A state perception system for multi-source data fusion of automotive electronic control units

Abstract

The present invention relates to the field of data processing technology, and in particular to a state perception system for multi-source data fusion of an automobile electronic control unit, the system comprising: collecting automobile driving speed, automobile vibration intensity and tire pressure data; clustering the driving speed and tire pressure data to obtain a number of clusters and continuous curve segments; obtaining the road condition similarity confidence of the continuous curve segments according to the difference in length and number of the continuous curve segments in each cluster; obtaining a driving state change index according to the difference between the driving speed at each moment and other moments; obtaining the possibility of abnormal vibration at each moment according to the difference in driving speed between different clusters at continuous moments; and judging the driving state of the automobile according to the correlation between the automobile vibration intensity and the possibility of abnormal vibration. The present invention analyzes the driving state of the automobile through multi-source data during the driving process of the automobile, so as to be used for the automobile electronic control system to control and manage the vehicle, thereby improving the accuracy of the automobile electronic control system.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a state perception system for multi-source data fusion of an automobile electronic control unit. Background Art

[0002] The automotive electronic control system is one of the most important components in modern cars, responsible for controlling and managing various electrical and electronic systems of the vehicle. With the development of the Internet of Things, intelligent driving has become more and more widespread, and the perception of vehicle driving status has become more and more important. It collects the driver's personal driving status information, judges the vehicle's driving status according to the different situations encountered by the driver while driving, and prompts the information when important situations occur in the vehicle.

[0003] Traditional automotive electronic control systems control and manage vehicles mainly by installing corresponding sensor elements on the vehicle to collect data for analysis. The data collected is mainly vibration sensor data. If the vehicle senses that a vibration value is greater than a preset threshold, a warning prompt will be issued to the automotive electronic control system, and the automotive electronic control system will control and manage the vehicle. However, when the vehicle is driving on an uneven road, the vehicle may be shaken due to the road surface, so that the vehicle's vibration sensor data may be greater than the preset threshold, thereby causing the vehicle to be falsely alarmed. On the other hand, vibration sensor data from a single source cannot ensure the accuracy of abnormal conditions, so that the vehicle's status information may be falsely alarmed due to the sensor, which interferes with the accuracy of the automotive electronic control system. Summary of the invention

[0004] The present invention provides a state perception system for multi-source data fusion of an automobile electronic control unit to solve the existing problems.

[0005] The state perception system of the automotive electronic control unit multi-source data fusion of the present invention adopts the following technical solution:

[0006] An embodiment of the present invention provides a state perception system for multi-source data fusion of an automotive electronic control unit, the system comprising the following modules:

[0007] Data acquisition module: used to collect the vehicle's driving speed, vehicle vibration intensity and tire pressure data at each moment;

[0008] Data processing module: used to cluster the driving speed and tire pressure data to obtain several clusters and several continuous curve segments in each cluster; obtain the road condition similarity confidence of the continuous curve segment at the current moment according to the difference in the length and number of the continuous curve segments in the cluster at the current moment; obtain the driving state change index at the current moment according to the difference in the driving speed at the current moment and the driving speed at other moments in the cluster; obtain the possibility of abnormal vibration at the current moment according to the difference in the driving speed on the previous continuous curve segment of the continuous curve segment at the current moment;

[0009] Abnormal detection module: used to obtain the driving state abnormality coefficient according to the correlation between the vehicle vibration intensity and the possibility of abnormal vibration; and to issue a vehicle driving vibration warning according to the driving state abnormality coefficient.

[0010] Furthermore, clustering the driving speed and tire pressure data to obtain a plurality of clusters and a plurality of continuous curve segments in each cluster includes:

[0011] The clustering space is constructed with the driving speed and tire pressure as the coordinate axes respectively, and all the driving speed and tire pressure data are mapped into the clustering space to form a number of discrete data points. The DBSCAN density clustering algorithm is used to cluster all the discrete data points according to the preset clustering radius and the preset number of samples to obtain a number of clusters;

[0012] The continuous data points collected at each cluster are connected to obtain several continuous curve segments in each cluster.

