Vehicle mass dynamic estimation system and method

Abstract

The invention relates to the technical field of vehicle dynamic parameter estimation, in particular to a vehicle mass real-time estimation method and system based on a composite weighting algorithm and embedded equipment. The method comprises the steps that the speed, acceleration and driving torque data in the vehicle running process are collected; calculating a mass estimation value of each data point through a preset dynamic model; a composite weighting algorithm is constructed by adopting a Gaussian weight method and an entropy method, the Gaussian weight method sets an expected value based on wagon balance data to carry out weight distribution, and the entropy method carries out objective weighting according to a data dispersion degree; carrying out fusion calculation on the two weighting results, and outputting a final quality estimation value; and wagon balance data are updated in real time through serial port communication, and algorithm parameters are dynamically optimized. The method is realized on an STM32 embedded platform, and can be effectively applied to commercial vehicle-mounted quality monitoring and intelligent logistics management systems.

Description

Technical Field

[0001] The present invention relates to the field of vehicle dynamic parameter monitoring technology, and more particularly to a system and method for dynamic vehicle mass estimation based on a composite weighted algorithm. Specifically, the present invention is applicable to commercial vehicle mass monitoring, intelligent logistics management, and vehicle dynamics control systems. It can accurately estimate vehicle mass changes in real time, providing key parameter support for vehicle safety control, energy consumption management, and intelligent scheduling. Background Art

[0002] Dynamic vehicle mass estimation is a key technology for intelligent logistics and commercial vehicle safety control. With the improvement of on-board sensor accuracy and the development of edge computing technology, vehicle mass estimation methods have made significant progress. Existing technologies primarily use analysis methods based on dynamic models and estimation algorithms based on machine learning.

[0003] However, existing vehicle mass estimation technologies still face numerous challenges and limitations. First, methods based on purely physical models experience significant declines in estimation accuracy under complex operating conditions (such as hilly driving and bumpy roads). Second, while deep learning-based methods perform well in certain scenarios, they suffer from insufficient model generalization and high computational resource requirements. Furthermore, existing methods struggle to effectively incorporate prior information, such as scale calibration data, resulting in insufficient stability in practical engineering applications.

[0004] Specifically, the existing technology has the following prominent problems:

[0005] 1. Traditional dynamic model methods have stringent requirements on sensor accuracy and cannot adapt to changes in vehicle configuration;

[0006] 2. Data-driven methods require a large amount of labeled data for training, which can be challenging to obtain in actual deployment.

[0007] 3. Although the hybrid method combines the advantages of the first two methods, it has high computational complexity and cannot meet the real-time requirements of vehicle systems;

[0008] 4. Existing solutions generally lack the effective use of prior information and are unable to achieve the organic integration of calibration data and real-time estimation.

[0009] Therefore, there is an urgent need to develop a new dynamic estimation method for vehicle mass that can take into account model accuracy and adaptability to working conditions while ensuring computational efficiency, while realizing the intelligent fusion of prior knowledge and real-time data, and providing reliable quality parameters for vehicle safety control and logistics management.

[0010] The above content is only used to assist in understanding the technical solution of the present invention and does not constitute an admission that the above content is prior art. Summary of the Invention

[0011] The purpose of the present invention is to provide a vehicle mass dynamic estimation method, system and embedded device based on a composite weighted algorithm, aiming to solve the problems existing in the existing technology such as insufficient real-time performance, difficulty in integrating prior knowledge, and poor adaptability to complex working conditions. By integrating the composite weighting strategy of the Gaussian weighting method and the entropy method, combined with embedded optimization technology, efficient, accurate and dynamically updateable vehicle mass estimation can be achieved, providing reliable parameter support for intelligent logistics management and vehicle safety control.

[0012] To achieve the above object, the present invention provides a method for dynamically estimating vehicle mass, the method comprising the following steps:

[0013] S1, data acquisition and preprocessing module

[0014] S11. Multi-source data collection:

[0015] Vehicle operation data is collected in real time through the vehicle bus (CAN bus / serial port), including vehicle speed v (unit: km / h), acceleration a (unit: m / s²), drive torque T (unit: N·m), wheel radius R, transmission efficiency η and other parameters.

