Dynamic vehicle mass calculation method and device and server

Abstract

The invention provides a dynamic vehicle mass calculation method and device and a server, and relates to the technical field of vehicle mass calculation, and the method comprises the steps: obtaining acceleration data of a dynamic vehicle through an acceleration sensor, and obtaining motor torque data through a controller local area network; carrying out calibration processing on the acceleration data and the motor torque data, carrying out signal processing on the calibrated data by using a recursive least square method, determining initial vehicle mass, and carrying out identification processing on a vehicle driving gradient by using a Kalman filtering algorithm, and determining an initial vehicle driving gradient; and through a preset vehicle driving force calculation model, the initial vehicle driving force is determined based on the initial vehicle mass and the initial vehicle driving gradient, and the compensated target vehicle mass is determined through error compensation processing. Error compensation is carried out through multi-source fusion data to determine the dynamic vehicle mass, and the accuracy of vehicle mass calculation can be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle mass calculation, and in particular to a dynamic vehicle mass calculation method, device and server. Background Art

[0002] At present, in the field of vehicle mass calculation, there are already a variety of methods for dynamic vehicle mass calculation. Relevant technologies propose that the vehicle mass can be directly measured based on sensors or calculated based on simple algorithms. This solution can quickly obtain vehicle mass data in simple scenarios, and the measurement principle is relatively intuitive. In addition, the vehicle mass can be calculated based on a simple algorithm, or the vehicle mass can be indirectly inferred by measuring tire deformation, vibration and other information during vehicle driving. However, the above schemes have large measurement errors under complex road conditions such as uphill and downhill slopes, curves or uneven roads. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a dynamic vehicle mass calculation method, device and server, which can significantly improve the accuracy of vehicle mass calculation by determining the dynamic vehicle mass through error compensation through multi-source fusion data.

[0004] In a first aspect, an embodiment of the present invention provides a dynamic vehicle mass calculation method, which is applied to a dynamic vehicle mass calculation system, and the method includes: acquiring acceleration data of a dynamic vehicle through an acceleration sensor, and acquiring motor torque data through a controller area network; calibrating the acceleration data and the motor torque data, and performing signal processing on the calibrated data using a recursive least squares method to determine an initial vehicle mass, and identifying the slope of vehicle travel using a Kalman filter algorithm to determine an initial vehicle travel slope; determining an initial vehicle driving force based on an initial vehicle mass and an initial vehicle travel slope using a preset vehicle driving force calculation model, and performing error compensation processing based on the initial vehicle driving force, the initial vehicle mass, and a preset vehicle reference mass to determine a compensated target vehicle mass.

[0005] In one embodiment, the step of calibrating the acceleration data and the motor torque data includes: obtaining the current ambient temperature data, and using the ambient temperature data to perform temperature compensation on the acceleration data and the motor torque data to determine the compensated data; performing zero drift calibration on the compensated data to determine the calibrated data.

[0006] In one embodiment, the step of performing temperature compensation processing on the acceleration data and the motor torque data using the ambient temperature data includes: performing temperature compensation processing on the acceleration data and the motor torque data based on a preset temperature compensation parameter table and the current ambient temperature data by using a preset temperature compensation correction function, wherein the preset temperature compensation correction function is:

[0007] in, is the compensated data, is the original collected data, is the temperature compensation coefficient determined by the calibration experiment, T is the current ambient temperature, It is the reference temperature for calibration experiments.

[0008] In one embodiment, a zero drift calibration is performed on the compensated data, and the step of determining the calibrated data includes: statically collecting the drift amounts corresponding to the acceleration data and the motor torque data for a preset number of times, and determining the mean of the drift amounts as the target drift amounts corresponding to the acceleration data and the motor torque data, respectively; and determining the difference between the compensated data and the target drift amount as the calibrated data.

[0009] In one embodiment, the step of performing signal processing on the calibrated data using recursive least squares method to determine the initial vehicle mass includes: obtaining a gain coefficient and extracting a characteristic vector of the calibrated acceleration data and motor torque data; inputting the gain coefficient and the characteristic vector into a recursive least squares method calculation model to determine the initial vehicle mass.

[0010] In one embodiment, the recursive least squares calculation model is:

[0011] in, is the estimated value of vehicle mass calculated for the kth time, is the vehicle mass calculated last time, is the gain coefficient, which is used to control the update amplitude of the model. To observe the driving force, is the characteristic vector containing acceleration and torque, is the parameter vector to be estimated.

[0012] In one embodiment, the Kalman filter algorithm is used to identify and process the slope of a vehicle, and the step of determining the initial vehicle driving slope includes: determining the estimated value of the current vehicle driving slope as the sum of the estimated value of the previous vehicle driving slope and the process noise; substituting the estimated value of the current vehicle driving slope into a preset nonlinear measurement function, and determining the initial vehicle driving slope as the sum of the substitution result and the measurement noise.

[0013] In one embodiment, the step of determining the initial vehicle driving force based on the initial vehicle mass and the initial vehicle driving slope by using a preset vehicle driving force calculation model includes: obtaining a preset drive axle speed ratio and a wheel radius of the vehicle, substituting the preset drive axle speed ratio, the wheel radius, the motor torque data, the initial vehicle mass and the initial vehicle driving slope into the preset vehicle driving force calculation model to determine the initial vehicle driving force, wherein the preset vehicle driving force calculation model is:

[0014] in, is the real-time torque of the motor obtained from the CAN bus, the drive axle ratio and wheel radius are the inherent parameters of the vehicle, M is the initial vehicle mass, f is the vehicle driving resistance coefficient, and β is the initial vehicle driving slope.