[0013] Furthermore, the obtaining of the road condition similarity confidence of the continuous curve segment at the current moment includes:

[0014] The current moment is recorded as the i-th moment, and the road condition similarity confidence of the continuous curve segment at the i-th moment is The calculation method is:

[0015]

[0016] In the formula, represents the number of continuous curve segments in the velocity cluster set at the i-th moment; Represents the number of special continuous curve segments in the velocity cluster set at the i-th moment.

[0017] Furthermore, the specific method of obtaining the speed cluster set includes:

[0018] The tire pressure range of all data points in each cluster is obtained, and the set consisting of all clusters that have an intersection with the tire pressure range of the cluster where the data point at the current moment is located is recorded as the speed cluster set at the current moment.

[0019] Furthermore, the specific method for obtaining the number of the special continuous curve segments includes:

[0020] In the speed cluster set at the i-th moment, the number of continuous curve segments whose length difference with the continuous curve segment at the i-th moment is less than a preset length threshold is recorded as the number of special continuous curve segments in the speed cluster set at the i-th moment.

[0021] Furthermore, the driving state change index at the current moment is obtained according to the difference between the driving speed at the current moment and the driving speed at other moments in the cluster, including:

[0022] The current moment is recorded as the i-th moment, and the calculation method of the driving state change index at the i-th moment is:

[0023]

[0024] In the formula, represents the driving state change index at the i-th moment; Represents the road condition similarity confidence of the continuous curve segment at the i-th moment; Indicates Driving speed at the moment; Indicates the cluster of the continuous curve segment at the i-th moment except the data point at the i-th moment. The driving speed of each data point; Represents the number of all data points in the cluster where the continuous curve segment at the i-th moment is located; Represents the Sigmoid function.

[0025] Furthermore, obtaining the possibility of abnormal vibration at the current moment according to the difference in driving speed on the previous continuous curve segment of the continuous curve segment at the current moment includes:

[0026] Record the current moment as the i-th moment, record the last moment of the nearest cluster of the i-th moment as the m-th moment in the nearest cluster of the i-th moment, and the probability of abnormal vibration at the i-th moment The calculation method is:

[0027]

[0028] In the formula, represents the driving state change index at the i-th moment; Indicates Driving speed at the moment; represents the driving speed at the mth moment in the nearest cluster at the i-th moment; represents the mean value of the driving speed of other data points in the cluster at the i-th time except the i-th time. represents the mean of the driving speeds of all data points in the nearest cluster at time i except the driving speed at time m; represents the Sigmoid function, Represents the absolute value function.

[0029] Furthermore, the specific method of obtaining the nearest neighbor cluster includes:

[0030] The previous continuous curve segment closest to the collection time of the continuous curve segment at the i-th moment is recorded as the nearest continuous curve segment at the i-th moment, and the cluster where the nearest continuous curve segment at the i-th moment is located is recorded as the nearest cluster at the i-th moment.

[0031] Furthermore, the driving state abnormality coefficient is obtained according to the correlation between the automobile vibration intensity and the possibility of abnormal vibration, including:

[0032] The abnormal vibration possibility at each moment is calculated to form an abnormal vibration possibility sequence, the Pearson correlation coefficient between the abnormal vibration possibility sequence and the sequence formed by the vehicle vibration intensity at each moment is calculated, and the Pearson correlation coefficient is recorded as the driving state abnormality coefficient.

[0033] Furthermore, the vehicle driving vibration warning according to the driving state abnormality coefficient includes:

[0034] A preset judgment threshold is set. When the driving state abnormality coefficient is less than or equal to the preset judgment threshold, and the vehicle vibration intensity of the vehicle at the current moment is greater than the preset vibration threshold, the vehicle electronic control system issues a warning prompt.

