Vehicle motion state judgment method and system

Abstract

The invention relates to a vehicle motion state judgment method and system, and the method comprises the following steps: obtaining original IMU data corresponding to a current frame image, and obtaining IMU average values in one-to-one correspondence with the images according to the original IMU data subjected to data processing; performing averaging operation on the IMU average value subjected to data processing, and completing IMU calibration based on the obtained IMU calibration data average value; and obtaining current IMU data corresponding to the current frame image through the calibrated IMU, and judging the motion state of the vehicle according to the current IMU data. According to the method, the IMU is calibrated through the multi-group and multi-dimensional original IMU data, the IMU data is further acquired in real time through the calibrated IMU, the vehicle motion state is judged quickly, the calculation process is simple, and the calculation amount is small.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a vehicle motion state judgment method and system. Background Art

[0002] In the prior art, when a vehicle-mounted camera executes a corresponding algorithm or control strategy, it needs to use an on-board IMU to obtain IMU data. However, due to installation differences, noise levels, and other reasons, the IMU data has a large error. As a result, there are technical solutions for correcting the IMU data to complete the calibration. However, these calibration processes require the collection of a large number of images, and the calculations are complex and difficult to implement. At the same time, the prior art does not link the IMU data with the judgment of the vehicle's motion state. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for judging the motion state of a vehicle. The method first calibrates the IMU through multiple groups of multi-dimensional original IMU data, and further obtains the IMU data in real time through the calibrated IMU, so as to quickly complete the judgment of the motion state of the vehicle. The calculation process is simple and the calculation amount is small.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] On the one hand, a method for determining a vehicle motion state is provided, which comprises the following steps:

[0006] Get the original IMU data corresponding to the current frame image, and obtain the IMU average value IMU corresponding to the image one by one based on the original IMU data after data processing sortcut_single ;

[0007] IMU average value IMU after data processing sortcut_single Perform averaging operation based on the average IMU calibration data obtained sortcut_single_mean , to complete the IMU calibration;

[0008] The current IMU data corresponding to the current frame image is obtained through the calibrated IMU, and the vehicle motion state is determined based on the current IMU data.

[0009] Preferably, each frame of image corresponds to a number k groups of original IMU data, and each group of original IMU data includes: x-axis angular velocity x_gyro, y-axis angular velocity y_gyro, z-axis angular velocity z_gyro and z-direction acceleration z_acc.

[0010] Preferably, the IMU calibration data average value IMU sortcut_single_mean Includes: x-axis angular velocity calibration data average value Average value of y-axis angular velocity calibration data Average value of z-axis angular velocity calibration data And the average value of the z-axis acceleration calibration data

[0011] Preferably, judging the vehicle motion state according to the current IMU data comprises the following steps:

[0012] Smooth the current IMU data to obtain the smoothed IMU data IMU KF ;

[0013] The smoothed IMU data IMU KF Bias of the upper limit of the vehicle motion state judgment data interval U And the lower limit Bias D A comparison is performed, and the vehicle motion state in the current frame image is determined based on the comparison result.

[0014] Preferably, the current IMU data is smoothed to obtain smoothed IMU data IMU KF , including the following steps:

[0015] According to formula (1), the IMU data change Pre is obtained diff :

[0016] Pre diff =(IMU shortcut_single -IMU hist ) / twenty one)

[0017] Among them, IMU short_single is the IMU average value corresponding to the current frame image; IMU hist is the IMU average value corresponding to the previous frame image;

[0018] According to formula (2), the IMU data change coefficient K is obtained:

[0019]

[0020] Among them, Pre var =Pre diff 2 , Kalman is the smoothing coefficient of Kalman filter;

[0021] According to formula (3), the smoothed IMU data IMU is obtained. KF :

[0022] IMU KF =IMU hist +K*(Pre diff ) (3).

[0023] Preferably, the smoothed IMU data IMU KF Bias of the upper limit of the vehicle motion state judgment data interval U And the lower limit Bias D The comparison is performed, and the vehicle motion state in the current frame image is determined according to the comparison result, including the following steps:

[0024] If IMU KF ∈[Bias D , Bias U ], the vehicle in the current frame image is considered to be stationary.