[0015] In one embodiment, error compensation is performed based on the initial vehicle driving force, the initial vehicle mass and the preset vehicle reference mass, and the step of determining the compensated target vehicle mass includes: performing root mean square error calculation processing on the initial vehicle mass and the preset vehicle reference mass to determine the average error value, and using the average error value to update the gain coefficient in the recursive least squares calculation model to determine the target gain coefficient; using the target gain coefficient and the initial vehicle driving force to update the recursive least squares calculation model, and using the updated model to recalculate the vehicle mass to determine the target vehicle mass.

[0016] In a second aspect, an embodiment of the present invention further provides a dynamic vehicle mass calculation device, which is applied to a dynamic vehicle mass calculation system, and the device includes: an information acquisition module, which acquires acceleration data of a dynamic vehicle through an acceleration sensor, and acquires motor torque data through a controller area network; a mass calculation module, which performs calibration processing on the acceleration data and the motor torque data, and uses a recursive least squares method to perform signal processing on the calibrated data to determine the initial vehicle mass, and uses a Kalman filter algorithm to identify the slope of the vehicle and determine the initial vehicle driving slope; an error compensation module, which determines the initial vehicle driving force based on the initial vehicle mass and the initial vehicle driving slope through a preset vehicle driving force calculation model, and performs error compensation processing based on the initial vehicle driving force, the initial vehicle mass and the preset vehicle reference mass to determine the compensated target vehicle mass.

[0017] In a third aspect, an embodiment of the present invention further provides a server, comprising a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement any one of the methods provided in the first aspect.

[0018] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement any one of the methods provided in the first aspect.

[0019] The embodiments of the present invention bring the following beneficial effects: A method, device and server for calculating dynamic vehicle mass are provided in an embodiment of the present invention. The method first obtains acceleration data of a dynamic vehicle through an acceleration sensor and obtains motor torque data through a controller area network. Then, the acceleration data and motor torque data are calibrated, and the calibrated data are signal processed using a recursive least squares method to determine the initial vehicle mass. The vehicle driving slope is identified using a Kalman filter algorithm to determine the initial vehicle driving slope. Finally, the initial vehicle driving force is determined based on the initial vehicle mass and the initial vehicle driving slope through a preset vehicle driving force calculation model. Error compensation is performed based on the initial vehicle driving force, the initial vehicle mass and the preset vehicle reference mass to determine the compensated target vehicle mass. The embodiment of the present invention can accurately calculate the vehicle mass by collecting, processing and fusing multi-source data during vehicle operation, combined with a specific algorithm and an error compensation strategy, and is widely used in quality dynamic monitoring and analysis scenarios of various vehicles, such as load monitoring of logistics and transportation vehicles, operation evaluation of engineering vehicles, etc., to provide data support for safe and efficient operation of vehicles.

[0020] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 A structural schematic diagram of a dynamic vehicle mass calculation system provided by an embodiment of the present invention; Figure 2A schematic flow chart of a dynamic vehicle mass calculation method provided by an embodiment of the present invention; Figure 3 A schematic diagram of a specific flow chart of a dynamic vehicle mass calculation method provided by an embodiment of the present invention; Figure 4 A schematic diagram of the structure of a dynamic vehicle mass calculation device provided by an embodiment of the present invention; Figure 5 A schematic diagram of the structure of a server provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described in combination with the embodiments below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] At present, in the field of vehicle mass calculation, there are already a variety of methods for dynamic vehicle mass calculation. Relevant technologies propose that the vehicle mass can be directly measured based on sensors or calculated based on simple algorithms. For example, by using pressure sensors and strain gauges installed in the vehicle suspension system, axles and other parts, the pressure or strain borne by these components when the vehicle is loaded is measured, and then the mass of the vehicle is calculated. This solution can quickly obtain vehicle mass data in simple scenarios, and the measurement principle is relatively intuitive. In addition, the vehicle mass can also be calculated based on a simple algorithm (for example, based on the vehicle's motion equation, combined with the vehicle's acceleration, driving force and other parameters), based on basic mechanical formulas to calculate the vehicle mass (for example, by measuring the acceleration of the vehicle during acceleration, and the known engine output torque, transmission system parameters, etc., using Newton's second law F=ma (where F is the resultant force on the vehicle, m is the vehicle mass, and a is the acceleration) to estimate the vehicle mass) or by measuring tire deformation, vibration and other information during vehicle driving, the vehicle mass can be indirectly inferred.

[0026] However, the above dynamic vehicle mass calculation method has many limitations. From the perspective of accuracy, in complex road conditions, such as when the vehicle is driving on uphill and downhill slopes, curves, and uneven roads, the method based on direct sensor measurement has large errors. Because in these cases, the forces acting on the vehicle include not only gravity and conventional driving resistance, but also additional lateral forces, components caused by slopes, etc. These factors will interfere with the measurement of the sensor and cause the measurement results to deviate from the true value; and simple algorithm calculations, because they are often based on ideal vehicle models, ignore many actual factors, such as energy losses in the vehicle transmission system, nonlinear friction characteristics between tires and the ground, etc., making it difficult for the accuracy of the calculation results to meet high-precision application scenarios. In terms of adaptability, the existing technology does not fully utilize multi-source data. The vehicle will generate multiple types of data during driving, such as motor torque data from the CAN bus, vehicle speed data, acceleration data collected by the acceleration sensor, vehicle driving posture data, etc., but most of the current methods only rely on a single or Vehicle mass calculation is performed based on only a few types of data, which fails to fully explore the correlation information between different data sources and cannot fully and accurately reflect the actual mass status of the vehicle. In addition, the existing technology has poor adaptability to changes in environmental factors. When the ambient temperature, humidity, air pressure and other conditions change significantly, the performance of the sensor will be affected, thereby affecting the accuracy of the vehicle mass calculation. Based on this, the dynamic vehicle mass calculation method, device and server provided by the present invention can accurately calculate the vehicle mass through the collection, processing and fusion of multi-source data during the operation of the vehicle, combined with specific algorithms and error compensation strategies, and are widely used in quality dynamic monitoring and analysis scenarios of various types of vehicles, such as load monitoring of logistics and transportation vehicles, operation evaluation of engineering vehicles, etc., to provide data support for the safe and efficient operation of vehicles.