[0035] The beneficial effects of the technical solution of the present invention are: collecting the driving speed, vehicle vibration intensity and tire pressure data of the vehicle, and selecting multi-source data for comprehensive analysis is more accurate; clustering the driving speed and tire pressure data to obtain a number of clusters and continuous curve segments, which is helpful for the subsequent analysis of the driving state of the vehicle under the same road condition environment; according to the difference in the length and number of the continuous curve segments in each cluster, the road condition similarity confidence of the continuous curve segments is obtained as a preliminary reference for the driving state analysis; according to the difference between the driving speed at each moment and other moments, the driving state change index is obtained; according to the difference in the change of driving speed between different clusters at continuous moments, the possibility of abnormal vibration at each moment is obtained, thereby improving the accuracy of the vehicle electronic control system; the driving state of the vehicle is judged according to the correlation between the vehicle vibration intensity and the possibility of abnormal vibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 The present invention is a structural block diagram of a state perception system for multi-source data fusion of an automotive electronic control unit. DETAILED DESCRIPTION

[0038] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a state perception system for multi-source data fusion of an automotive electronic control unit proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0039] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0040] The specific scheme of the state perception system of multi-source data fusion of an automobile electronic control unit provided by the present invention is described in detail below with reference to the accompanying drawings.

[0041] See also Figure 1 , which shows a structural block diagram of a state perception system for multi-source data fusion of an automotive electronic control unit provided by an embodiment of the present invention, the system includes the following modules:

[0042] Data collection module 101: collects the vehicle's driving speed, vehicle vibration intensity and tire pressure data.

[0043] The purpose of this embodiment is to analyze the driving state of the car through multi-source data during the driving process of the car, so as to use the car electronic control system to control and manage the vehicle. Therefore, the driving speed and vibration intensity of the car at each moment are collected through the car electronic control system, and a pressure sensor is installed in the tire of the vehicle to collect the tire pressure of the vehicle at each moment during the driving process. The collection time in this embodiment is half a year, and the collection frequency is once per minute. This example is used for description. Other values ​​can be set in other embodiments, and this embodiment does not specifically limit it. It should be noted that since there are multiple tires on the vehicle, the tire pressure data described in this embodiment is the average of multiple tire pressures.

[0044] Data processing module 102:

[0045] (1) According to different speeds under different road conditions, the driving speed and tire pressure are clustered to obtain several clusters and several continuous curve segments; according to the differences between the continuous curve segments of different clusters, the road condition similarity confidence of the continuous curve segment at the current moment is obtained.

[0046] It should be noted that under different road conditions, the tire pressure of a car is different when it is driving, and the tire pressure changes accordingly as the road conditions change; for example, on a highway, the normal driving speed of a car is stable, the road conditions are good, and the speed fluctuation is small, so the tire pressure changes less. On a country road with poor road conditions, because the road conditions are unknown and the road surface is uneven, the car's driving speed is low, and the tire pressure fluctuates greatly due to the different degrees of squeezing caused by the road surface. Therefore, the tire pressure is classified according to the size of the car's tire pressure. The same tire pressure represents the same road condition, and the driving state of the car can be roughly judged according to the driving speed size under each tire pressure.

[0047] Specifically, a clustering space is constructed with driving speed and tire pressure as coordinate axes respectively, all driving speed and tire pressure data are mapped into the clustering space to form a number of discrete data points, and the DBSCAN density clustering algorithm is used to cluster all discrete data points according to a preset clustering radius and a preset number of samples to obtain a number of clusters; it should be noted that this embodiment is described using a preset clustering radius of 10 and a clustering sample number of 5 as an example, the DBSCAN density clustering algorithm is an existing well-known technology, other embodiments may use other methods, which will not be described in detail in this embodiment.

[0048] Furthermore, the driving speed and tire pressure of the data points in each cluster after clustering are similar. Since the collected data is time series, and the road condition and speed will not change greatly when the vehicle is driving normally, multiple data points representing the same road condition and speed in each cluster constitute a segment of time series data. Based on the above logic, this embodiment connects the continuous data points at the collection moment in each cluster to obtain several continuous curve segments, each continuous curve segment represents a driving state, and the length of the continuous curve segment represents the maintenance time of the driving state.

[0049] It should be further explained that the driving states of vehicles under the same road conditions are similar, that is, when the road conditions are the same, the vehicles maintain similar driving speeds, the tire pressures of the vehicles are similar, and the maintenance time of the stable driving state is similar. If the data at the current moment is clustered in a cluster with similar driving speeds, tire pressures, and maintenance time, and the larger the cluster size, the higher the confidence of the road condition similarity of the data at the current moment, and it can be used as a reference when analyzing vehicle abnormalities. If the cluster where the data at the current moment is located is a small cluster, it means that the data at the current moment is more likely to be due to the vehicle being in different driving states such as acceleration and deceleration, and there is an offset from the data generated under normal road conditions, thereby causing it to be far away from the cluster where the normal data is located. Therefore, this embodiment obtains the road condition similarity confidence of the continuous curve segment at the current moment based on the length difference between the cluster where the continuous curve segment at the current moment is located and other continuous curve segments.