[0025] If IMU KF <Bias D or IMU KF >Bias U , then it is considered that the vehicle in the current frame image is in a moving state;

[0026] When the proportion of images in which the vehicle is in a moving state among several frames is greater than or equal to a threshold and they are continuous frame images, the vehicle is considered to be in motion; if it is less than the threshold, the vehicle is considered to be stationary.

[0027] Preferably, the upper limit Bias of the vehicle motion state judgment data interval U And the lower limit Bias D Obtained through the following formula:

[0028]

[0029] Among them, γ is an adjustable parameter; Bias is the IMU data difference sequence, and Bias_x_gyro, Bias_y_gyro, Bias_z_gyro, and Bias_z_acc are the IMU calibration data sequences. calib , IMU calibration data series average IMU sortcut_single_mean The average angular velocity of each x-axis is Average value of x-axis angular velocity calibration data The average value of the difference, the average value of each y-axis angular velocity Average value of y-axis angular velocity calibration data The average value of the difference, the average value of each z-axis angular velocity Average value of z-axis angular velocity calibration data The average value of the difference and the average value of the z-direction acceleration Average value of z-axis acceleration calibration data The average of the differences.

[0030] Preferably, after the vehicle is considered to be in motion, if the IMU KF <Bias D , then the vehicle motion deviation result ME cur =IMU KF -Bias D , in several frames of images, the vehicle motion deviation results ME cur If the proportion of images less than or equal to the first turning threshold is greater than or equal to the preset value, the vehicle is considered to be in a left turn state; if the IMU KF >Bias U , then the vehicle motion deviation result ME cur =IMU KF -Bias U , and in several frames of images, the vehicle motion deviation result ME cur When the proportion of the number of images greater than or equal to the second turning threshold is greater than or equal to a preset value, it is considered that the vehicle is in a right turn state; if the vehicle is not in a turning state, it is considered that the vehicle is in a straight state.

[0031] On the other hand, a vehicle motion state determination system is also provided, which includes:

[0032] An IMU unit, which is used to obtain raw IMU data corresponding to the current frame image in real time;

[0033] The raw data processing unit is used to process the raw IMU data to obtain the IMU data after data processing. sortcut ;

[0034] The raw data mean acquisition unit is used to obtain the IMU data after data processing. sortcut Perform an averaging operation to obtain the IMU average value IMU corresponding to the current frame image sortcut_single ;

[0035] Calibration data acquisition unit, which is used to calibrate the IMU containing the average value of all IMUs sortcut_single IMU calibration data sequence IMU calib Perform data processing to obtain the IMU calibration data sequence IMU after data processing calib_sortcut ;

[0036] The calibration data mean acquisition unit is used to calibrate the IMU data sequence after data processing. calib_sortcut Perform an averaging operation to obtain the average value of the IMU calibration data IMU sortcut_single_mean ;

[0037] The first motion state judgment unit is used to judge whether the vehicle is in a motion state according to the current IMU data.

[0038] Preferably, the vehicle motion state determination system further includes:

[0039] The second motion state judgment unit is used to determine whether the vehicle is in a turning state or a straight state according to the vehicle operation deviation result after the vehicle is considered to be in a motion state.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present application first calibrates the IMU through multiple groups of multi-dimensional original IMU data, and further obtains IMU data in real time through the calibrated IMU, and thereby quickly completes the judgment of the vehicle's motion state. The calculation process is simple and the amount of calculation is small. Furthermore, on the premise that the vehicle has been judged to be in a motion state, it can further judge the specific motion state of the vehicle, such as turning left, turning right, going straight, etc., which is conducive to achieving precise motion control of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of the steps of the vehicle motion state determination method in the present invention;

[0043] Figure 2 It is a structural schematic diagram of the vehicle motion state judgment system in the present invention. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] Example 1

[0046] like Figure 1 As shown, this embodiment provides a method for determining the motion state of a vehicle, which includes the following steps:

[0047] S1. Obtain the original IMU (i.e., Inertial Measurement Unit) data corresponding to the current frame image. In this embodiment, each frame image corresponds to a number of k groups of original IMU data, and each group of original IMU data includes: x-axis angular velocity x_gyro, y-axis angular velocity y_gyro, z-axis angular velocity z_gyro, and z-axis acceleration z_acc. For example, when k=15, the original IMU data corresponding to the current frame image is as follows:

[0048]

[0049] S2. Calibrate the IMU according to the original IMU data; in this embodiment, the IMU can be installed on the vehicle;