[0027] To facilitate understanding of this embodiment, a dynamic vehicle mass calculation method disclosed in an embodiment of the present invention is first described in detail. The method is applied to a dynamic vehicle mass calculation system. To facilitate understanding of the dynamic vehicle mass calculation system, an embodiment of the present invention provides a structural schematic diagram of a dynamic vehicle mass calculation system, as shown in FIG. Figure 1As shown, it is used to receive signals from a signal source, process the signals, and issue an alarm through a prompt system. Specifically, the collected data can be input into the signal processing module of the device, wherein the signal processing module is used to preprocess and calibrate the signals, and perform preliminary calculations to estimate the vehicle mass and slope. The information signal processing module first calibrates the signals collected by the acceleration sensor and the motor torque signal obtained from the CAN bus. After the calibration is completed, the calibrated signals are processed using the recursive least squares method (RLS) to preliminarily identify the vehicle mass. At the same time, the extended Kalman filter algorithm (EKF) is used to identify the slope of the vehicle and calculate the vehicle driving force. Multiple groups of preliminary identification masses and corresponding vehicle driving force data can be obtained by collecting and repeating the above calculations for multiple times. These data are statistically analyzed, and the identification masses obtained by multiple calculations are compared with the actual known half-load mass of the vehicle. According to the error situation, the relevant parameters in the recursive least squares (RLS) algorithm are adjusted, such as the gain coefficient, to achieve error compensation for the identification mass, and finally determine the accurate vehicle mass after compensation.

[0028] Furthermore, the compensated vehicle mass data and related data in the calculation process, such as acceleration, motor torque, slope, etc., can be recorded and stored through the recording system of the equipment. At the same time, the prompt system of the equipment will give corresponding prompts according to the calculation results. For example, when the vehicle mass is close to or exceeds the specified load, the instrument will flash and the sound alarm will be sounded. The equipment reserves an interface so that other systems can call the recorded data and calculation results later.

[0029] based on Figure 1 The structural diagram of the dynamic vehicle mass calculation system is shown in FIG. 1 , and the dynamic vehicle mass calculation method is described in detail in the embodiment of the present invention. Figure 2 The flowchart of a dynamic vehicle mass calculation method shown in FIG. 1 mainly includes the following steps S202 to S206: Step S202, the acceleration data of the dynamic vehicle is obtained through the acceleration sensor, and the motor torque data is obtained through the controller local area network. In one embodiment, the data collection process includes: during the operation of the vehicle, the device connects itself to the vehicle CAN bus, and the acceleration sensor in the device collects the acceleration data of the vehicle in real time. At the same time, the device obtains the real-time torque data of the motor from the CAN bus. In various scenarios of vehicle driving, including flat driving, uphill and downhill driving, etc., data collection is continuously and uninterruptedly. The frequency of the acceleration sensor collecting acceleration data is dynamically adjusted according to the driving state of the vehicle. In scenarios where the vehicle accelerates, decelerates, or the slope changes significantly, the collection frequency is increased; in scenarios where the vehicle is driving at a constant speed and other stable states, the collection frequency is reduced.

[0030] Step S204, calibrate the acceleration data and motor torque data, and use recursive least squares method to perform signal processing on the calibrated data to determine the initial vehicle mass, and use Kalman filtering algorithm to identify the slope of the vehicle to determine the initial vehicle slope. In one embodiment, the process of calibrating the signal includes: temperature compensation calibration and zero drift calibration. The acceleration sensor and motor torque signal are calibrated through the calibration parameter table pre-stored in the device and combined with the ambient temperature data during collection.

[0031] Step S206, by using a preset vehicle driving force calculation model, based on the initial vehicle mass and the initial vehicle driving slope, the initial vehicle driving force is determined, and error compensation is performed based on the initial vehicle driving force, the initial vehicle mass and the preset vehicle reference mass to determine the compensated target vehicle mass. In one embodiment, after determining the target vehicle mass, the compensated vehicle mass data and related data in the calculation process, such as acceleration, motor torque, slope, etc., can be recorded and stored through the recording system of the device. At the same time, the prompt system of the device gives corresponding prompts according to the calculation results. For example, when the vehicle mass approaches or exceeds the specified load, the instrument flashes and the sound alarm prompts. The device reserves an interface so that other systems can call the recorded data and calculation results later. The recording system adopts a circular storage method. When the storage capacity reaches the upper limit, the earliest recorded data is automatically overwritten to ensure that the latest data within a certain period of time is always stored.

[0032] The above-mentioned dynamic vehicle mass calculation method provided by the embodiment of the present invention can accurately calculate the vehicle mass during the operation of the vehicle by collecting, processing and fusing multi-source data, combined with specific algorithms and error compensation strategies, and is widely used in dynamic quality monitoring and analysis scenarios of various types of vehicles, such as load monitoring of logistics and transportation vehicles, operation evaluation of engineering vehicles, etc., to provide data support for the safe and efficient operation of vehicles.

