70results about How to "Real-time estimate" patented technology

The invention belongs to the field of automotive electronic measurement, and in particular relates to an automobile tire monitoring device based on an organic piezoelectric material. The automobile tire monitoring device consists of a probe in a tire, a circuit part in the tire (slave) and a main control circuit part, wherein, the probe is formed by matching a PVDF (polyvinylidene fluoride) film with a specific support. In the invention, change of the operating distance x in the case of under-pressure of the automobile tire during the running process is converted into pressure and deformation acting on the PVDF film through the probe so as to generate a piezoelectric signal for reflecting the status of the tire; and meanwhile, data correction based on multi-physical quantity is carried out on an induction signal, and then the signal is transmitted in a wireless communication manner, so that the monitoring system can effectively measure and estimate the tire pressure in real time, and can also judge truly dangerous situations of the tire, such as air leakage and under voltage, and give an alarm.

The invention discloses a vehicle quality estimation method considering gear shifting and road slope factors. The vehicle quality estimation method comprises the following steps that 1, data are collected, specifically, (11), the vehicle driving state data are collected through a data collecting device, and (12), parameters required by a model are calculated and obtained by combining vehicle intrinsic parameters with the vehicle driving state data; 2, based on the obtained vehicle driving state data, a vehicle longitudinal dynamic model is established; 3, a vehicle quality estimation model considering the gear shifting factor is established; 4, a transmission system mechanical efficiency estimation model is established, and the transmission system mechanical efficiency estimation model and the vehicle quality estimation model which considers the gear shifting factor and is provided in the third step jointly form a vehicle quality estimation model which considers the gear shifting and road slope factors simultaneously; 5, the use condition of a vehicle quality real-time estimation system is determined; and 6, the collected vehicle driving state data and relevant model parameters are input into the vehicle quality real-time estimation system to estimate the real-time vehicle quality.

The invention discloses a heavy-duty combination vehicle parameter estimation method. The method aims at estimating the road slope, the whole vehicle mass and the location of the center of mass online in real time according to the environment sensing device information and sensor information of a heavy-duty combination vehicle. According to the method, the influence on slope estimation precision of the pitch angle is sufficiently considered, and the real road slope is obtained by subtracting the pitch angle, calculated by an air spring height sensor, of a motor tractor from the estimated road slope calculated according to the environment sensing device information; according to mass estimation, the influence on slope resistance and air resistance is sufficiently considered, the air resistance is estimated as the unknown quantity, and the estimation precision is improved; according to electric control air braking system pressure sensor information, the location of the center of mass is estimated.

The invention discloses a remaining capacity estimation method and system of a lithium battery. The method comprises the following steps that S1, the working mode of the lithium battery is regularly judged, if the working mode is a charging mode, the step S21 is carried out, and if the working mode is a discharging mode, the step S22 is carried out; S21, whether or not the chargeable capacity change rate of the lithium battery reaches a wave crest or a wave trough is judged, if yes, the step S3 is carried out, and if not, the step S22 is carried out; S22, the current SOC value is calculated bymeans of the ampere-hour integral method, and the step S4 is carried out; S3, the current SOC value is calculated by means of the ampere-hour integral method, and SOCp is used for rectifying the current SOC value, wherein the SOCp is the SOC value of the corresponding extreme point when the chargeable capacity change rate of the lithium battery reaches the wave crest or the wave trough; S4, the current SOC value is output. The method is easy and practicable, the calculation amount is small, and on the premise that the current hardware condition is not improved, the estimation precision of theSOC can be effectively improved.

A gas-liquid dynamic model-based accurate lithium-ion battery SOC estimation method, the gas-liquid dynamic model being a cylindrical closed container, a pipe and a valve that can be opened or closed being mounted at the top of the container, a liquid having a volume of Vw being held in the cylindrical container, and the remaining volume, V, being the volume of a gas having a pressure of P. According to the accurate SOC estimation method, a model, different from conventional RC equivalent circuit model and electrochemical model, is established from the perspective of a gas-liquid coexistence system. The method comprises the following steps: setting up an ideal gas state equation, a gas continuous motion equation, a gas-liquid dissolution equilibrium equation, etc. according to a gas-liquid coexistence system model; deriving steady state and transient state recursion formula of the gas-liquid coexistence system according to the model equations; performing intermittent charging and discharging stationary test on a certain model of battery; and fitting the expression of the relation between SOC and open circuit voltage, and recognizing parameter values of the gas-liquid coexistence system model, so as to achieve online estimation of a battery SOC value. The recursion formula is simple, has self-convergence characteristics, and can accurately describe the nonlinear characteristics of the battery.

