Vehicle speed prediction method under special function activation condition
By measuring the initial wheel speed under special function activation and executing a loop to detect the peak vehicle speed, the peak vehicle speed is calculated using the wheel speed difference matrix and the matrix, thus solving the problem of vehicle speed fluctuation in Hall effect sensors under special driving conditions and achieving accurate vehicle speed prediction.
Patent Information
- Application Number
- CN202411029146.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2026-01-30
AI Technical Summary
Under special driving conditions, such as ABS operation, the wheel speed measured by the Hall effect sensor fluctuates rapidly between the maximum and minimum values, making it impossible to accurately calculate or predict the vehicle speed.
By activating special functions and measuring the initial wheel speed at the initial time point, a loop is executed to perform vehicle speed prediction steps and detect the peak vehicle speed. The peak vehicle speed is calculated using the wheel speed difference matrix, peak factor matrix, and correction factor matrix. The peak vehicle speed is compared with a predetermined threshold to determine the effective peak value. The time point and wheel speed are then updated for accurate prediction.
When the special function is activated, it can accurately predict vehicle speed, relying solely on wheel speed measurement without requiring other input, thus improving the accuracy of vehicle speed prediction.
Smart Images
Figure CN121425239A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle control technology, and more specifically to a method for predicting vehicle speed when a special function, such as a special braking function, is activated. For example, the special function could be something like ABS provided in a driver assistance system. Background Technology
[0002] To improve the active safety of vehicles while driving, it is necessary to measure the state parameters of the vehicle during driving, and vehicle speed is one of the important parameters of the vehicle active safety control system.
[0003] Measuring wheel speed using Hall effect sensors and calculating vehicle speed based on that speed is a common method for vehicle speed prediction. Specifically, these sensors generate pulse signals with a frequency proportional to the wheel speed during vehicle movement, and the vehicle control system uses this pulse frequency to calculate the wheel speed. However, under certain driving conditions, such as ABS operation, the vehicle's wheel speed characteristics change, and the wheel speed measured by the Hall effect sensor fluctuates rapidly between its maximum and minimum values. In such cases, the vehicle speed cannot be calculated or predicted using the wheel speed measured by the sensor as in normal driving conditions. Summary of the Invention
[0004] The purpose of this application is to provide a new vehicle speed prediction technology when a special function is activated.
[0005] The objective of this application is achieved through a method for predicting vehicle speed when a special function is activated. This method includes: activating the special function and measuring the initial wheel speed at an initial time point; then simultaneously performing a vehicle speed prediction step to predict the vehicle speed at a target time point and executing a loop to detect the peak vehicle speed.
[0006] The vehicle speed prediction step is performed as follows: before the effective peak value is detected cyclically, it is based on the initial wheel speed, a predetermined slope value depending on the vehicle model, and a target time point; after the effective peak value is detected cyclically, it is performed based on the peak time point and peak wheel speed corresponding to the current effective peak value, the peak time point and peak wheel speed corresponding to the previous effective peak value, or, when the current effective peak value is the first effective peak value, based on the initial time point and initial wheel speed, and based on the time interval between the target time point and the peak time point corresponding to the current effective peak value or the initial time point.
[0007] The loop includes: a first step of repeatedly measuring and recording wheel speeds and corresponding time points; a second step of selecting a set of consecutive time points and a corresponding set of wheel speeds and generating a wheel speed difference matrix; a third step of calculating a vehicle speed peak based on the wheel speed difference matrix, a predetermined peak factor matrix depending on the vehicle model, and a predetermined correction factor matrix depending on environmental factors; a fourth step of comparing the vehicle speed peak with a predetermined peak threshold depending on the vehicle model to determine whether the vehicle speed peak is a valid peak; and a fifth step: when the vehicle speed peak is determined to be a valid peak, determining the peak time point and peak wheel speed corresponding to the valid peak and providing them to the vehicle speed prediction step, and repeating steps two through five when the vehicle speed peak is determined not to be a valid peak, wherein the set of time points selected in the second step is the same number of consecutive time points starting from the second time point in the previous set of time points.
[0008] In the wheel speed difference matrix, the element in the i-th row and j-th column represents the difference between the i-th wheel speed and the j-th wheel speed in the set of wheel speeds. Here, i and j are natural numbers between 1 and the number of time points in the set of consecutive time points, including endpoints.
[0009] In one embodiment, the slope value, the peak factor matrix and the correction factor matrix, the preset threshold, and the number of time points included in the set of time points are obtained or calibrated through experimentation or empirical means.
