot and the target point on the workplane. First, based on the RGB-D camera and the human motion capture system VICON, a time-synchronized visual pointing gesture position dataset is established. Each sample contains the RGB-D image of the pointing gesture and the true value of the pointing gesture pose. Secondly, a multi-level neural network model is formulated for estimating pointing gesture pose by combining semantic and geometric information. Thirdly, a ray approximation loss function is designed, which combines the position errorΔP and direction angle error Δθ. The pointing gesture pose estimation model is trained based on the constructed dataset. Finally, human-robot interaction experiments and model validation are implemented in the laboratory environment. In the range of 5 m from the camera, results show that the average precision of pointing gesture detection is 98. 4% , the average position error of pointing gesture pose is 34 mm, and the average angle error is 9. 94°. The average error of gesture pointing to the target point on the workplane is 0. 211 m.

Abstract:Because Lidar can directly obtain ranging information and is more robust than visual sensors to environmental changes such as illumination, the technology of laser synchronous location and mapping ( SLAM) has been widely developed in recent years. The traditional laser SLAM has made a lot of research achievements. But, it only uses geometric features, has limited understanding of the scene, and is difficult to deal with complex tasks. In addition, the current SLAM application scenarios have transited from traditional static scenes to complex dynamic scenes, and traditional methods are mostly difficult to achieve good performance due to interference of dynamic elements. Therefore, the 3D laser SLAM technology of semantic information enhancement has attracted more and more attention of researchers. The point cloud semantic tags are integrated with pure geometric features. On the one hand, the potential moving objects are filtered out with semantic information to solve the problem of static environmental assumptions. On the other hand, semantic information is used to assist the laser odometer to obtain high-precision positioning and mapping. This article summarizes the research progress of 3D laser SLAM technology for semantic information enhancement, puts forward a general framework for this technology, focuses on the outstanding research achievements and applications in this field in modules, and finally summarizes and prospects the development direction of this field.

Abstract:The penetration weapons can effectively damage the enemy′s underground works and other multi-layer structural targets. To achieve the precise control of the blast point and achieve the best damage effect, the acquisition and identification of the multilayer penetration signal are crucial. In this paper, a penetration simulation experiment method based on array sensor is proposed about several key technologies such as acquisition of penetration signal and layer identification algorithm. The interval of three impact times is controlled to be 9. 15 and 5. 41 ms. Compared with the finite element simulation results, the reliability of the multi-channel signal acquisition circuit and the feasibility of the simulation experiment method are verified. Meanwhile, to improve the accuracy of layer identification, short-time Fourier transform, wavelet transform and Winger-Vile distribution were used to analyze the experimental data at meter layer. The analysis results show that Winger-Vie algorithm can be comprehensively analyzed from multiple dimensions of time, frequency and energy, with high recognition rate and intuitiveness. However, in terms of information extraction, it still needs to improve the time-frequency analysis for optimization. The research of this paper provides an important basis for realizing high-precision penetration signal meter layer and solving signal adhesion problem.

Abstract:To address the problem of low synchronization trigger accuracy in synchronous acquisition of mechanical vibration in wireless sensor networks, a cross-layer synchronous acquisition trigger method with proportional compensation is proposed. Firstly, the ultra wide band mechanical vibration wireless sensor network nodes based on cross-layer synchronous architecture are designed, and the composition of the actual acquisition clock of the acquisition node is analyzed. Then, a hardware timer is used as the acquisition control clock to obtain the synchronization information across layers and carry out the delay of synchronous trigger. Finally, the time scale model between nodes is formulated according to the periodic synchronization information, and the delay time of synchronous trigger is proportionally compensated according to the uplink time scale of the acquisition node to reduce the synchronous trigger error. Experimental results show that the mean value and maximum value of synchronous acquisition trigger error in single-hop network are 20 and 50 ns, and the mean value and maximum value of synchronous acquisition trigger error in two-hop network are 37 and 76 ns, which effectively improve the synchronous acquisition trigger accuracy of mechanical vibration wireless sensor network.