[0016] S12. Data screening mechanism:

[0017] A multi-level screening strategy was used to ensure that the data met the applicable conditions of the kinetic model:

[0018] Primary screening: Based on motion state threshold (vehicle speed v < 30km / h, acceleration 0.04m / s 2 ≤ a ≤1.0m / s 2 ) Eliminate invalid data segments.

[0019] Secondary screening: Further optimize data reliability through signal quality analysis (e.g., noise level, signal continuity).

[0020] S13. Data standardization:

[0021] The filtered data is normalized to eliminate dimensional differences. The formula is:

[0022]

[0023] in is the mass estimate of the i-th data point, min(m) and max(m) are the minimum and maximum values ​​of the mass data, respectively.

[0024] S2, quality estimation and composite weighted calculation module

[0025] S21. Kinetic model calculation:

[0026] This section starts from the vehicle driving equation and derives the calculation formula of the mass estimation value in detail.

[0027] S211. Vehicle driving equation: According to vehicle dynamics theory, the driving force of a vehicle must overcome all driving resistance and satisfy the force balance relationship:

[0028]

[0029] in:

[0030] F is the driving force (N);

[0031] ·F f is the rolling resistance (N);

[0032] ·F a is the air resistance (N);

[0033] ·F i is the slope resistance (N);

[0034] m is the vehicle mass (kg);

[0035] a is the vehicle acceleration (m / s²).

[0036] S212, resistance expansion:

[0037] Driving force F:

[0038] It is determined by the power system output torque T (N·m), transmission efficiency η, final reducer transmission ratio i0, and wheel radius R (m), and the formula is:

[0039]

[0040] Rolling resistance F f :

[0041] Related to vehicle mass m, rolling resistance coefficient f, and acceleration due to gravity g (9.81 m / s²):

[0042]

[0043] Air resistance F a :

[0044] The product of air density ρ (kg / m³), vehicle speed v (m / s), air resistance coefficient and frontal area Related:

[0045]

[0046] Slope resistance F i :

[0047] If the slope is θ, then:

[0048]

[0049] S213. Simplified force balance equation:

[0050] Ignore the slope resistance (F i =0), and substitute each resistance term into the force balance equation:

[0051]

[0052] S214. Calculate mass m:

[0053] Rearranging the equation into a linear expression for m:

[0054]

[0055] Finally, the quality estimation formula is obtained:

[0056]

[0057] in:

[0058] T: driving torque (N·m), measured by the vehicle-mounted sensor;

[0059] i0: final drive ratio, hardware parameter (dimensionless);

[0060] η: Transmission efficiency (0~1), requires calibration;

[0061] R: wheel radius (m), hardware parameter;

[0062] ρ: air density (1.225 kg / m³, under standard atmospheric conditions);

[0063] v: vehicle speed (m / s), obtained by the vehicle speed sensor;

[0064] ·C d A: The product of the air resistance coefficient and the frontal area (m²), which needs to be calibrated according to the vehicle model;

[0065] f: rolling resistance coefficient (dimensionless), typically 0.01-0.02;

[0066] a: acceleration (m / s²), obtained by the accelerometer.

[0067] S215. Extended explanation of slope effects:

[0068] If the slope resistance F is to be considered i , the mass formula is modified to:

[0069]

[0070] S22, composite weighted algorithm:

[0071] S221, Gaussian weight method:

[0072] The weight is assigned based on the scale calibration value μ (prior knowledge), and the formula is:

[0073]

[0074] Normalized weights:

[0075]

[0076] S222, Entropy Method:

[0077] Objectively assign weights based on the degree of data dispersion. The steps include:

[0078] S2221 Calculate probability matrix :

[0079]

[0080] S2222 Calculate Information Entropy :

[0081]

[0082] S2223 Determine weight:

[0083]

[0084] S223, Weight Fusion and Quality Estimation:

[0085] The two sets of weights are fused through the dynamic credibility coefficient α to output the final quality estimate:

[0086]

[0087] Where α is dynamically adjusted according to the data update frequency and discreteness:

[0088] When the scale data is updated frequently, α↑ (biased towards Gaussian weighting);

[0089] When the data dispersion is large, α↓ (biased entropy method).