[0050] Specifically, the IMU is calibrated according to the original IMU data, including the following steps:

[0051] S21, sorting the x-axis angular velocity x_gyro, the y-axis angular velocity y_gyro, the z-axis angular velocity z_gyro, and the z-direction acceleration z_acc in the original IMU data corresponding to the current frame image from small to large / from large to small;

[0052] S22, processing the x-axis angular velocity x_gyro, y-axis angular velocity y_gyro, z-axis angular velocity z_gyro and z-direction acceleration z_acc of the sorted original IMU data, thereby obtaining the IMU data IMU after data processing sortcut , wherein the data processing includes data elimination, that is, eliminating the first m data and the last n data in the order of data value from large to small, wherein m and n can be the same or different. For example, in this embodiment, if m=n=3, k=15, then according to the order of data value from large to small, the 1st to 3rd and 13th to 15th data are eliminated, and the IMU data after data elimination is IMU sortcut as follows:

[0053]

[0054] And the IMU data after data processing sortcut The x-axis angular velocity x_gyro, y-axis angular velocity y_gyro, z-axis angular velocity z_gyro and z-axis acceleration z_acc in the image are all averaged to obtain the IMU average value IMU corresponding to the current frame image sortcut_single :

[0055]

[0056] in, They correspond to the average angular velocity of the x-axis, the average angular velocity of the y-axis, the average angular velocity of the z-axis, and the average acceleration in the z-direction;

[0057] S23, repeat steps S21-S22 to obtain the IMU average value IMU corresponding to the N frames of images one by one sortcut_single , and construct an IMU containing the average value of all IMUs sortcut_single IMU calibration data sequence IMU calib :

[0058]

[0059] In this embodiment, the value range of N is 50-200, preferably N=100;

[0060] S24, IMU calibration data sequence IMU calib The average x-axis angular velocity in Average y-axis angular velocity Average z-axis angular velocity And the average z-axis acceleration All are sorted from small to large / from large to small;

[0061] And the sorted IMU calibration data sequence IMU calib The average x-axis angular velocity in Average y-axis angular velocity Average z-axis angular velocity And the average z-axis acceleration Data processing is performed on all of them, thereby obtaining the IMU calibration data sequence IMU after data processing calib_sortcut Similarly, the data processing includes data elimination, that is, according to the order of data values ​​from large to small, the first p data are eliminated, and the last q data are eliminated, where p and q can be the same or different. For example, in this embodiment, if p=q=20, N=100, then according to the order of data values ​​from large to small, the data of the 1st to 20th and the 81st to 100th bits are eliminated. After the data is eliminated, the IMU calibration data sequence IMU calib_sortcut as follows:

[0062]

[0063] S25, IMU calibration data sequence IMU after data processing calib_sortcut The average x-axis angular velocity in Average y-axis angular velocity Average z-axis angular velocity And the average z-axis acceleration The average value is calculated to obtain the average value of IMU calibration data. sortcut_single_mean :

[0064]

[0065] in, They correspond to the average value of the x-axis angular velocity calibration data, the average value of the y-axis angular velocity calibration data, the average value of the z-axis angular velocity calibration data, and the average value of the z-axis acceleration calibration data;

[0066] This completes the IMU calibration. Since the noise level and installation position of each IMU are different, there are various calculation errors. In order to improve the accuracy and precision of the subsequent vehicle status judgment, the IMU needs to be calibrated first;

[0067] S3, obtaining the current IMU data corresponding to the current frame image in real time through the calibrated IMU, and judging the vehicle motion state according to the current IMU data, which specifically includes the following steps:

[0068] S31, smoothing the current IMU data corresponding to the current frame image to obtain smoothed IMU data, which specifically includes the following steps:

[0069] According to formula (1), the IMU data change Pre is obtained diff :

[0070] Pre diff =(IMU shortcut_single -IMU hist ) / twenty one)

[0071] Among them, IMU short_single is the IMU average value corresponding to the current frame image; IMU hist is the IMU average value corresponding to the previous frame image;

[0072] According to formula (2), the IMU data change coefficient K is obtained:

[0073]

[0074] Among them, Pre var =Pre diff 2 , Kalman is the smoothing coefficient of Kalman filter;

[0075] According to formula (3), the smoothed IMU data is obtained:

[0076] IMU KF =IMU hist +K*(Pre diff ) (3)

[0077] Among them, IMU KF is the IMU data corresponding to the current frame image after smoothing. Due to the existence of IMU noise, the IMU data corresponding to each frame image may have jumps, so it is necessary to filter the jump points first to make the IMU data smooth.