[0033] See also Figure 3 The specific flow chart of a dynamic vehicle mass calculation method is shown in FIG. 1 . The embodiment of the present invention also provides an implementation method of dynamic vehicle mass calculation, and the details are as follows (1) to (4): (1) First, the collected data is input into the signal processing module of the device. The signal processing module first calibrates the signal collected by the acceleration sensor and the motor torque signal obtained from the CAN bus. That is to say, the current ambient temperature data is obtained, and the ambient temperature data is used to perform temperature compensation processing on the acceleration data and the motor torque data to determine the compensated data. The compensated data is calibrated for zero drift to determine the calibrated data. Signal calibration can improve the accuracy of the data, which mainly includes temperature compensation calibration and zero drift calibration.

[0034] In one embodiment, since the measurement accuracy of the acceleration sensor and the motor torque sensor will be affected by the change of ambient temperature, the system will use the stored temperature compensation parameter table in combination with the current ambient temperature data to correct the measurement signal. Specifically, the acceleration data and the motor torque data can be temperature compensated based on the preset temperature compensation parameter table and the current ambient temperature data by using a preset temperature compensation correction function, wherein the preset temperature compensation correction function is:

[0035] in, is the compensated data, is the original collected data, is the temperature compensation coefficient determined by the calibration experiment, T is the current ambient temperature, It is the reference temperature for calibration experiments.

[0036] In one embodiment, since the acceleration sensor and the torque sensor have zero drift, the system will perform self-calibration in a static state, calculate the drift and deduct it. Specifically, the drift corresponding to the acceleration data and the motor torque data for a preset number of times can be statically collected, and the mean of the drift can be determined as the target drift corresponding to the acceleration data and the motor torque data, respectively. Finally, the difference between the compensated data and the target drift can be determined as the calibrated data, wherein the calculation model of the zero drift is:

[0037] in, is the acceleration signal after final calibration; ΔA is the zero drift compensation value, which is usually calculated by taking the average value of multiple static acquisitions.

[0038] (2) After the calibration is completed, the calibrated signal is processed using the recursive least squares method (RLS) to preliminarily identify the vehicle mass. At the same time, the extended Kalman filter algorithm (EKF) is used to identify the slope of the vehicle. In other words, the calibrated data can be input into the recursive least squares method (RLS) model to estimate the vehicle mass M, and the error weight is iteratively updated through the recursive least squares method to continuously optimize the mass estimation value. Specifically, the gain coefficient can be obtained, and the characteristic vector of the calibrated acceleration data and motor torque data can be extracted. Then, the gain coefficient and the characteristic vector are input into the recursive least squares method calculation model to determine the initial vehicle mass. The recursive least squares method calculation model is:

[0039] in, is the estimated value of vehicle mass calculated for the kth time, is the vehicle mass calculated last time, is the gain coefficient, which is used to control the update amplitude of the model. To observe the driving force, is the characteristic vector containing acceleration and torque, is the parameter vector to be estimated, mainly including the vehicle mass.

[0040] Furthermore, The calculation method is as follows:

[0041] in, is the covariance matrix; λ is the forgetting factor, which ranges from 0<λ≤1 and is usually 098-099.

[0042] In one embodiment, during vehicle driving, the change of slope β will affect the calculation of driving force. Therefore, an extended Kalman filter (EKF) can be used to identify the slope in real time. The EKF iteratively updates the state variables so that the slope estimation value gradually converges to the true value. Specifically, the sum of the estimated value of the previous vehicle driving slope and the process noise is determined as the estimated value of the current vehicle driving slope. Then, the estimated value of the current vehicle driving slope is substituted into a preset nonlinear measurement function, and the sum of the substitution result and the measurement noise is determined as the initial vehicle driving slope. The state equation of the EKF is as follows:

[0043] Measurement equation:

[0044] in, is the current estimated slope, is the estimated slope at the previous moment, is the process noise, is the measured value (such as the slope information calculated by combining the acceleration signal with the ground speed), To measure noise, It is a nonlinear measurement function that describes the relationship between the slope and the measurement variable.

[0045] In actual applications, if an electric vehicle is subjected to mass identification under different slopes and loads, the input data is as follows: acceleration sensor reading (uncalibrated): 0.25m / s²; ambient temperature: 30℃; temperature compensation coefficient kT: 0.002m / s² / ℃; reference temperature Tref: 25℃; zero drift compensation value ΔA: 0.01m / s²; motor torque: 300Nm; drive axle speed ratio: 4.5; wheel radius: 0.3m; slope (initial estimate): 5°; rolling resistance coefficient f: 0.01; initial mass estimate: 1500kg, then the preliminary calculation of mass and slope is: Signal calibration:

[0046]

[0047] Preliminary mass estimate (RLS update): Assume the estimated mass at the previous moment , gain coefficient = 0.1, then:

[0048] If calculated but:

[0049] Slope calculation (EKF update): The initial slope β=5∘, and the corrected slope βk=5.2∘ is obtained after EKF iterative update.

[0050] Finally, the device outputs the current estimated vehicle mass as 1501kg and the slope as 5.2°.