[0090] S3, embedded optimization and dynamic update module

[0091] S31, multi-level cache architecture:

[0092] In view of the memory limitation of the STM32F1 platform (64KB RAM), a block data caching mechanism is designed to retain only the valid data in the current calculation window to avoid memory overflow.

[0093] S32, Parallel Computing Optimization:

[0094] The Gaussian weighting method and the entropy method are executed in parallel in independent threads to reduce computational latency.

[0095] Use fixed-point operations instead of floating-point operations to improve real-time performance.

[0096] S33, parameter update:

[0097] By receiving the calibration data of the weighbridge, the expected value μ and standard deviation σ are dynamically adjusted to ensure that the algorithm parameters match the actual working conditions.

[0098] S34, anomaly detection and verification:

[0099] Continuously monitor the fluctuation trend of the quality estimate value. If it exceeds the threshold (such as a single jump > 5%), trigger data re-collection;

[0100] ·Ensure the integrity of communication data through CRC check.

[0101] S4, system output and application module

[0102] S41, real-time quality display:

[0103] Output the final mass estimate m through the OLED screen or vehicle terminal final And confidence index:

[0104]

[0105] S42, intelligent control interface:

[0106] Upload quality data to the cloud logistics management system to optimize cargo scheduling;

[0107] Trigger safety control strategies (such as overload alarm and speed limit control).

[0108] Optionally, after outputting the final quality estimation value through the composite weighted algorithm, the method further includes:

[0109] Embedding the vehicle mass dynamic estimation system into an onboard controller (such as an STM32F1 microcontroller);

[0110] Real-time collection of vehicle operation data through the vehicle bus, including vehicle speed, acceleration, driving torque and environmental parameters;

[0111] Use a multi-level cache management architecture to dynamically store valid data segments to avoid memory overflow;

[0112] Inputting real-time data into the composite weighted algorithm module to calculate the current quality estimate and confidence index;

[0113] The mass estimation results are combined with vehicle status information to trigger safety strategies (such as overload alarm and power limitation) through the intelligent control interface.

[0114] To achieve the above objectives, the present invention proposes a vehicle mass dynamic estimation system based on a composite weighted algorithm, the system comprising:

[0115] Data acquisition module: configured to obtain vehicle operation data through CAN bus or serial port;

[0116] Composite weighted calculation module: realizes the parallel calculation and weight fusion of Gaussian weight method and entropy method;

[0117] Dynamic calibration module: receives scale data via serial communication and updates expected value μ and standard deviation σ in real time;

[0118] Embedded optimization module: Design lightweight algorithms for the STM32 platform, supporting multi-threaded parallel computing and memory block management.

[0119] In addition, to achieve the above-mentioned object, the present invention further provides an on-vehicle electronic control unit, the electronic control unit comprising:

[0120] Memory, storing vehicle mass estimation program and dynamic calibration parameters;

[0121] A processor, configured to implement the composite weighted algorithm as claimed in claim 1 when executing the program;

[0122] Real-time data acquisition interface, connecting vehicle sensors and communication buses;

[0123] Output module, which displays the quality estimation results and confidence level through an OLED display or an in-vehicle terminal.

[0124] In addition, to achieve the above-mentioned object, the present invention further provides a computer-readable storage medium, wherein a vehicle mass estimation program is stored on the storage medium, and when the program is executed by a processor, the following steps are implemented:

[0125] 1. Collect vehicle operation data and perform multi-level screening;

[0126] 2. Calculate the mass estimate of each data point based on the kinetic model;

[0127] 3. Calculate weights using the Gaussian weighting method and entropy method in parallel;

[0128] 4. Dynamically fuse weights and output the final quality estimate;

[0129] 5. Update calibration parameters through serial communication to optimize algorithm performance.