[0010] In one embodiment, when the special function is activated, the initial time point and the initial wheel speed are stored as a first time point and a first wheel speed;
[0011] In each loop, after determining that the peak vehicle speed is a valid peak, the peak time point and peak wheel speed corresponding to the valid peak are stored as the second time point and the second wheel speed and sent to the vehicle speed prediction step. Before re-executing the loop, the first time point and the first wheel speed are updated using the second time point and the second wheel speed.
[0012] In one embodiment, the loop is returned to its starting step immediately after the update or after a predetermined period of time, and the loop is re-executed.
[0013] In one embodiment, the set of time points and the set of wheel speeds in the loop are respectively denoted as t1, t2…t n-1 and v1, v2...v n-1 Indicated, n is the number of time points; the peak factor matrix is a 1xn matrix; the correction factor matrix is an nx1 matrix;
[0014] The calculation of peak vehicle speed includes multiplying the wheel speed difference matrix by the peak factor matrix and the correction factor matrix to obtain the peak vehicle speed.
[0015] In one embodiment, the correction factor matrix is a constant matrix based on wheel noise, where all elements in the correction factor matrix are 1 when the wheel noise is less than a noise threshold.
[0016] In one embodiment, when it is determined that the peak vehicle speed is a valid peak, one of the time points is selected from the set of time points as the peak time point and the corresponding wheel speed is selected as the peak wheel speed.
[0017] In one embodiment, the peak time point is the last time point in the set of time points, or the middle time point in the set of time points, or a time point adjacent to the middle time point.
[0018] This application provides a computer program product, including a computer-readable program that, when a computer or processor is run, causes the computer or processor to perform the methods described above.
[0019] The vehicle speed prediction method of this application starts at the initial time point when a special function is activated, and simultaneously performs repeated operations to predict vehicle speed and performs a cyclic operation to detect vehicle speed peaks. During a first time period from the initial time point until the first valid (vehicle speed) peak is detected in the cyclic detection of vehicle speed peaks, vehicle speed prediction is performed according to a first linear change law of vehicle speed with a specific slope obtained or calibrated based on a specific vehicle model through experience or experimentation, based on the time interval between the target time point and the initial time point. After a valid peak is detected in the cyclic detection of vehicle speed peaks, vehicle speed prediction is performed according to a second linear change law of vehicle speed defined by the peak time point and peak wheel speed corresponding to the valid peak, and by the peak time point and peak wheel speed corresponding to the previous valid peak, based on the time interval between the target time point and the peak time point corresponding to the valid peak. When the valid peak is the first valid peak, the peak time point and peak wheel speed corresponding to the previous valid peak are the initial time point and initial wheel speed. The specific steps of detecting peak vehicle speed include: selecting a set of consecutive time points and a corresponding set of consecutive wheel speeds to generate a wheel speed difference matrix; multiplying the wheel speed difference matrix with a peak factor matrix obtained or calibrated based on a specific vehicle model through experience or experimentation, and a correction factor matrix obtained or calibrated based on environmental factors through experience or experimentation to calculate the peak vehicle speed; comparing the calculated peak vehicle speed with a predetermined threshold obtained or calibrated based on a specific vehicle model through experience or experimentation; determining peak vehicle speeds greater than or equal to the threshold as valid peak speeds; and selecting a time point from the set of time points used to calculate the peak vehicle speed, providing this time point and the corresponding wheel speed as the peak time point and peak wheel speed to the above-mentioned vehicle speed prediction operation.
[0020] The vehicle speed prediction technology or method of this application requires no other input besides the measured wheel speed, and can obtain a relatively accurate vehicle speed prediction value when a special function is activated. Attached Figure Description
[0021] The foregoing advantages and features of this application, as well as other advantages and features, will be better understood in the following detailed description with reference to the accompanying drawings.
[0022] Figure 1 This is a flowchart of the vehicle speed prediction method according to this application.
[0023] Figures 2a-2c The curve is used to illustrate the principles of this application. Detailed Implementation
[0024] To make the purpose, technical solution, and advantages of this application clearer and easier to understand, the following will be combined with Figure 1 Flowcharts and Figures 2a-2cThe curves provided further illustrate the technology of this application in detail. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit the scope of this application.
[0025] The technology in this application relates to predicting vehicle speed when a special function of the vehicle (such as a special braking function) is activated. The "special function" in this application may be an anti-lock braking system (ABS), or a special braking function that is already included in, or will be updated or added in the future, for example, in an advanced driver assistance system for the vehicle.