[0051] (3) Calculating the vehicle driving force. Specifically, the preset drive axle speed ratio and the wheel radius of the vehicle may be obtained, and the preset drive axle speed ratio, wheel radius, motor torque data, initial vehicle mass, and initial vehicle driving slope may be substituted into a preset vehicle driving force calculation model to determine the initial vehicle driving force. The preset vehicle driving force calculation model is:

[0052] in, is the real-time torque of the motor obtained from the CAN bus, the drive axle speed ratio and wheel radius are the inherent parameters of the vehicle, M is the initial vehicle mass, f is the vehicle driving resistance coefficient, which is used to describe the coefficient of ground rolling resistance, generally between 0.01 (flat road) and 0.03 (rough road), and β is the initial vehicle driving slope.

[0053] (4) Collect data multiple times and repeat the above calculations to obtain multiple sets of preliminary identification masses and corresponding vehicle driving force data. Perform statistical analysis on these data and compare the identification masses obtained by multiple calculations with the actual known half-load mass of the vehicle (set as ), calculate the error. According to the error situation, adjust the relevant parameters in the RLS algorithm, such as adjusting the gain coefficient , in order to realize the error compensation of the identification quality, and finally determine the accurate vehicle quality after compensation. Specifically, the root mean square error calculation processing can be performed on the initial vehicle quality and the preset vehicle reference quality to determine the average error value, and the average error value can be used to update the gain coefficient in the recursive least squares calculation model to determine the target gain coefficient. Then, the target gain coefficient and the initial vehicle driving force are used to update the recursive least squares calculation model, and the updated model is used to recalculate the vehicle quality to determine the target vehicle quality. Further, for the error calculation method, it is assumed that in the n groups of measurements, the preliminary identification quality set obtained is { , , ..., }, the actual known half-load mass of the vehicle (i.e., the reference mass) is set to The error metric is calculated using the root mean square error (RMSE):

[0054] in, is the recognition quality obtained by the i-th group of measurements, is the actual known half-load mass of the vehicle, n is the number of data collections, and in physical terms, the RMSE index is used to measure the quality of identification. and actual quality The smaller the value, the closer the calculated quality is to the true value and the smaller the error is.

[0055] In one implementation, after the RMSE error is calculated, the gain coefficient of the recursive least squares method (RLS) needs to be adjusted to compensate for the mass calculation error so that the new identified mass is closer to the true mass.

[0056] In the RLS algorithm, the update rules are as follows:

[0057] in, is the corrected mass calculated in the k+1th round, is the mass calculated in the kth round, is the gain factor, which determines the weight distribution of the new and old quality estimates, It is the vehicle driving force calculated according to the driving force formula. It is a matrix related to vehicle dynamics parameters and affects the accuracy of mass calculation.

[0058] The gain coefficient adjustment strategy is set to increase if the RMSE error is large. To enhance the weight of new data, the quality estimation converges faster. If the RMSE error is small, reduce To reduce the impact of new data on existing estimates and improve stability.

[0059] The dynamic adjustment formula of its gain coefficient is:

[0060] in, is the gain coefficient of the previous round, α is the adjustment factor (usually 0.01-0.1), which is used to control the correction amplitude, and RMSE is the root mean square error calculated currently.

[0061] In practical applications, if the actual half-load mass of a vehicle is 1600 kg, the initial identification masses obtained in 5 rounds of measurement are as follows: 1550; 1620; 1580; 1640; 1595; then the RMSE is calculated as:

[0062]

[0063]

[0064]

[0065] The calculated RMSE is 314 kg, indicating that the mass identification error is large and the gain factor of RLS needs to be adjusted.

[0066] Further, adjust the gain coefficient: Assuming the initial gain coefficient =0.5, adjustment factor α=0.05:

[0067] The adjusted gain factor is approximately 0.1945. In the next round of mass calculation, the new gain factor will be used for correction to make the estimated mass closer to the true value.

[0068] In actual application, verification under conventional driving scenarios: In the verification of routine driving scenarios, a common logistics transport truck was selected. The truck was tested for uniform speed driving under normal road conditions. The test section was a 10-kilometer-long straight highway with a flat road surface and good traffic conditions. The vehicle basically maintained a uniform speed during driving.

[0069] During the data collection phase, the data collection device is connected to the truck's CAN bus, and the acceleration sensor begins to collect vehicle acceleration data in real time. Since the vehicle is in a stable state of uniform driving, the acceleration sensor works at a lower acquisition frequency, that is, it collects data every 500 milliseconds. At the same time, the device obtains the real-time torque data of the motor from the CAN bus, and this data is transmitted to the data collection device at a stable rate through the CAN bus. During the entire 10-kilometer driving process, data is collected continuously and uninterruptedly, and a total of 200 sets of acceleration data and corresponding motor torque data are collected.

[0070] The collected data is transmitted to the signal processing module. The signal processing module first calibrates the signal collected by the acceleration sensor and the motor torque signal obtained from the CAN bus. The signal is calibrated for temperature compensation and zero drift by combining the calibration parameter table pre-stored in the device with the ambient temperature data during the acquisition. During the acquisition process, the ambient temperature is 25°C. According to the calibration parameter table, the acceleration sensor signal is subjected to corresponding temperature compensation calculation to ensure the accuracy of the signal. After the calibration is completed, the calibrated signal is processed using the recursive least squares method (RLS) to preliminarily identify the vehicle mass. The recursive least squares method preliminarily concludes that the vehicle mass is 10500kg based on the input acceleration and motor torque signals after multiple iterative calculations. At the same time, the extended Kalman filter algorithm (EKF) is used to identify the slope of the vehicle. Since the vehicle is traveling on a straight highway, the extended Kalman filter algorithm accurately identifies the slope as 0 degrees by fusing information such as acceleration and speed.