[0130] This invention discloses a dynamic vehicle mass estimation method, system, and embedded device based on a composite weighted algorithm. The method includes: collecting vehicle operating data via onboard sensors; calculating instantaneous mass estimates based on a dynamic model; concurrently calculating weights using a Gaussian weighting method (relying on scale calibration values) and an entropy method (relying on data dispersion); dynamically fusing the weights to output a final mass estimate; and implementing embedded optimization technology to achieve low-latency, high-reliability, real-time estimation. This method provides precise mass parameter support for vehicle overload warnings, intelligent logistics scheduling, and energy consumption management, significantly improving estimation accuracy and system robustness under complex operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0131] Figure 1 This is a schematic diagram of the overall structure of the system involved in the embodiment of the present invention;

[0132] Figure 2 This is an overall framework diagram of an embodiment of a vehicle mass estimation method of the present invention;

[0133] Figure 3 A flowchart of a dynamics module model construction for an embodiment of a vehicle mass estimation method according to the present invention;

[0134] Figure 4 A flow chart for constructing a composite weighted algorithm model for an embodiment of a vehicle mass estimation method according to the present invention;

[0135] Figure 5 This is a schematic diagram of the embedded architecture of an embodiment of the vehicle mass estimation method of the present invention;

[0136] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0137] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0138] Reference Figure 1 , Figure 1 Schematic diagram of the overall structure of a vehicle mass estimation system according to an embodiment of the present invention.

[0139] like Figure 1As shown, the vehicle mass estimation system may include: a vehicle electronic control unit (ECU) 1001, onboard sensors 1002, a weighbridge 1003, a single-chip microcomputer (MCU) 1004, and an output application module 1005. The required communication method between components varies depending on the hardware interface. The ECU 1001 may include a traditional microprocessor, embedded memory, and a vehicle bus interface to receive and pre-process velocity, acceleration, engine torque, and gyroscope data collected by the onboard sensors 1002. The weighbridge 1003 connects to the system via a human-machine interface to measure static axle weight. The MCU 1004 utilizes an ARM Cortex-M architecture and integrates an analog-to-digital converter (ADC) to perform fusion calculations on multi-source sensor data. The output application module 1005 may include an OLED, an onboard display, an audible and visual alarm, and an interface with vehicle actuators to trigger a warning when the mass estimate exceeds a threshold. Data exchange between modules is achieved through the Controller Area Network (CAN) bus, and vehicle-mounted Bluetooth or T-box can be expanded for remote data transmission needs.

[0140] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation to the vehicle mass estimation system, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0141] like Figure 1 As shown, the single-chip microcomputer 1004 may also include an operating system, a communication module, a user interface module, a vehicle mass estimation system program, and a data processing program.

[0142] exist Figure 1 In the vehicle mass estimation system shown, the on-board sensor 1002 mainly collects the torque fed back by the strain gauge on the half-axle; the vehicle electronic control unit 1001 records the vehicle's driving speed, acceleration, and gear status; the single-chip microcomputer 1004 mainly includes a data screening program, a dynamic model calculation program, and a composite weighting method program, and executes the vehicle mass estimation method provided by the embodiment of the present invention.

[0143] Reference Figure 2 , Figure 2 The figure is an overall framework diagram of an embodiment of the vehicle mass estimation method of the present invention, and an embodiment of the vehicle mass estimation method of the present invention is proposed.

[0144] In an embodiment, the vehicle mass estimation method comprises the following steps:

[0145] Step S10: Acquire data collected by vehicle sensors.

[0146] It should be understood that the execution subject of this embodiment is a vehicle mass estimation device, which has functions such as data processing, data communication, and program execution.

[0147] In specific implementation, the driving data provided by a car company is used to judge the stability and quality of the test data output according to the length of the data output, and a longer section is intercepted as the test data set.

[0148] Step S20: A vehicle dynamics model is constructed by combining automobile theory, automobile structure, mechanical dynamics and other theories with empirical formulas and engineering practice.

[0149] In a specific embodiment, rolling resistance and air resistance are taken into account, and dynamics-related theories are used to construct vehicle motion equations. Newton's second and third laws, as well as empirical formulas, are used to complete the overall dynamics model for vehicle mass estimation.

[0150] Steps S30 and S40: Input the data obtained from the current vehicle dynamics model into the composite weighted algorithm for weight distribution and fusion calculation to facilitate the convergence of the results, and finally output the comprehensive vehicle mass estimation result.

[0151] It should be noted that the composite weighted algorithm includes two parts: subjective weighting method (Gaussian weight method) and objective weighting method (entropy method).

[0152] In practice, data is fed into both the subjective weighting module (Gaussian weighting) and the objective weighting module (entropy method) to obtain corresponding weights, which are then further integrated, calculated, and judged. This diversity of weighting methods helps the model improve the accuracy and credibility of assessments in complex environments.