[0026] Specifically, this application relates to predicting the vehicle speed V at a time point T after a special function is activated, based on measurements of the vehicle's wheel speed. T In this paper, the wheel speed (value) of a vehicle measured at any given time point is denoted by v, which is obtained using wheel speed sensors of any configuration or type known in the art. A series of measurement time points and the corresponding series of wheel speeds can be stored in a sufficiently large time (point) array t = [t0 t1 t2 … t a The wheel speed (value) array v = [v0 v1 v2 ... v a In this context, 'a' is a positive integer sufficient to satisfy the storage function. An example of a wheel speed sensor that can be used is a Hall effect sensor. In this paper, the vehicle speed is denoted by V, and the vehicle speed prediction method of this application is used to calculate the vehicle speed V at time point T. T .
[0027] The vehicle speed prediction method of this application begins upon activation of a special function, first including a starting step S1 for activating the special function. In this step, the special function is activated ( Figure 2a The A0 position is determined, and the wheel speed v0 at that time point t0 is measured, along with the storage of the time point t0 and wheel speed v0 (which can be referred to as the initial time point and initial wheel speed, respectively, in this paper). Although Figures 2a-2c The initial time point t0 is shown as an offset from the coordinate zero point; however, optionally, since this application begins when a special function is activated, for simplicity, the initial time point t0 in this method can be set to 0. The start step S1 also includes setting the state parameter U to 0. The start step S1 may further include assigning the initial time point t0 and the initial wheel speed v0 to the first time point tt1 parameter and the first wheel speed vv1 parameter. The state parameter U will be described later.
[0028] After starting step S1, this method will proceed to vehicle speed prediction step S2, in which the vehicle speed V at the predicted time point T (referred to as the "target time point") is repeatedly predicted at predetermined time increments or time intervals. TThe target time point T is either the time elapsed from the initial time point t0, or the time elapsed since the initial time point t0. The speed prediction algorithm will select from two pre-stored algorithms or formulas F1 and F2 based on the value of the state parameter U. When the state parameter U is 0, the speed prediction algorithm selects formula F1:
[0029] V T =v (0) -k0×T (F1).
[0030] Formula F1 corresponds to Figure 2a In the case described above, no valid peak value was detected at the target time T after the initial time point t0. Figure 2b and 2c Time point t p1 Before point A1, the vehicle speed prediction follows a predetermined linear law Q0 based on the vehicle model. The coefficient k0 in formula F1 is the slope of the straight line Q0, and therefore depends on the vehicle model; it is a constant coefficient that can be obtained or calibrated through experience, experimentation, or other methods known in the art. For example, for a certain vehicle model, coefficient k0 can be equal to 0.6. For another vehicle model, coefficient k0 can be equal to 1.8. That is, the vehicle speed V at the target time point T after the initial time point t0... T Calculate or predict according to a predetermined linear law Q0.
[0031] Optionally, the vehicle speed prediction step S2 further includes an ending step S3, which may include at least one of the following: storing all time points T where the vehicle speed was predicted or calculated and the corresponding calculated vehicle speed V. T Based on all stored time points T and vehicle speed V T Plot curves for reporting or analysis purposes; notify the driver when the predicted speed exceeds the predetermined value.
[0032] like Figure 1 As shown, after starting step S1, the loop L for detecting peak vehicle speed is performed simultaneously with the vehicle speed prediction step S2. In general, the loop L for detecting peak vehicle speed includes: continuously and repeatedly measuring wheel speed (values); selecting a set of consecutive time points (sometimes referred to as "moments" in this document) and a set of wheel speed values corresponding to that set of time points; calculating the maximum vehicle speed, or peak speed, within a specific period (time interval) defined by the set of consecutive time points; determining whether the peak speed is a valid peak available in the vehicle speed prediction step S2; and when the peak speed is a valid peak, sending the time point corresponding to the valid peak (called the peak time point) and the measured wheel speed corresponding to that time point (called the peak wheel speed) to the vehicle speed prediction step S2 for vehicle speed prediction.
[0033] The number of time points selected in loop L (represented by n in this paper) can be selected or set through trial and error based on the vehicle model, the expected calculation accuracy, and the type of special function (ABS). For example, commonly used values for n are 6, 7, 8, 9, 10, etc. Of course, n can also be any other possible positive integer.
[0034] The aforementioned state parameter U is a parameter set to indicate whether a valid peak value has been detected in the cycle L for detecting the peak vehicle speed, and it can be 0 or 1. In the above-described start step S1, setting the state parameter U to 0 indicates that no valid (vehicle speed) peak value has been identified or detected, and setting the state parameter U to 1 indicates that at least one true peak value has been detected. In this document, the terms "peak value" and "peak vehicle speed" are used interchangeably, referring to the maximum vehicle speed calculated in the cycle L for detecting the peak vehicle speed, indicating the maximum vehicle speed within a specific period defined by that set of consecutive time points. A "valid peak value" refers to a value equal to or greater than a predetermined threshold F. THD The peak value is used for vehicle speed prediction in step S2 of the vehicle speed prediction method of this application. Conversely, values less than the threshold F THD The peak value is not used in the vehicle speed prediction step S2 of the vehicle speed prediction method in this application, and is not a "valid peak value," but can be called an "invalid peak value." The predetermined threshold F... THD These are preset values obtained or calibrated through experience or experimentation for a specific vehicle model; for example, they can be constant.