[0071] The vehicle driving force was calculated according to the core driving force calculation formula, and the vehicle driving force was 17550N. Data was collected multiple times and the above signal processing, preliminary calculation and vehicle driving force calculation steps were repeated 10 times, and 10 sets of preliminary identification quality and corresponding vehicle driving force data were obtained. These data were statistically analyzed, and the identification quality obtained by multiple calculations was compared with the actual known half-load mass of the vehicle (set to 8000kg). The root mean square error (RMSE) was used as the error measurement indicator, and the formula was:

[0072] Where n is the number of groups of collected data 10, . is the recognition quality calculated for the i-th group of data, . The half-load mass of the vehicle is 8000kg. After calculation, the root mean square error is 320kg. According to the error situation, the gain coefficient in the RLS algorithm is . Make adjustments. In this embodiment, The initial value was adjusted from 0.8 to 0.9. After recalculation, the root mean square error was reduced to 180kg, and the accurate vehicle mass after compensation was finally determined to be 10300kg. After comparison with the actual vehicle mass, the error was within an acceptable range, verifying the feasibility and accuracy of this method in conventional driving scenarios.

[0073] In actual application, verification under complex driving scenarios: For verification of complex driving scenarios, an engineering vehicle was selected for driving test on mountain roads with uphill and downhill, frequent acceleration and deceleration. The test section includes a 5-kilometer uphill road with a slope of about 10°, followed by a 3-kilometer downhill road with a slope of about -8°, and frequent acceleration and deceleration operations during driving.

[0074] During the data collection phase, the acceleration sensor dynamically adjusts the collection frequency according to the vehicle's driving state due to frequent changes in the vehicle's driving state. In scenarios where the vehicle's state changes significantly, such as acceleration, deceleration, and climbing, the collection frequency is increased to collect data once every 100 milliseconds. At the same time, the device continuously obtains the motor's real-time torque data from the CAN bus. During the entire driving process, a total of 800 sets of acceleration data and corresponding motor torque data were collected.

[0075] The collected data is input into the signal processing module for calibration. The signal processing module also performs temperature compensation calibration and zero drift calibration on the signals collected by the acceleration sensor and the motor torque signal obtained from the CAN bus. After the calibration is completed, the recursive least squares method (RLS) is used to preliminarily identify the vehicle mass, and the preliminary calculation shows that the vehicle mass is 12,000 kg. The extended Kalman filter algorithm (EKF) is used to identify the slope of the vehicle, and the slope is accurately identified as 10° in the climbing stage and -8° in the downhill stage.

[0076] The vehicle driving force is calculated based on the core formula. During the climbing phase, is 3500N・m, the drive axle ratio is 4.2, the wheel radius is 0.48m, M is the initially identified vehicle mass of 12000kg, f is 0.06, β is 10°, and the vehicle driving force is 25347N. In the downhill stage, The driving force of the vehicle is calculated to be -10245N (the negative sign indicates that the driving force is in the opposite direction of travel and acts as a brake).

[0077] Data was collected multiple times and the calculation steps were repeated for a total of 15 times to obtain multiple sets of preliminary identification mass and corresponding vehicle driving force data. The identification mass obtained by multiple calculations was compared with the vehicle half-load mass of 8500kg, and the root mean square error (RMSE) was used to calculate the error. The initial root mean square error was calculated to be 450kg. According to the error situation, the gain coefficient in the RLS algorithm was adjusted. After multiple adjustments and calculations, the root mean square error was finally reduced to 250kg, and the accurate vehicle mass after compensation was determined to be 11800kg. Compared with the actual vehicle mass, the error was within a reasonable range, indicating that this method can still accurately calculate the vehicle mass in complex driving scenarios, showing good adaptability and reliability.

[0078] In summary, the present invention can significantly improve the accuracy of vehicle mass calculation through multi-source data fusion and error compensation strategy. Compared with traditional methods, it can more accurately reflect the actual mass of the vehicle under various complex working conditions, and the error is greatly reduced, meeting the demand for high-precision monitoring of vehicle mass. In addition, real-time and accurate vehicle mass data provides a strong guarantee for vehicle driving safety. When the vehicle mass is close to or exceeds the specified load, the timely reminder function can let the driver know in advance and take corresponding measures, effectively avoiding safety problems such as longer braking distance, increased tire wear, and decreased vehicle handling performance caused by overloading, and reducing the probability of traffic accidents. Furthermore, accurate vehicle mass calculation helps to optimize vehicle performance. In terms of vehicle power system control, the engine output power, gearbox shifting strategy, etc. can be more reasonably adjusted according to the accurate vehicle mass, thereby improving energy utilization efficiency and reducing fuel consumption. In the adjustment of the vehicle suspension system and the braking system, it can also be optimized according to the actual mass of the vehicle to improve the comfort and handling stability of the vehicle.

[0079] For the dynamic vehicle mass calculation method provided in the above embodiment, an embodiment of the present invention provides a dynamic vehicle mass calculation device, which is applied to a dynamic vehicle mass calculation system, see Figure 4 A structural schematic diagram of a dynamic vehicle mass calculation device is shown, the device includes the following parts: An information acquisition module 402 acquires acceleration data of a dynamic vehicle through an acceleration sensor and acquires motor torque data through a controller area network; The mass calculation module 404 performs calibration processing on the acceleration data and the motor torque data, and uses the recursive least square method to perform signal processing on the calibrated data to determine the initial vehicle mass, and uses the Kalman filter algorithm to identify the slope of the vehicle to determine the initial vehicle slope; The error compensation module 406 determines the initial vehicle driving force based on the initial vehicle mass and the initial vehicle driving slope by means of a preset vehicle driving force calculation model, and performs error compensation processing based on the initial vehicle driving force, the initial vehicle mass and the preset vehicle reference mass to determine the compensated target vehicle mass.