[0078] S32, the IMU data corresponding to the current frame image and after smoothing KFBias of the upper limit of the vehicle motion state judgment data interval U And the lower limit Bias D Performing comparison, and determining the vehicle motion state in the current frame image according to the comparison result;

[0079] Specifically, if IMU KF ∈[Bias D , Bias U ], the vehicle in the current frame image is considered to be stationary.

[0080] If IMU KF <Bias D or IMU KF >Bias U , then it is considered that the vehicle in the current frame image is in a moving state;

[0081] S33. Repeat steps S31-S32. When the proportion of images in which the vehicle is in a moving state in a number of frames is greater than or equal to a threshold and is a continuous frame image, the vehicle is considered to be in motion. If the proportion is less than the threshold, the vehicle is considered to be stationary. For example, in this embodiment, the threshold value ranges from 60% to 95%.

[0082] Furthermore, the upper limit Bias of the vehicle motion state judgment data interval U And the lower limit Bias D Obtain it by following the steps below:

[0083] IMU based on the average IMU calibration data sortcut_single_mean The average value of the x-axis angular velocity calibration data in Average value of y-axis angular velocity calibration data Average value of z-axis angular velocity calibration data And the average value of the z-axis acceleration calibration data Construct the IMU data difference sequence Bias, and

[0084] Among them, Bias_x_gyro, Bias_y_gyro, Bias_z_gyro, and Bias_z_acc are the IMU calibration data sequences IMU calib , IMU calibration data series average IMU sortcut_single_mean The average angular velocity of each x-axis is Average value of x-axis angular velocity calibration data The average value of the difference, the average value of each y-axis angular velocity Average value of y-axis angular velocity calibration data The average value of the difference, the average value of each z-axis angular velocity Average value of z-axis angular velocity calibration data The average value of the difference and the average value of the z-direction acceleration Average value of z-axis acceleration calibration data The average of the differences;

[0085] And Bias_x_gyro, Bias_y_gyro, Bias_z_gyro, and Bias_z_acc are obtained by formulas (4-1)-(4-4) respectively:

[0086]

[0087] Among them, mean means taking the average value;

[0088] Determine the upper limit Bias of the vehicle motion state judgment data interval based on the IMU data difference sequence Bias U And the lower limit Bias D , where the upper limit Bias U And the lower limit Bias D Obtained by formulas (5-1)-(5-2) respectively:

[0089]

[0090] Wherein, γ is an adjustable parameter, and its value range is 0.5-10. For example, in this embodiment, γ=4;

[0091] Furthermore, the smoothing coefficient Kalman of the Kalman filter can be obtained according to the IMU data difference sequence Bias, wherein the smoothing coefficient Kalman=Bias 2 .

[0092] Therefore, this embodiment first obtains multiple groups of original IMU data corresponding to each frame of image, and each original IMU data contains 4-dimensional information (i.e., x angular velocity, y angular velocity, z angular velocity, and z acceleration). The original IMU data is further processed (such as data elimination, etc.), averaged, and other operations are performed to complete the IMU calibration, thereby reducing the result error when the vehicle motion state is subsequently judged by the IMU data. The calibrated IMU is further used to obtain IMU data in real time, and the vehicle motion state is quickly judged. The calculation process is simple, the calculation amount is small, and the judgment result is highly accurate.

[0093] Embodiment 2:

[0094] The difference between this embodiment and the first embodiment is that in step S3, after the vehicle is considered to be in motion, it is determined whether the vehicle is in a turning state or in a straight state according to the vehicle operation deviation result;

[0095] Specifically, if IMU KF <Bias D , then the vehicle motion deviation result ME cur =IMU KF -Bias D , and in several frames of images, the vehicle motion deviation result ME cur When the proportion of images less than or equal to the first turning threshold is greater than or equal to a preset value, it is considered that the vehicle is in a left turn state;

[0096] If IMU KF >Bias U , then the vehicle motion deviation result ME cur =IMU KF -Bias U , and in several frames of images, the vehicle motion deviation result ME cur When the proportion of images with a value greater than or equal to the second turning threshold is greater than or equal to a preset value, it is considered that the vehicle is in a right turn state;

[0097] The absolute values ​​of the first turning threshold and the second turning threshold are the same, and one of them is a positive value and the other is a negative value;

[0098] In addition, if the vehicle is not in a turning state (ie, the vehicle is neither in a left turn state nor in a right turn state), it is considered that the vehicle is in a straight-moving state.