[0153] Step S50: Dynamically optimize and verify the convergence of the fusion calculation results through a composite weighted algorithm to generate a comprehensive quality estimation value.

[0154] It should be understood that this step requires addressing weight drift caused by spatiotemporal asynchrony in multi-source data. In practice, a sliding window mechanism is used to iteratively correct the fusion results: every 100 milliseconds, the subjective weight coefficients (based on sensor confidence) output by the Gaussian weighting method and the objective weight coefficients (reflecting data dispersion) calculated by the entropy method are dynamically normalized, and the weight ratio is optimized using the least squares method. The Mahalanobis distance criterion is also introduced to eliminate outlier data points caused by sensor failure, ensuring the mathematical convergence of the fusion results.

[0155] Step S60: Implementing protocol encapsulation and multi-modal output of the mass estimation result through the vehicle microcontroller.

[0156] In a specific embodiment, a single-chip microcomputer 1004 converts the final quality estimate into a CAN data frame compliant with the SAE J1939 standard, which contains a 32-bit floating-point total quality value, an 8-bit quality status code, and a 16-bit CRC checksum. The output application module 1005 simultaneously performs three operations: displaying the quality on an OLED screen; sending a PWM modulated signal to an audible and visual alarm via a bus to trigger a level 3 warning when the estimate exceeds a threshold; and uploading the encrypted quality message to a cloud-based monitoring platform via a T-box. A 10ms watchdog timer is implemented in the system to ensure the real-time nature of output commands and the integrity of the execution chain.

[0157] like Figure 3 As shown, in the process of constructing the dynamic model of the vehicle mass estimation system, the core force balance equation is first established based on the vehicle longitudinal dynamics theory. The input parameters include the drive shaft torque T, the real-time vehicle speed v (calculated by the wheel speed pulse signal collected by the vehicle sensor 1002 after data processing), the acceleration a (obtained by fusing the three-axis accelerometer data with the gyroscope attitude angle), the comprehensive driving force F, the rolling resistance F f , air resistance F a , slope resistance F i And the inertia force ma, construct the total force balance relationship:

[0158]

[0159] Among them, the driving force F is accurately modeled by the transmission system parameters as follows:

[0160]

[0161] i0 is the measured transmission ratio, calculated from the engine output shaft speed and tire speed, and dynamically calibrated in real time. η is the transmission efficiency, calculated by interpolation of the speed-efficiency curve fitted by bench testing; R is the tire rolling radius, obtained based on data from a certain automaker. Rolling resistance F f Adopting the graded friction coefficient model:

[0162]

[0163] (f is dynamically adjusted according to the road surface type: 0.012-0.015 for asphalt roads and 0.02-0.025 for gravel roads)

[0164] Air resistance F a The Reynolds number correction term is introduced to improve the accuracy of high-speed conditions:

[0165]

[0166] (ρ is the air density after altitude compensation, updated by the pressure sensor data; Cd The dynamic drag coefficient calibrated by wind tunnel test; v ref is the reference wind speed). Slope resistance F i Combined with the pitch angle θ output by the inertial navigation unit (IMU), the following is calculated:

[0167]

[0168] (The slope angle θ is filtered by a sliding average to eliminate instantaneous bump interference)

[0169] In the simplified model, the air resistance term is ignored for the urban low-speed scenario, and the equation is reconstructed as:

[0170]

[0171] Then the quality calculation formula is derived:

[0172]

[0173] (μ is the dynamic correction factor for the rolling resistance coefficient, which is set and updated through the human-computer interaction interface)

[0174] To improve the adaptability of the model, an extended interface is added to support dynamic switching of multiple working conditions, such as introducing slope or center of mass offset conditions.

[0175] Finally, the model parameters (η, f, C d The data (e.g., data from a single-chip microcomputer) is written into the memory (FRAM) partition of the single-chip microcomputer 1004, supporting an erase and write lifespan of 100,000 times and stable operation in the industrial temperature range of -40°C to 125°C, forming a scalable and high-precision dynamics framework for vehicle mass estimation.