[0080] The above-mentioned dynamic vehicle mass calculation device provided in the embodiment of the present application determines the dynamic vehicle mass by performing error compensation through multi-source fusion data, which can significantly improve the accuracy of vehicle mass calculation.

[0081] In one embodiment, when performing the step of calibrating the acceleration data and the motor torque data, the mass calculation module 404 is also used to: obtain the current ambient temperature data, and use the ambient temperature data to perform temperature compensation on the acceleration data and the motor torque data to determine the compensated data; perform zero drift calibration on the compensated data to determine the calibrated data.

[0082] In one embodiment, when performing the step of performing temperature compensation processing on the acceleration data and the motor torque data using the ambient temperature data, the mass calculation module 404 is further used to: perform temperature compensation processing on the acceleration data and the motor torque data based on the preset temperature compensation parameter table and the current ambient temperature data by using a preset temperature compensation correction function, wherein the preset temperature compensation correction function is:

[0083] in, is the compensated data, is the original collected data, is the temperature compensation coefficient determined by the calibration experiment, T is the current ambient temperature, It is the reference temperature for calibration experiments.

[0084] In one embodiment, when performing a zero-point drift calibration on the compensated data and determining the calibrated data, the mass calculation module 404 is also used to: statically collect the drift amounts corresponding to the acceleration data and the motor torque data for a preset number of times, and determine the mean of the drift amounts as the target drift amounts corresponding to the acceleration data and the motor torque data, respectively; and determine the difference between the compensated data and the target drift amount as the calibrated data.

[0085] In one embodiment, when performing signal processing on the calibrated data using the recursive least squares method to determine the initial vehicle mass, the mass calculation module 404 is also used to: obtain a gain coefficient and extract a characteristic vector of the calibrated acceleration data and motor torque data; input the gain coefficient and the characteristic vector into the recursive least squares method calculation model to determine the initial vehicle mass.

[0086] In one implementation, the quality calculation module 404 further includes a recursive least squares calculation model, and the recursive least squares calculation model is:

[0087] in, is the estimated value of vehicle mass calculated for the kth time, is the vehicle mass calculated last time, is the gain coefficient, which is used to control the update amplitude of the model. To observe the driving force, is the characteristic vector containing acceleration and torque, is the parameter vector to be estimated.

[0088] In one embodiment, when using the Kalman filter algorithm to identify and process the slope of the vehicle and determine the initial vehicle driving slope, the above-mentioned mass calculation module 404 is also used to: determine the estimated value of the current vehicle driving slope by summing the estimated value of the previous vehicle driving slope and the process noise; substitute the estimated value of the current vehicle driving slope into a preset nonlinear measurement function, and determine the initial vehicle driving slope by summing the substitution result and the measurement noise.

[0089] In one embodiment, when performing the step of determining the initial vehicle driving force based on the initial vehicle mass and the initial vehicle driving slope by using a preset vehicle driving force calculation model, the error compensation module 406 is further used to: obtain a preset drive axle speed ratio and a wheel radius of the vehicle, substitute the preset drive axle speed ratio, wheel radius, motor torque data, initial vehicle mass and initial vehicle driving slope into the preset vehicle driving force calculation model to determine the initial vehicle driving force, wherein the preset vehicle driving force calculation model is:

[0090] in, is the real-time torque of the motor obtained from the CAN bus, the drive axle ratio and wheel radius are the inherent parameters of the vehicle, M is the initial vehicle mass, f is the vehicle driving resistance coefficient, and β is the initial vehicle driving slope.

[0091] In one embodiment, when performing error compensation processing based on the initial vehicle driving force, the initial vehicle mass and the preset vehicle reference mass to determine the compensated target vehicle mass, the above-mentioned error compensation module 406 is also used to: perform root mean square error calculation processing on the initial vehicle mass and the preset vehicle reference mass to determine the average error value, and use the average error value to update the gain coefficient in the recursive least squares calculation model to determine the target gain coefficient; use the target gain coefficient and the initial vehicle driving force to update the recursive least squares calculation model, and use the updated model to recalculate the vehicle mass to determine the target vehicle mass.

[0092] The device provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0093] An embodiment of the present invention provides a server. Specifically, the server includes a processor and a storage device. The storage device stores a computer program, and when the computer program is executed by the processor, it executes the method described in any one of the above-mentioned embodiments.

[0094] Figure 5 A structural diagram of a server provided in an embodiment of the present invention, the server 100 includes: a processor 50, a memory 51, a bus 52 and a communication interface 53, wherein the processor 50, the communication interface 53 and the memory 51 are connected via the bus 52; the processor 50 is used to execute an executable module stored in the memory 51, such as a computer program.

[0095] The memory 51 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 53 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0096] The bus 52 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0097] Among them, the memory 51 is used to store programs, and the processor 50 executes the program after receiving the execution instruction. The method executed by the device for flow process definition disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 50 or implemented by the processor 50.

[0098] The processor 50 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 50. The above processor 50 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor to be executed, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 51, and the processor 50 reads the information in the memory 51 and completes the steps of the above method in combination with its hardware.

[0099] The computer program product of the readable storage medium provided in the embodiment of the present invention includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the methods described in the previous method embodiments. The specific implementation can be referred to the previous method embodiments, which will not be repeated here.