[0099] Therefore, this embodiment can further determine the specific motion state of the vehicle, such as turning left, turning right, going straight, etc., based on the premise that the vehicle has been determined to be in a motion state, which is conducive to achieving precise motion control of the vehicle.

[0100] Embodiment 3:

[0101] This embodiment provides a vehicle motion state determination system that can implement the vehicle motion state determination method described in Embodiment 1 or 2. Figure 2 As shown, it includes:

[0102] IMU unit 1, which is used to obtain the original IMU data corresponding to the current frame image in real time;

[0103] The raw data processing unit 2 is used to process the raw IMU data to obtain the IMU data after data processing. sortcut , the process is the same as step S22;

[0104] The original data mean acquisition unit 3 is used to obtain the IMU data after data processing. sortcut Perform an averaging operation to obtain the IMU average value IMU corresponding to the current frame imagesortcut_single , the process is the same as step S22;

[0105] The calibration data acquisition unit 4 is used to calibrate the IMU including the average value of all IMUs sortcut_single IMU calibration data sequence IMU calib Perform data processing to obtain the IMU calibration data sequence IMU after data processing calib sortcut , the process is the same as steps S23-S24;

[0106] The calibration data mean acquisition unit 5 is used to calibrate the IMU data sequence after data processing. calib_sortcut Perform an averaging operation to obtain the average value of the IMU calibration data IMU sortcut_single_mean , the process is the same as step S25;

[0107] A first motion state judgment unit 6, which is used to judge whether the vehicle is in motion according to the current IMU data, and the process is the same as step S3;

[0108] The second motion state judgment unit 7 is used to determine whether the vehicle is in a turning state or a straight-moving state according to the vehicle operation deviation result after the vehicle is considered to be in a motion state. The process is the same as that of embodiment 2.

[0109] To summarize, the present application first calibrates the IMU through multiple groups of multi-dimensional original IMU data to reduce the result error when subsequently judging the vehicle's motion state through the IMU data, and further obtains the IMU data in real time through the calibrated IMU, and thereby quickly completes the judgment of the vehicle's motion state. The calculation process is simple and the amount of calculation is small. Furthermore, on the premise that the vehicle has been judged to be in a motion state, it can further judge what specific motion state the vehicle is in, such as turning left, turning right, going straight, etc., which is conducive to achieving precise motion control of the vehicle.

[0110] 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 spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining a vehicle motion state, characterized in that: The steps include: Get the original IMU data corresponding to the current frame image, and obtain the IMU average value IMU corresponding to the image one by one based on the original IMU data after data processing sortcut_single ; IMU average value IMU after data processing sortcut_single Perform averaging operation based on the average IMU calibration data obtained sortcut_single_mean , to complete the IMU calibration; The current IMU data corresponding to the current frame image is obtained through the calibrated IMU, and the vehicle motion state is determined based on the current IMU data.

2. The vehicle motion state determination method according to claim 1, characterized in that: Each frame of image corresponds to several k groups of raw IMU data, and each group of raw IMU data includes: x-axis angular velocity x_gyro, y-axis angular velocity y_gyro, z-axis angular velocity z_gyro and z-axis acceleration z_acc.

3. The vehicle motion state determination method according to claim 1, characterized in that: The IMU calibration data average IMU sortcut_single_mean Includes: x-axis angular velocity calibration data average value Average value of y-axis angular velocity calibration data Average value of z-axis angular velocity calibration data And the average value of the z-axis acceleration calibration data 4. The vehicle motion state determination method according to claim 3, characterized in that: Judging the vehicle motion state according to the current IMU data includes the following steps: Smooth the current IMU data to obtain the smoothed IMU data IMU KF ; The smoothed IMU data IMU KF Bias of the upper limit of the vehicle motion state judgment data interval U And the lower limit Bias D A comparison is performed, and the vehicle motion state in the current frame image is determined based on the comparison result.