Abstract:The mass unbalancing is one of the main defects in the manufacturing process of the hemispherical resonator, which has important impact on the vibration performance. To formulate the evaluation method of resonance mass unbalance and provide the basis for balancing, the structural characteristics and mass unbalance of the resonator are characterized. Through the coupled vibration physical model and simulation analysis, the mechanism of the coupled vibration of the resonator and the influence of the mass unbalance on the frequency splitting and support loss are clarified. Through experiments, the frequency splitting characteristic and coupling vibration characteristics of the resonator under different removal masses are measured, and the relationship between the mass unbalance of the resonator and the vibration characteristics is obtained. The distribution characteristics of the 1st, the 3rd and the 2nd harmonics of the resonator without balance and the support loss introduced by them are 1. 6×10 -8 and 3. 7×10 -9 . The corresponding relationship between frequency splitting and 4th harmonic unbalance mass is 0. 01 Hz/ μg. Therefore, the mass unbalance of the resonator can be evaluated and the balancing process is formulated.

Abstract:The raster map generated by the particle filter-based SLAM algorithm has the problem of low storage efficiency. To address this issue, a SLAM algorithm based on Huffman run-length coding is proposed. The coding redundancy and space redundancy problems of the original raster map are solved. Based on the particle filter-based SLAM algorithm, the algorithm uses Huffman run-length coding map representation. According to different application scenarios, two storage methods of Huffman run-length coding maps are proposed. When the scale of the grid map is small, the fixed-length coding is used. When the scale of the raster map is relatively large, such as a large shopping mall environment, the variable-length coding is utilized, which further expands the application range of the map representation. Simulation and real scene experiments show that, under certain conditions, the SLAM algorithm based on Huffman run-length coding can reduce memory consumption by up to 94. 8% , which proves the feasibility and effectiveness of the algorithm.

Abstract:The internal complex structure precision measurement of key part is a challenge in the field of high quality manufacturing. When the industrial CT technology is used to achieve precise measurement of the internal structure of the object, it faces problems of grayscale inhomogeneity, blurred edges, and artifacts of the target image. In view of these, the local energy minimization model (RSF) image segmentation method is investigated in this article. The natural gradient and AdamW algorithms are used to improve the convergence speed and parameter adaptivity of the RSF model, respectively. First, the approximate natural gradients are computed on the statistical manifold to improve gradient descent efficiency and RSF model convergence speed. Secondly, the AdamW algorithm is utilized to realize the adaptive control of the scale of the Gaussian kernel function of the RSF model. Compared with the classical RSF model, the improved RSF model reduces the number of iterations by 1 353, the number of iterations by about 76. 79% , the iteration time by about 43. 61% , and the low measurement errors of the probe-radius and the diameter of jet fuel nozzle cylinder, which not only maintains the sub-pixel segmentation accuracy of the original model, but also significantly improves the convergence speed and robustness of the model.

Abstract:The cuboid primitive fitting is a typical problem of 3D point cloud geometric fitting, which is widely used in practical scenes, such as 3D reconstruction, reverse engineering and industrial 3D measurement. In practical applications, it is usually impossible to obtain complete cuboid point cloud data due to occlusion and equipment blind area, which makes it difficult to accurately fit cuboid structure in modeling or measurement. To solve this problem, this article proposes a cuboid parameterized fitting method which combines plane-fitting projection segmentation and incomplete plane vertical angle detection, and realizes the cuboid parameterized fitting. Firstly, the plane point cloud with strong contour information is obtained by the plane fitting projection segmentation algorithm. Then, the detection algorithm of incomplete plane vertical angle is designed to fit the real corner points of cuboids. Finally, the non-coplanar fourpoint method is used to calculate and complete the incomplete corner points of cuboids to obtain the complete cuboid parameter information. Experiments show that the proposed method can accurately detect and complete cuboid corner point information and plane parameter information in all kinds of situations. The accuracy rate and recall rate of corner point are both 100% , and the average error is only 1. 204×10 -3 m. The proposed method can achieve accurate cuboid parameterization fitting.

Abstract:The hand-eye calibration is a key technology in robot vision guided precision operation. The traditional hand-eye calibration and the robot tool center point (TCP) calibration are carried out separately, which has a large cumulative error. Meanwhile the handeye calibration for depth cameras has a disadvantage of insufficient accuracy. In this article, a new method for simultaneous calibration of hand-eye relationship of depth camera and TCP calibration is proposed. Based on the calibration plane in the camera coordinate system and the calibration plane in the robot arm coordinate system are actually the same plane, and the hand-eye relationship is calculated through the transformation between the two plane equations. This method reduces the cumulative error effect of independent calibration of TCP and hand-eye relationship, and saves calibration time and calibration cost. The results of simulation and measurement show that the proposed method can improve the hand-eye calibration accuracy of depth camera, and the average measured position error after calibration is 0. 2 mm, which provides a new idea for the calibration required for high-precision operation of the robot vision control system.