[0050] Specifically, the current moment is recorded as the i-th moment, the tire pressure range of all data points in each cluster is obtained, and the set consisting of all clusters that have an intersection with the tire pressure range of the cluster where the data point at the current moment is located is recorded as the speed cluster set at the current moment. Then, the road condition similarity confidence of the continuous curve segment at the i-th moment is calculated as follows:

[0051]

[0052] In the formula, Represents the confidence level of the road condition similarity of the continuous curve segment at the i-th moment; represents the number of continuous curve segments in the velocity cluster set at the i-th moment whose length difference with the continuous curve segment at the i-th moment is less than the preset length threshold; Indicates the number of continuous curve segments in the speed cluster set at the i-th moment; in this embodiment, the preset length threshold is 2 minutes, which is used as an example for description, and other embodiments may set other values.

[0053] Indicates the continuous curve segment at the i-th moment in the velocity cluster set and the The proportion of continuous curve segments with similar continuous curve segments at the time to all continuous curve segments in the velocity cluster set is When the value is larger, the frequency of the continuous curve segment in the speed cluster set where the continuous curve segment at the i-th moment is located in the historical speed change is higher, the current continuous curve segment can better represent the approximate driving state of the vehicle under the same road condition environment, and the road condition similarity confidence of the continuous curve segment at the i-th moment is higher.

[0054] (2) Based on the difference in driving speed between the current moment and other moments and the confidence level of the road condition similarity of the continuous curve segment at the current moment, the driving state change index at the current moment is obtained.

[0055] It should be noted that, since the driving states of vehicles under the same road conditions are basically similar and will not change over time, the operation of the vehicle under the same road conditions is similar, and the corresponding continuous curve segments are similar; the higher the similarity between the continuous curve segment at the current i-th moment and the continuous curve segment under the same road conditions, the higher the similarity between the i-th moment and the continuous curve segment under the same road conditions. The more similar the driving state of the car at the moment is to the driving state under the same road condition, the higher the possibility that the car is normal. The more different the continuous curve segment at the i-th moment is from the driving state under the same road condition, the higher the possibility that the car is normal. The higher the probability that the car at the i-th moment has a fault; that is, the greater the difference between the driving speed at the i-th moment and the driving speed at other moments in the same cluster, the more dissimilar the driving state is, and the more likely the car is abnormal.

[0056] Specifically, the dissimilarity between the driving state at the i-th moment and the driving state at other moments in the same cluster is recorded as the driving state change index at the i-th moment; then the driving state change index at the i-th moment is calculated as follows:

[0057]

[0058] In the formula, represents the driving state change index at the i-th moment; Represents the confidence level of the road condition similarity of the continuous curve segment at the i-th moment; Indicates Driving speed at the moment; Indicates the cluster of the continuous curve segment at the i-th moment except the data point at the i-th moment. The driving speed of each data point; Represents the number of all data points in the cluster where the continuous curve segment at the i-th moment is located; Represents the Sigmoid function.

[0059] Under the same road conditions, The vehicle speed at the time The difference in vehicle speed at the time The larger the The greater the difference between the continuous curve segment at the time and the corresponding continuous curve segment under the same road condition, the The greater the possibility that the car has abnormalities.

[0060] (3) Based on the similarity of the driving speed changes between different clusters at adjacent moments, the possibility of abnormal vibration at the current moment is obtained.

[0061] It should be noted that the speed of the car changes greatly during the acceleration or deceleration stage. When the car is in a normal state, the driving speed changes at consecutive moments under the same road conditions. Since the driving states under the same road conditions are similar, the changes in driving speed during the acceleration and deceleration stages are also similar, so the changes between different adjacent clusters at consecutive moments are similar. The greater the difference in driving speed changes between the cluster at the i-th moment and the adjacent cluster, the less similar the driving speeds at consecutive moments are, and the greater the difference in driving states under the same road conditions.