5. The vehicle motion state determination method according to claim 4, characterized in that: Smooth the current IMU data to obtain the smoothed IMU data IMU KF , including the following steps: According to formula (1), the IMU data change Pre is obtained diff : Pre diff =(IMU shortcut_single -IMU hist ) / 2 (1) Among them, IMU short_single is the IMU average value corresponding to the current frame image; IMU hist is the IMU average value corresponding to the previous frame image; According to formula (2), the IMU data change coefficient K is obtained: Among them, Pre var =Pre diff 2 , Kalman is the smoothing coefficient of Kalman filter; According to formula (3), the smoothed IMU data IMU is obtained. KF : IMU KF =IMU hist +K*(For diff ) (3).

6. The vehicle motion state determination method according to claim 4, characterized in that: The smoothed IMU data IMU KF Bias of the upper limit of the vehicle motion state judgment data interval U And the lower limit Bias D The comparison is performed, and the vehicle motion state in the current frame image is determined according to the comparison result, including the following steps: If IMU KF ∈[Bias D , Bias U ], the vehicle in the current frame image is considered to be stationary. If IMU KF <Bias D or IMU KF >Bias U , then it is considered that the vehicle in the current frame image is in a moving state; When the proportion of images in which the vehicle is in a moving state among several frames is greater than or equal to a threshold and they are continuous frame images, the vehicle is considered to be in motion; if it is less than the threshold, the vehicle is considered to be stationary.

7. The vehicle motion state determination method according to claim 6, characterized in that: The upper limit Bias of the vehicle motion state judgment data interval U And the lower limit Bias D Obtained through the following formula: Among them, γ is an adjustable parameter; Bias is the IMU data difference sequence, and Bias_x_gyro, Bias_y_gyro, Bias_z_gyro, and Bias_z_acc are the IMU calibration data sequences. calib , IMU calibration data series average IMU sortcut_single_mean The average angular velocity of each x-axis is Average value of x-axis angular velocity calibration data The average value of the difference, the average value of each y-axis angular velocity Average value of y-axis angular velocity calibration data The average value of the difference, the average value of each z-axis angular velocity Average value of z-axis angular velocity calibration data The average value of the difference and the average value of the z-direction acceleration Average value of z-axis acceleration calibration data The average of the differences.

8. The vehicle motion state determination method according to claim 6, characterized in that: After the vehicle is considered to be in motion, if the IMU KF <Bias D , then the vehicle motion deviation result ME cur =IMU KF -Bias D , in several frames of images, the vehicle motion deviation results ME cur If the proportion of images less than or equal to the first turning threshold is greater than or equal to the preset value, the vehicle is considered to be in a left turn state; if the IMU KF >Bias U , then the vehicle motion deviation result ME cur =IMU KF -Bias U , and in several frames of images, the vehicle motion deviation result ME cur When the proportion of the number of images greater than or equal to the second turning threshold is greater than or equal to a preset value, it is considered that the vehicle is in a right turn state; if the vehicle is not in a turning state, it is considered that the vehicle is in a straight state.

9. A vehicle motion state judgment system, characterized in that: include: An IMU unit, which is used to obtain raw IMU data corresponding to the current frame image in real time; The raw data processing unit is used to process the raw IMU data to obtain the IMU data after data processing. sortcut ; The raw data mean acquisition unit is used to obtain the IMU data after data processing. sortcut Perform an averaging operation to obtain the IMU average value IMU corresponding to the current frame image sortcut_single ; Calibration data acquisition unit, which is used to calibrate the IMU containing the average value of all IMUs sortcut_single IMU calibration data sequence IMU calib Perform data processing to obtain the IMU calibration data sequence IMU after data processing calib_sortcut ; The calibration data mean acquisition unit is used to calibrate the IMU data sequence after data processing. calib_sortcut Perform an averaging operation to obtain the average value of the IMU calibration data IMU sortcut_single_mean ; The first motion state judgment unit is used to judge whether the vehicle is in a motion state according to the current IMU data.

10. The vehicle motion state determination system according to claim 9, characterized in that: The vehicle motion state judgment system also includes: The second motion state judgment unit is used to determine whether the vehicle is in a turning state or a straight-moving state according to the vehicle operation deviation result after it is considered that the vehicle is in a motion state.

Images