The invention discloses a tea tree LAI and nitrogen content estimation method based on hyperspectral images. The method comprises acquiring multiple tea plants as test samples, carrying out correlation and regression analysis on LAI, nitrogen content, hyperspectral position variable, hyperspectral area variable and vegetation index variable of a tea tree as a test sample, building a final tea treeLAI estimation model and a final tea tree nitrogen content estimation model, acquiring hyperspectral position variable, hyperspectral area variable and vegetation index variable of the tea tree needing the LAI and nitrogen content test, respectively inputting the above variables of the tea tree into the final tea tree LAI estimation model and the final tea tree nitrogen content estimation model and calculating the LAI and nitrogen content of the tea tree. The method has the advantages of real-time estimation of LAI and nitrogen content of a tea tree, fastness and no damage.

Aiming at the problems and through application experience of a practical monitoring system, the invention provides a vehicle weight measuring system based on a GPS (Global Position System) and an inertial sensor, which is low in measuring device cost, is convenient to install behind and achieves supervision and administration effects on overload illegal behaviors, so as to realize a using effect of a full monitoring system. The vehicle weight measuring system comprises an ARM (Automatic Route Management) embedded system and a data acquisition module, wherein the data acquisition module is connected with an acceleration sensor and an angular velocity sensor; and the data acquisition module is integrated with a GPS module.

The invention discloses a method for estimating the SOC of a lithium battery by a data driving algorithm considering internal resistance. The method comprises the following steps: firstly, carrying out a charge-discharge test on the lithium ion battery by using test equipment, measuring voltage, current, temperature and internal resistance data of the battery in different working states, and preprocessing the obtained data; then, building a bidirectional GRU network, wherein one part of processed data serves as a training set to train the network, and the other part of the processed data serves as a test set to evaluate network performance; and finally, in order to improve the performance of the constructed network, optimizing the bidirectional GRU network by using an NAG algorithm. The input of the constructed bidirectional GRU-NAG network is the voltage, the current, the temperature and the internal resistance of the battery, and the output of the constructed bidirectional GRU-NAG network is the residual electric quantity of the battery, so that the method has the advantages of high estimation speed and simple process, and is a data-driven battery residual electric quantity estimation model.

The invention discloses an unmanned vehicle obstacle avoidance method and system based on obstacle configuration reconstruction. The method comprises steps of reconstructing obstacle data under a global coordinate system, building a corresponding target function, and solving the optimal solution of the target function to obtain an unmanned vehicle obstacle avoidance optimal path. The method is advantaged in that compared with the prior art, an expansion configuration (double splayed, circular, elliptical or rectangular configuration) with a safety area is reconstructed, the reconstructed expansion configuration is reconstructed according to the detected obstacle data, therefore, even if the reconstructed expansion configuration deviates when the obstacle data is detected for the first time, the obstacle data can be continuously detected in real time, and obstacle configuration is continuously adjusted according to the obstacle data, so the size of the obstacle is estimated more accurately in real time, and finally, obstacle avoidance path planning of the unmanned vehicle is realized.

Methods for combining anatomical data and physiological data on a single image are provided. The methods include obtaining an image, for example, a raw near-infrared (NIR) image or a visible image, of a sample. The image of the sample includes anatomical structure of the sample. A physiologic map of blood flow and perfusion of the sample is obtained. The anatomical structure of the image and the physiologic map of the sample are combined into a single image of the sample. The single image of the sample displays anatomy and physiology of the sample in the single image in real time. Related systems and computer program products are also provided.

The invention provides a middle-distance and long-distance approximate position estimation method of a space tethered robot. According to the middle-distance and long-distance approximate position estimation method, a camera is used for measuring a target azimuth angle to obtain an azimuth angle of a target relative space tethered robot; a space tethering and rope releasing mechanism of the space tethered robot is combined to obtain tethering length information of the space tethered robot; a lossless Kalman filtering method is used for realizing position estimation of the space tethered robot. According to the middle-distance and long-distance approximate position estimation method of the space tethered robot provided by the invention, the tethering length information of the space tethered robot is sufficiently utilized and field angle information of a monocular camera is combined, so as to realize real-time estimation of the position of the space tethered robot.