[0176] like Figure 4 As shown, Figure 4 This is a flow chart for constructing a composite weighted algorithm model for the vehicle mass estimation method embodiment of the present invention. During the specific implementation of the composite weighted algorithm in the vehicle mass estimation system, the system achieves high-precision mass estimation through multi-level data fusion and dynamic weight adjustment mechanisms.

[0177] The first step is sensor data collection and calibration. In a specific embodiment, the system first obtains the static axle weight measurement value provided by the scale 1003. It also receives the raw data collected by the vehicle-mounted sensors 1002 (including wheel speed sensors and gyroscopes) in real time.

[0178] The second is the dual-channel weight distribution module:

[0179] Subjective Weight Channel (Gaussian Weighting): A sensor benchmark library is established based on historical data statistics. Each sensor is assigned an expected value n (e.g., the theoretical output value of an accelerometer at a constant speed on a flat road) and a standard deviation σ (reflecting the sensor's long-term drift). When new data arrives, its deviation from the expected value is calculated to generate an initial weight. For example, if an accelerometer's instantaneous measurement deviates from the expected value by more than 3σ, its weight is automatically reduced to 30% of the baseline value.

[0180] Objective Weight Channel (Entropy Method): Within each 5-second window, the extreme value distribution of each sensor data is statistically analyzed and the data is normalized. The degree of chaos in the sensor data is assessed by calculating information entropy. Sensors with higher entropy values ​​(such as torque signals affected by electromagnetic interference) have their objective weights dynamically lowered.

[0181] Next, the calculation results are dynamically integrated and the credibility is evaluated:

[0182] 1. Working condition adaptive weight fusion

[0183] The system dynamically adjusts the subjective and objective weights based on real-time working parameters:

[0184] Low-speed steady-state condition (vehicle speed <30 km / h): Prioritizes subjective weighting (α=0.7), relying on preset sensor characteristics to ensure stability;

[0185] High-speed dynamic conditions (vehicle speed ≥ 30 km / h): Switches to objective weight dominance (α=0.3) to suppress the influence of environmental noise through data distribution characteristics;

[0186] Extreme environmental conditions (such as -20°C or heavy rain): Introduce temperature and humidity sensor data to correct the α value and reduce the weight of sensors that are susceptible to environmental interference (such as optical gyroscopes).

[0187] 2. Abnormal data suppression mechanism

[0188] Outlier detection: A sliding window chi-square test (window size 50 sampling points) is used to monitor the residual distribution of the fusion results in real time. If the residuals of three consecutive sampling points exceed the 2σ threshold, it is determined to be an abnormal condition and the following actions are triggered:

[0189] Pause the current fusion process and roll back to the last valid state;

[0190] Read the vehicle ECU fault code through the CAN bus and locate abnormal sensors (such as tire pressure sensor failure);

[0191] Force the weights of tagged sensors to zero in weight assignment and enable backup data sources (e.g., using weighbridge data to compensate for failed acceleration signals).

[0192] It should be noted that in extreme environmental conditions, it is necessary to use the extended interface of the dynamic model formula to upgrade the dynamic model to introduce new variables.

[0193] like Figure 5 As shown, Figure 5 This is a schematic diagram of the embedded architecture of an embodiment of the vehicle mass estimation method of the present invention. During the specific implementation of the embedded architecture of the vehicle mass estimation system, the system adopts a layered modular design to achieve efficient data processing and real-time control. The specific implementation method is as follows:

[0194] Data input module:

[0195] The system connects to onboard sensors 1002 (including wheel speed sensors and gyroscopes) via a CAN interface, acquiring raw signals at a 10 ms cycle. Hardware-level preprocessing of the raw data involves applying an RC low-pass filter to the wheel speed signal to eliminate high-frequency electromagnetic interference. Accelerometer data is converted to digital form using an on-chip ADC module (12-bit resolution) and compensated for temperature drift using a lookup table correction method using a built-in temperature sensor. CAN bus data integration involves the CAN bus component receiving engine torque, transmission ratio i0, and tire pressure data from the vehicle's electronic control unit 1001. The message format complies with the SAE J1939 standard (ID range 0x18F00 to 0x18FFF). The data reception rate is configured at 500 kbps, and a double buffer mechanism (ping-pong buffering) is used to prevent data loss.