[0062] The previous continuous curve segment closest to the collection time of the continuous curve segment at the i-th moment is recorded as the nearest continuous curve segment at the i-th moment. Since the tire pressure and driving speed in a cluster are connected to form a continuous curve, the curve segment before the collection time of the continuous curve segment at the i-th moment must not be in the same cluster as the i-th moment. The cluster where the nearest continuous curve segment at the i-th moment is located is recorded as the nearest cluster at the i-th moment, and the last moment of the nearest cluster is recorded as the m-th moment in the nearest cluster.

[0063] Specifically, the calculation method of the abnormal vibration possibility at the i-th moment is:

[0064]

[0065] In the formula, represents the possibility of abnormal vibration at the i-th moment; represents the driving state change index at the i-th moment; Indicates Driving speed at the moment; represents the driving speed at the mth moment in the nearest cluster at the i-th moment. The mean of the driving speed of all data points in the cluster at the i-th time except the i-th time, represents the mean of the driving speeds of all data points in the nearest cluster at time i except the driving speed at time m; represents the Sigmoid function, Represents the absolute value function.

[0066] Indicated in The speed of time to The change in speed at a moment, It represents the change in the mean of the driving speed of the data points in the cluster at the i-th time except the driving speed at the i-th time, and the mean of the driving speed of the data points in the nearest cluster at the i-th time except the driving speed at the m-th time. So when The larger the The greater the difference between the driving speed change at the moment and the driving speed change in the near future, the more different it is from the historical data. The higher the probability that there is an abnormality in the car at that moment.

[0067] Abnormality detection module 103: obtains the driving state abnormality coefficient according to the correlation between the vibration intensity of the vehicle and the possibility of abnormal vibration, and judges the driving state of the vehicle.

[0068] The possibility of abnormal vibration at each moment in the acquisition time is calculated to form an abnormal vibration possibility sequence, and the Pearson correlation coefficient of the abnormal vibration possibility sequence and the sequence formed by the automobile vibration intensity at each moment in the acquisition time is calculated, and the calculated Pearson correlation coefficient is recorded as the driving state abnormality coefficient. A preset judgment threshold value is set. The judgment threshold value preset in this embodiment is 0.8, which is described as an example. Other implementation methods can be set to other values, which are not limited in this embodiment; if the driving state abnormality coefficient is greater than the preset judgment threshold value, the current automobile vibration has a strong correlation with the possibility of abnormal vibration. The automobile vibration at this time may be caused by the automobile driving process. If the automobile vibration intensity collected by the vehicle at this time is greater than the preset vibration threshold value, the automobile electronic control system does not issue an early warning prompt; if the driving state abnormality coefficient between the automobile vibration intensity and the possibility of abnormal vibration is less than or equal to the preset judgment threshold value, the current automobile vibration has a weak correlation with the possibility of abnormal vibration. The automobile vibration at this time may be caused by an abnormal fault in the automobile. If the automobile vibration intensity collected by the vehicle at this time is greater than the preset vibration threshold value, the automobile electronic control system issues an early warning prompt.

[0069] So far, the present invention is completed.