The invention relates to a free floating target pose prediction method based on hybrid Kalman filtering. The method comprises steps of 1, establishing a free floating target kinetic model; 2, designing a hybrid Kalman filter, and estimating a motion state and kinetic model parameters by using the hybrid Kalman filter; and 3, predicting a future period of time track of a floating target by utilizing the estimated motion parameters. The method is advantaged in that aPose parameters of the target are represented by axial angles, the positions are recorded as independent motions of three axes, theestimated motion information is used for predicting the position and the pose in a future period of time, and the pose of a certain moment in the future is obtained. The method is advantaged in thatlong-period prediction of the free floating target track is carried out in real time in a rolling mode, accuracy is good, and thereby a foundation is laid for follow-up mechanical arm capturing.

The invention discloses a storage battery capacity online dynamic estimation method based on impact load response characteristics. The storage battery capacity online dynamic estimation method comprises the steps of: keeping carrying out online real-time monitoring on a voltage, a current and a temperature of a battery; and when an impact load occurs, i.e., when an instantaneous increase amplitudeof a current of the storage battery is greater than a preset value, transmitting storage voltage, current and temperature data acquired during the load sudden change to an online estimator, by a storage battery capacity online estimation algorithm fusing wavelet analysis with a cerebellar model neural network, estimating surplus capacity of the storage battery in real time. According to the invention, the storage battery does not need to be subjected to long-time full charge and full discharge; other harmonic signals also do not need to be actively injected into the storage battery; by directly utilizing the voltage, current and temperature data of the battery, which is monitored in real time under the action of the impact load, the capacity of the storage battery can be rapidly and accurately estimated in real time; the storage battery capacity online dynamic estimation method is suitable for an uninterrupted power system; battery maintenance cost is reduced; a problem of reduction or failure of the capacity of the storage battery is found timely; and reliability of the system is improved.

The invention discloses an engine indicating torque estimating method. The method comprises the steps of calculating the instantaneous rotation speed of an engine according to a signal of a crankshaft position sensor; collecting time slices of each tooth of the crankshaft position sensor, recoding the time slices during an acting process and storing a time slice array; utilizing the currently stored time slice array to calculate the current rotation speed acceleration; according to the current fuel-injection quantity Q of the engine, the current rotation speed acceleration and the current instantaneous rotation speed of the engine, calculating and outputting the engine indicating torque TInd. The invention discloses an engine indicating torque calculating circuit. According to the invention, the engine indicating torque can be calculated and output by utilizing the crankshaft position sensor and introducing the concept of rotation speed acceleration based on tooth time slices. Therefore, the indicating torque of the engine can be precisely estimated in real time.

The invention provides a road adhesion coefficient estimating method based on cloud data sharing. The method comprises the steps of: (S100): building of a first cloud data image blocks acquired by a vision sensor are matched with adhesion coefficient estimating value based on dynamics to form data pairs, and the data pairs are uploaded to a cloud; (S200) training of classifiers: the cloud receives the data pairs, and arranges and trains the data pairs to generate useful classifiers; and (S300) matching of the classifiers: a second cloud database is built in the actual driving process of avehicle, and is matched with the trained classifiers to obtain the optimal-matched classifier for real-time road adhesion coefficient estimation. The method carries out the data management and training at the cloud completely, reduces demands on a vehicle-mounted calculating unit, can match with all the same vehicle types after completion of training, and supports online updating.

The invention discloses a device and a method for detecting head postures based on a monocular camera, and belongs to the field of computer vision. The device for detecting the head posture comprisesan AprilTag label attached to the top of the head of a subject, a monocular camera above the AprilTag label, a notebook computer, a USB transmission line for connecting the monocular camera with the notebook computer, a tripod for fixing the monocular camera, and a telescopic ring for adjusting the height of the tripod. According to the method for detecting the head posture, a monocular camera isused for shooting an AprilTag label attached to the top of the head. The AprilTag label moves along with the head, the spatial pose of the AprilTag label is obtained through the AprilTag algorithm, and the pose of the head is judged through the change of the rotation angle of the AprilTag label. The device and the method are a novel head posture detection method, and the human-computer interactioncapability can be improved.

The invention discloses a regional vehicle density estimation method based on a dynamic sampling mechanism and an RBF neural network, and the method comprises the steps of preliminarily constructing avehicle density database according to the sampling information of the vehicle density in a target region; comparing the estimation values of the vehicle density for many times, taking the stored sampling data as an input variable of the RBF neural network, defining a group of activation functions, and establishing an estimation model based on the RBF neural network; applying the Kalman filteringalgorithm to an RBF neural network estimation algorithm; estimating the vehicle density of any point in the target area according to the weight coefficient of each neural network and the correlation function of the vehicle density in the target area; and finally, dynamically estimating the vehicle density in the target area by judging whether the estimation result meets the task requirement or not. The method has the advantages of higher estimation efficiency, lower operation load and higher estimation precision, can effectively estimate the time-varying vehicle density in the target area in real time, and has wide application space and practical range.