[0196] CAN filters and data parsing, including filtering key messages through programmable CAN filters:

[0197] Set the ID whitelist (for example, 0x18F005 is a torque message, and 0x18F010 is a transmission ratio message). The data parsing unit converts the original CAN message into an engineering value. For example, the 2-byte data of the 0x18F005 message is converted into a torque value T (range 0-2000 Nm, resolution 0.1 Nm) using a linear mapping formula.

[0198] Memory management module, including static storage area: allocated to ferroelectric memory (FRAM) partition, storage of firmware code, dynamic model parameters (rolling friction coefficient f, drag coefficient C d ) and a safety threshold database, supporting stable read and write access over a wide temperature range of -40°C to 125°C. The dynamic computation area utilizes a DMA dual-channel SRAM (256 KB) to cache raw sensor data, intermediate fusion results, and alarm event logs in real time. The memory allocation strategy is as follows: high-priority tasks (such as the core quality estimation algorithm) occupy 60% of the memory to ensure real-time performance; low-priority tasks (such as historical data storage) are allocated on demand.

[0199] Static interface data processing involves executing floating-point-intensive operations through hardware (e.g., the ARM Cortex architecture): dynamic model solutions (iterating force balance equations every 5 ms) and matrix operations in composite weighted algorithms (e.g., entropy method probability matrix calculations). Static interfaces are configured in exclusive access mode to avoid data conflicts caused by multitasking.

[0200] Output and user interaction module, including multi-modal output control, such as protocol encapsulation, encapsulating the final mass estimate into two types of outputs: CAN bus message (ID 0x18F100, including total mass, status code and CRC check);

[0201] User Interface: A 7-inch industrial-grade touchscreen (800×480 resolution) displays dynamic quality curves, allows users to select road conditions, and input nominal weighing values. Physical buttons support manual query of historical overload records (stored in the FRAM partition).

[0202] Other embodiments or specific implementations of the vehicle mass estimation method and device of the present invention can refer to the above-mentioned method embodiments, and therefore have at least all the beneficial effects brought by the technical solutions of the above-mentioned embodiments, which will not be repeated here.

[0203] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0204] The serial numbers of the embodiments of the present invention are for descriptive purposes only and do not represent superiority or inferiority of the embodiments. In a unit claim that enumerates several means, several of these means may be embodied by the same item of hardware. The use of the terms first, second, and third, etc., does not denote any order; these terms should be interpreted as designations.

[0205] Through the above description of the embodiments, those skilled in the art will clearly understand that the methods of the above embodiments can be implemented by software plus the necessary general hardware platform. Of course, hardware can also be used, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The program of this product is stored in a storage medium (such as a read-only memory image (ROM) / random access memory (RAM), a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a single-chip microcomputer, an on-board computer, a computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0206] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A vehicle mass dynamic estimation method based on a composite weighted algorithm, characterized in that: The following steps are involved: S1. Real-time collection of vehicle operation data through the vehicle bus, and selection of valid data segments that meet the applicable conditions of the dynamic model; S2. Calculate the mass estimate of each data point based on the vehicle dynamics equation; S3, using a composite weighted algorithm to process the quality estimate; S4, fusing and calculating the two sets of weighted results according to a preset credibility coefficient, and outputting a final quality estimation value; S5. Establish a data communication link with the external calibration system to achieve dynamic parameter updates; S6. Set up an anomaly detection mechanism to start the data verification process when abnormal fluctuations in the estimated value are identified.

2. The method according to claim 1, characterized in that The step S1 comprises: S11. Collect multiple operating parameters including vehicle speed, acceleration, and driving torque through the vehicle bus protocol; S12. Adopt a multi-level data screening mechanism, including primary screening based on motion status and secondary screening based on signal quality; S13. The motion state-based screening includes identifying and classifying vehicle operating conditions.

3. The method according to claim 1, characterized in that The vehicle dynamics equation in step S2 includes: S21, driving force calculation unit, comprehensively considering the power system output characteristics and transmission parameters; S22, a resistance calculation unit, including a composite calculation model of rolling resistance and air resistance; S23, mass calculation unit, mass estimation model based on the force balance principle.