[0100] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0101] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A dynamic vehicle mass calculation method, characterized in that: The method is applied to a dynamic vehicle mass calculation system, and the method comprises: Acquire dynamic vehicle acceleration data through acceleration sensors and motor torque data through controller area networks; Calibrate the acceleration data and the motor torque data, perform signal processing on the calibrated data using a recursive least squares method to determine an initial vehicle mass, and identify the vehicle driving slope using a Kalman filter algorithm to determine an initial vehicle driving slope; By presetting a vehicle driving force calculation model, the initial vehicle driving force is determined based on the initial vehicle mass and the initial vehicle driving slope, and error compensation processing is performed based on the initial vehicle driving force, the initial vehicle mass and the preset vehicle reference mass to determine the compensated target vehicle mass.

2. The dynamic vehicle mass calculation method according to claim 1, characterized in that: The step of calibrating the acceleration data and the motor torque data comprises: Acquire current ambient temperature data, and use the ambient temperature data to perform temperature compensation processing on the acceleration data and the motor torque data to determine compensated data; Perform zero drift calibration on the compensated data to determine the calibrated data.

3. The dynamic vehicle mass calculation method according to claim 2, characterized in that: The step of performing temperature compensation processing on the acceleration data and the motor torque data using the ambient temperature data comprises: The acceleration data and the motor torque data are subjected to temperature compensation processing by a preset temperature compensation correction function based on a preset temperature compensation parameter table and the current ambient temperature data, wherein the preset temperature compensation correction function is: in, is the compensated data, is the original collected data, is the temperature compensation coefficient determined by the calibration experiment, T is the current ambient temperature, It is the reference temperature for calibration experiments.

4. The dynamic vehicle mass calculation method according to claim 2, characterized in that: The step of performing zero drift calibration on the compensated data and determining the calibrated data comprises: By statically collecting the drift amounts corresponding to the acceleration data and the motor torque data for a preset number of times, and determining the average of the drift amounts as the target drift amounts corresponding to the acceleration data and the motor torque data; The difference between the compensated data and the target drift amount is determined as the calibrated data.

5. The dynamic vehicle mass calculation method according to claim 1, characterized in that: The step of using the recursive least square method to perform signal processing on the calibrated data to determine the initial vehicle mass includes: Obtaining a gain coefficient, and extracting a characteristic vector of the calibrated acceleration data and the motor torque data; The gain coefficient and the characteristic vector are input into a recursive least squares calculation model to determine the initial vehicle mass.

6. The dynamic vehicle mass calculation method according to claim 5, characterized in that: The recursive least squares calculation model is: in, is the estimated value of vehicle mass calculated for the kth time, is the vehicle mass calculated last time, is the gain coefficient, which is used to control the update amplitude of the model. To observe the driving force, is the characteristic vector containing acceleration and torque, is the parameter vector to be estimated.

7. The dynamic vehicle mass calculation method according to claim 1, characterized in that: The step of using the Kalman filter algorithm to identify the slope of the vehicle and determine the initial vehicle slope includes: The sum of the estimated value of the previous vehicle driving slope and the process noise is determined as the estimated value of the current vehicle driving slope; The estimated value of the current vehicle driving slope is substituted into a preset nonlinear measurement function, and the sum of the substitution result and the measurement noise is determined as the initial vehicle driving slope.

8. The dynamic vehicle mass calculation method according to claim 1, characterized in that: The step of determining the initial vehicle driving force based on the initial vehicle mass and the initial vehicle driving slope by presetting the vehicle driving force calculation model comprises: Obtain a preset drive axle speed ratio and a wheel radius of the vehicle, substitute the preset drive axle speed ratio, the wheel radius, the motor torque data, the initial vehicle mass and the initial vehicle driving slope into a preset vehicle driving force calculation model to determine the initial vehicle driving force, wherein the preset vehicle driving force calculation model is: in, is the real-time torque of the motor obtained from the CAN bus, the drive axle ratio and wheel radius are the inherent parameters of the vehicle, M is the initial vehicle mass, f is the vehicle driving resistance coefficient, and β is the initial vehicle driving slope.

9. The dynamic vehicle mass calculation method according to claim 1, characterized in that: The step of performing error compensation processing based on the initial vehicle driving force, the initial vehicle mass and a preset vehicle reference mass to determine a compensated target vehicle mass comprises: Performing root mean square error calculation processing on the initial vehicle mass and the preset vehicle reference mass to determine an average error value, and using the average error value to update a gain coefficient in a recursive least squares method calculation model to determine a target gain coefficient; The target gain coefficient and the initial vehicle driving force are used to update the recursive least squares calculation model, and the updated model is used to recalculate the vehicle mass to determine the target vehicle mass.

10. A dynamic vehicle mass calculation device, characterized in that: The device is applied to a dynamic vehicle mass calculation system, and comprises: An information acquisition module, which acquires acceleration data of a dynamic vehicle through an acceleration sensor and acquires motor torque data through a controller area network; A mass calculation module, which performs calibration processing on the acceleration data and the motor torque data, and uses a recursive least squares method to perform signal processing on the calibrated data to determine an initial vehicle mass, and uses a Kalman filter algorithm to identify a vehicle driving slope to determine an initial vehicle driving slope; The error compensation module determines the initial vehicle driving force based on the initial vehicle mass and the initial vehicle driving slope by using a preset vehicle driving force calculation model, and performs error compensation processing based on the initial vehicle driving force, the initial vehicle mass and a preset vehicle reference mass to determine the compensated target vehicle mass.

11. A server, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the method according to any one of claims 1 to 9.

Images