The invention discloses a positioning method and device adopting visual inertial data deep fusion, wherein the method comprises the following steps of: S1: enabling a system to be in a static state atan initial moment, and acquiring a measured value of an accelerometer and a measured value of a gyroscope in an initialization time period so as to estimate an initial state of the system; S2: afterobtaining the initial state, carrying out propagation on a system state according to the measured value of the accelerometer and the measured value of the gyroscope, and carrying out updating on a covariance matrix of the system; S3: after acquiring an image, tracking feature points by utilizing an IMU (Inertial Measurement Unit) assisted external point elimination method; and S4: for feature points failed in tracking, constructing visual measurement according to visual measurement information, and carrying out updating on the system state. The method sufficiently utilizes various sensor datain a visual inertial fusion system, can effectively utilize IMU data to promote a tracking effect and efficiency, and can accurate estimate a pose of the system in real time.

The invention discloses a lightweight human body posture estimation method and system based on streaming attention. The method comprises the steps that a target image needing posture estimation is acquired; carrying out standardized clipping and normalization processing on the target image; inputting the target image into a trained lightweight human body posture estimation network based on streaming attention, and outputting a posture estimation heat map result; a model training step: performing down-sampling on training images in the training data set through convolution of different scales, obtaining four first feature maps of different scales as input of an up-sampling module through multiple groups of ghost lightweight modules with streaming attention, and performing convolution respectively to obtain four second feature maps of the same number of channels; and carrying out bilinear up-sampling operation to obtain a double feature map, carrying out feature fusion, outputting a third feature map, and converting to obtain a heat map required by human body posture estimation. The method is high in recognition precision and low in complexity.

The invention discloses an estimation method for a dynamic frequency response function of a voltage transformer and a realizing device thereof. The estimation method comprises the steps of using a triangular-rectangular mutual-convolution window to carry out time-domain weighting on discrete stimulation and response; carrying out discrete fourier transformation on the weighted stimulation and response; and correcting a peak spectral frequency of the stimulation and response through a discrete peak spectral frequency weighted average algorithm to acquire an estimated result of the frequency response function of the voltage transformer as a ratio of the corrected response frequency spectrum to the stimulation peak spectral frequency. According to the estimation method for the dynamic frequency response function of the voltage transformer and the realizing device thereof, mutual interference of spectrum leakage and harmonic wave in the estimation on the frequency response function of the voltage transformer can be restrained, the estimation stability for the dynamic frequency response function of the voltage transformer is improved, and the real-time accurate estimation on the dynamic frequency response function of the voltage transformer is achieved.

The invention discloses a human behavior classification method based on multi-station radar micro Doppler motion direction finding, and belongs to the field of radar target detection signal processing. According to the implementation method, azimuth angles of human motion relative to different radar receivers are measured in real time by using a human motion direction estimation method based on multi-station radar Doppler frequency. The method comprises the following steps: performing interval division on multi-station data according to different influences of measured angle values and different azimuth angles on classification performance, performing data-level fusion according to divided intervals, and performing feature extraction and classification on the data of different intervals by using a dual-channel convolutional principal component analysis network (CPCAN), and carrying out adaptive weighted decision level fusion on classification results of the two channels to obtain a final behavior category result. According to the method, the influence of the micro Doppler data caused by the movement azimuth change is fully considered, so that a multi-station classification network structure can be simplified, and better classification performance and a more stable classification effect are achieved.

The invention discloses a road gradient estimation method, system and equipment for an electric vehicle, and belongs to the field of road gradient calculation.The road gradient estimation method comprises the steps that firstly, an acceleration value of a controller is processed through a Gaussian Newton method and first-order filtering, a system zero offset error and a scale error are eliminated, and an accurate and useful acceleration value is obtained; secondly, the slope change rate is estimated through a forgetting factor least square method, and the slope can be estimated on the road surface with the steep slope and the fast slope change in consideration of the change of the slope change rate of the road surface; and finally, in combination with a Kalman filtering method, influence of a gradient change rate factor is estimated through a previous forgetting factor least square method, and the estimation real-time performance of the time-varying gradient is very good through the Kalman filtering method. According to the method, the slope value can be estimated quickly in real time and the accuracy of the calculated slope value can be ensured during gear shifting and on a road with a large slope and quick change.