4. The method according to claim 1, wherein The step S3 of processing the quality estimation value using a composite weighted algorithm includes: S31, calculating a first set of weights by a Gaussian weighting method, wherein the weights are assigned based on a degree of deviation between the mass estimate and the calibration reference value; S32. Calculate a second set of weights using an entropy method, where the weight allocation is based on the discrete characteristics of the quality data distribution.

5. The method according to claim 4, characterized in that The step S31 includes: S311, receiving a reference mass value provided by an external calibration system; S312. adopting a probability weight allocation method based on normal distribution; S313, setting adaptively adjustable weight distribution parameters; The weight allocation method considers the following factors: The absolute deviation of the mass estimate from the reference value; The vehicle's operating status at the time of data collection; Statistical properties of historical data.

6. The method according to claim 4, characterized in that The step S32 includes: S321, performing standardization processing on the quality estimation sequence; S322. Evaluate the information content of each data point based on information entropy theory; S323, objectively assign weights based on the amount of information; The weight distribution process includes: Calculate the information entropy value of each data point; Determine the weight distribution ratio based on the entropy value; Smoothing of weights to improve stability.

7. The method according to claim 1, characterized in that In step S4, the two sets of weighted results are fused and calculated based on the preset credibility coefficients to output a final quality estimation value, wherein the fusion calculation process includes: S41, normalizing the two sets of weighted results to ensure that the sum of the weights is 1; S42. Use dynamic credibility coefficient to adjust the contribution ratio of the two groups of weights; S43, calculating a final mass estimate by weighted summation; S44. Calculate the confidence index of the estimation result.

8. The method according to claim 7, characterized in that The fusion calculation process in step S4 specifically includes: S411. Normalize the weights obtained by the Gaussian weighting method: ； S412. Normalize the weights obtained by the entropy method: ； S421. Dynamically adjust the credibility coefficient α based on data quality. The adjustment strategy includes: When the scale data is updated frequently, increase the α value; When the data dispersion is large, reduce the α value; S431. Calculation of final mass estimate: ； S441. Confidence index calculation: confidence 。 9. The method according to claim 1, characterized in that The data communication link with the external calibration system is established in step S5 to achieve dynamic parameter update. The update strategy includes: S51, regularly receiving weighbridge calibration data; S52, adjusting the reference mass value according to the calibration data; S53: Update weight distribution parameters.

10. The method according to claim 1, characterized in that The abnormality detection mechanism in step S6 is set to start the data verification process when abnormal fluctuations in the estimated value are identified, including: S61, continuous trend of the detection estimate; S62, triggering data re-collection when an abnormal pattern is detected; S63. Mark and exclude abnormal data.

11. A vehicle mass dynamic estimation system, characterized in that: include: a data acquisition module configured to acquire operating data via a vehicle bus; A data processing module, configured to execute the method according to any one of claims 1 to 7, comprising: Weighted calculation unit, which realizes the parallel calculation of Gaussian weight method and entropy method; Fusion calculation unit to complete weight normalization and weighted summation; Confidence assessment unit, which calculates the reliability of the quality estimate; Human-computer interaction module, used to display estimation results and system status; Calibration interface module, used to connect external calibration equipment.

12. The system according to claim 8, wherein: The data processing module is implemented on an embedded control unit and has the following features: Multi-level cache management architecture; Parallel computing capabilities; Real-time performance optimization; Supports online parameter calibration and algorithm update, the update methods include: Receive new parameters through the calibration interface; Automatically adjust computing strategies; Maintain the continuity of historical data.

13. A vehicle-mounted electronic control unit, characterized in that: The electronic control unit comprises a processor, a memory, and a computer program stored in the memory, wherein when the processor executes the program, the method according to any one of claims 1 to 7 is implemented, and the electronic control unit has: Real-time data acquisition interface; High-speed computing capability; Multi-task scheduling function; The program execution process includes: Initialization phase: loading algorithm parameters and calibration data; Operation phase: real-time processing of vehicle operation data; Maintenance phase: Receive parameter updates and algorithm optimizations.

14. A computer-readable storage medium, characterized in that A vehicle mass estimation program is stored, and when the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented and is applicable to a vehicle embedded system environment. The program on the storage medium includes: Data acquisition program module; Weighted calculation program module; Fusion computing program module; exception handler module; The program modules interact with each other through well-defined interfaces.

Images