[0070] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A state perception system for multi-source data fusion of an automotive electronic control unit, characterized in that: The system includes the following modules: Data acquisition module: used to collect the vehicle's driving speed, vehicle vibration intensity and tire pressure data at each moment; Data processing module: used to cluster the driving speed and tire pressure data to obtain several clusters and several continuous curve segments in each cluster; obtain the road condition similarity confidence of the continuous curve segment at the current moment according to the difference in the length and number of the continuous curve segments in the cluster at the current moment; obtain the driving state change index at the current moment according to the difference in the driving speed at the current moment and the driving speed at other moments in the cluster; obtain the possibility of abnormal vibration at the current moment according to the difference in the driving speed on the previous continuous curve segment of the continuous curve segment at the current moment; Abnormal detection module: used to obtain the driving state abnormality coefficient according to the correlation between the vehicle vibration intensity and the possibility of abnormal vibration; and to issue a vehicle driving vibration warning according to the driving state abnormality coefficient; The clustering of the driving speed and tire pressure data to obtain a plurality of clusters and a plurality of continuous curve segments in each cluster includes: The clustering space is constructed with the driving speed and tire pressure as the coordinate axes respectively, and all the driving speed and tire pressure data are mapped into the clustering space to form a number of discrete data points. The DBSCAN density clustering algorithm is used to cluster all the discrete data points according to the preset clustering radius and the preset number of samples to obtain a number of clusters; Connect the continuous data points in each cluster at the time of collection to obtain several continuous curve segments in each cluster; The obtaining of the road condition similarity confidence of the continuous curve segment at the current moment includes: The current moment is recorded as the i-th moment, and the road condition similarity confidence of the continuous curve segment at the i-th moment is The calculation method is: In the formula, represents the number of continuous curve segments in the velocity cluster set at the i-th moment; represents the number of special continuous curve segments in the velocity cluster set at the i-th moment; The specific method of obtaining the speed cluster set includes: Obtain the tire pressure range of all data points in each cluster, and record the set of all clusters that have an intersection with the tire pressure range of the cluster where the data point at the current moment is located as the speed cluster set at the current moment; The specific method for obtaining the number of special continuous curve segments includes: In the speed cluster set at the i-th moment, the number of continuous curve segments whose length difference with the continuous curve segment at the i-th moment is less than the preset length threshold is recorded as the number of special continuous curve segments in the speed cluster set at the i-th moment; The driving state change index at the current moment is obtained according to the difference between the driving speed at the current moment and the driving speed at other moments in the cluster, including: The current moment is recorded as the i-th moment, and the calculation method of the driving state change index at the i-th moment is: In the formula, represents the driving state change index at the i-th moment; Represents the road condition similarity confidence of the continuous curve segment at the i-th moment; Indicates Driving speed at the moment; Indicates the cluster of the continuous curve segment at the i-th moment except the data point at the i-th moment. The driving speed of each data point; Represents the number of all data points in the cluster where the continuous curve segment at the i-th moment is located; Represents the Sigmoid function; The method of obtaining the possibility of abnormal vibration at the current moment according to the difference in driving speed between the continuous curve segment before the continuous curve segment at the current moment comprises: Record the current moment as the i-th moment, record the last moment of the nearest cluster of the i-th moment as the m-th moment in the nearest cluster of the i-th moment, and the probability of abnormal vibration at the i-th moment The calculation method is: In the formula, represents the driving state change index at the i-th moment; Indicates Driving speed at the moment; represents the driving speed at the mth moment in the nearest cluster at the i-th moment; represents the mean value of the driving speed of other data points in the cluster at the i-th time except the i-th time. represents the mean of the driving speeds of all data points in the nearest cluster at time i except the driving speed at time m; represents the Sigmoid function, Represents the absolute value function.

2. According to claim 1, a state perception system for multi-source data fusion of an automotive electronic control unit is characterized in that: The specific method of obtaining the nearest neighbor cluster includes: The previous continuous curve segment closest to the collection time of the continuous curve segment at the i-th moment is recorded as the nearest continuous curve segment at the i-th moment, and the cluster where the nearest continuous curve segment at the i-th moment is located is recorded as the nearest cluster at the i-th moment.

3. According to claim 1, a state perception system for multi-source data fusion of an automotive electronic control unit is characterized in that: The abnormal coefficient of driving state is obtained according to the correlation between the vibration intensity of the vehicle and the possibility of abnormal vibration, including: The abnormal vibration possibility at each moment is calculated to form an abnormal vibration possibility sequence, the Pearson correlation coefficient between the abnormal vibration possibility sequence and the sequence formed by the vehicle vibration intensity at each moment is calculated, and the Pearson correlation coefficient is recorded as the driving state abnormality coefficient.

4. According to claim 1, a state perception system for multi-source data fusion of an automotive electronic control unit is characterized in that: The vehicle driving vibration warning according to the driving state abnormality coefficient includes: A preset judgment threshold is set. When the driving state abnormality coefficient is less than or equal to the preset judgment threshold, and the vehicle vibration intensity of the vehicle at the current moment is greater than the preset vibration threshold, the vehicle electronic control system issues a warning prompt.

Images