The invention provides a density traffic flow counting method and system based on space-time counting characteristics. The density traffic flow counting method comprises the steps: obtaining a trafficvideo, and carrying out the vehicle detection in a selected salient region; constructing a line-crossing probability function to evaluate the probability that the detected vehicle may pass through the counting line, selecting the vehicle needing to be tracked in each frame of image according to the probability value, and extracting the spatial feature and the time feature of the vehicle; constructing a counting model based on a long-short-term recurrent neural network, and analyzing bidirectional space-time counting features in the video to obtain a classification and a corresponding countingresult; and estimating the flow based on the counting result, and calculating to obtain the flow traffic volume, density and speed. According to the density traffic flow counting method, the whole dense traffic flow can be described, and the traffic flow can be accurately determined.

The invention aims at providing a linear frequency modulation (FM) signal parameter estimation method based on small-wave-packet denoising and power spectral entropy, which includes the following steps: denoising signals in a multi-dimension small-wave-packet mode, and determining small-wave-packet function and small-wave-packet decomposition level; calculating power spectral entropy of signals denoised in a small-wave-packet mode, and setting an entropy feature data base of linear FM signals with different FM slope under the condition of different signal to noise ratios; carrying out interpolation operation for obtained discretized entropy features; fitting the curve after interpolation of a cubic spline function with a polynomial function, and obtaining FM slop of linear FM signals under the condition of different signal to noise ratios and relational expression with input entropy features; and estimating FM slop of linear frequency modulation (LFM) signals received by a receiver by utilizing the fit expression. The linear FM signal parameter estimation method based on small-wave-packet denoising and power spectral entropy is small in calculated amount and capable of estimating FM slop of LFM in real time under the premise of meeting the requirement for parameter estimation accuracy.

The invention discloses an electric vehicle motor torque estimation method and system and a vehicle. The method comprises the following steps of extracting a motor d-axis flux linkage change value deltapsid at the current moment from a d-axis flux linkage, and because deltapsif=deltapsid, substituting deltapsif into a formula Te=deltapsif*K+Te0, namely estimating a torque value Te at the current moment. The curve function Te=deltapsif*K+Te0 is obtained by connecting a torque sensor and a flux linkage measuring instrument on a rack, acquiring different flux linkages and torques corresponding to the flux linkages and fitting the flux linkages and the torques, and Te is a torque value at the current moment; deltapsif is a permanent magnet flux linkage change value of the motor; Te0 is an execution torque; and K is a coefficient value obtained by fitting. The method for estimating the motor torque of the electric vehicle can estimate the motor torque in real time, and the estimation result is more accurate.

The invention provides a transformer substation inspection robot indoor positioning method based on information fusion. The method comprises the steps: carrying out scanning and photographing on a two-dimensional code, and processing an obtained two-dimensional code image; calculating the pose of a to-be-tested robot in a two-dimensional code positioning mode; estimating the linear velocity and the angular velocity of the to-be-tested robot in the two-dimensional code positioning mode; measuring the linear velocity and the angular velocity of the to-be-tested robot in an inertial measurement positioning mode; estimating the pose of the to-be-tested robot in the inertial measurement positioning mode; constructing an observer; iteratively calculating a pose deviation estimated value of the to-be-tested robot in the inertial measurement positioning mode; and calculating the final pose of the to-be-detected robot. According to the invention, information fusion is carried out on a two-dimensional code positioning mode and an inertial measurement positioning mode, the observer model based on the attitude kinematics of the substation inspection robot is constructed, and the real-time pose of the substation inspection robot can be rapidly and accurately estimated in real time.

The embodiment of the invention relates to a method and a terminal device for self-interference estimation. The method comprises the steps that the terminal device obtains a self-interference signal estimated for the ith time according to a received signal and an uplink signal, the received signal comprises the self-interference signal and a downlink signal received by the terminal device, the uplink signal carries self-interference information, i is a positive integer, and i is a value from 1 and is increased progressively according to 1; the terminal device determines a downlink signal estimated for the ith time based on the received signal and the self-interference signal estimated for the ith time; the terminal device judges whether the ith residual signal is converged to the background noise or not, and determines whether the self-interference signal estimated for the ith time is an expected self-interference signal, wherein the ith residual signal is the difference between the received signal and the received signal estimated for the ith time, and the received signal estimated for the ith time is the sum of the self-interference signal estimated for the ith time and the downlink signal estimated for the ith time. According to the self-interference estimation method and the terminal device provided by the embodiment of the invention, self-interference can be suppressed.