[0035] The following describes in detail the cycle L for detecting peak vehicle speed.
[0036] The loop L first includes an initial step L1. This step may include parameter setting operations: setting the state parameter U to 1 and the loop count counter r to 1; and includes repeated measurement operations: repeatedly measuring the vehicle's wheel speed at predetermined time intervals or increments and storing the measured wheel speeds and corresponding time points. The measurement operations in this step can be performed using sensors already configured on the vehicle, which are generally equipped with wheel speed sensors and are constantly performing wheel speed measurement operations. This application can be configured to receive continuously detected wheel speed values from other systems or sensors on the vehicle.
[0037] At each time point t a and corresponding to each time point t a Wheel speed V a The parameters can be stored in any way. As an example, this application can store the above parameters as a velocity matrix (or array) M1:
[0038]
[0039] In another example, at each time point ta and corresponding to each time point t a Wheel speed V a They can be stored as two separate arrays or matrices:
[0040] t = [t1 t2 …t a ];
[0041] v = [v1 v2 …v] a ].
[0042] Next, loop L may include step L2, which generates the wheel speed difference matrix. This step involves selecting the corresponding n time points t based on the value of the loop count counter r. r , t r+1 , ...t r+n-1 and n wheel speeds v r v r+1 ,...v r+n-1 And generate an n*n wheel speed difference matrix M2:
[0043]
[0044] In this matrix, each element M ij (the i-th row and j-th column, where i and j are both positive integers between 1 and n) can be expressed as: M ij =v r+i-1 -v r+j-1 .
[0045] For example, if the cycle count counter r is 1, then the wheel speed difference matrix M2 is determined by the time points t1, t2...t... n-1 The measured wheel speeds v1, v2...v n-1 Composition. If the cycle count counter r is 3, then the wheel speed difference matrix M2 is composed of time points t3, t4...t... n+2 The measured wheel speeds v3, v4...v n+2 constitute.
[0046] The loop L then includes calculating the peak vehicle speed F based on the wheel speed difference matrix M2, the peak factor matrix K, and the correction factor matrix m. p The calculation step L3 is calculated using the formula F. p = K·M2·m, that is:
[0047]
[0048]
[0049] An nx1 constant matrix of environmental impacts, where each element m i It can also be obtained through experience or experimental methods:
[0050]
[0051] In one example, for a certain vehicle model, when n=7, the peak factor matrix K obtained through empirical or experimental methods can be set as: K=[1 2 3 3 3 2 1]. In another example, for another vehicle model, when n=4, the peak factor matrix K obtained through empirical or experimental methods can be set as: K=[1 2 2 1].
[0052] In one example, the correction factor matrix *m* is a correction parameter matrix based on wheel speed noise. Through experimentation, when wheel speed noise or disturbances are excessive, such as when n=6, the correction factor matrix *m* can be set to the following values: *m* = [0.5 0.5 11 0.5 0.5]. When wheel speed noise is normal, if n=4, all elements of the correction factor matrix *m* can be set to 1, i.e., *m* = [1 1 1]. In fact, in most cases, all elements of the correction factor matrix *m* are set to 1.
[0053] Optionally, the environmental factors considered in obtaining or calibrating the correction factor matrix m are not limited to wheel noise, but may also include environmental factors (such as weather) and road factors.
[0054] The peak vehicle speed F is calculated in calculation step L3. p Then, this method performs a comparison calculation of the peak value F. p Use a predetermined threshold to determine the peak value F p Step L4 involves comparing whether a peak value is valid. Peak value F p If the value is greater than or equal to a predetermined threshold, the peak value F is determined. p It is a valid peak value that can be used to predict vehicle speed, and can be used in the vehicle speed prediction step S2 of this application. Otherwise, the peak value F is determined. p It is an invalid peak, that is, at the selected time point t. r , t r+1 , ...t r+n-1 No valid peak values for predicting vehicle speed were detected during that period. The predetermined (peak) threshold is also a constant value predetermined based on the vehicle model and can be obtained through trial or experience.
[0055] If the peak value F is determined in comparison step L4 p If the peak value is greater than or equal to a predetermined threshold, this method executes update step L5: select time point t. r , t r+1 , ...t r+n-1 One of the time points is used as the effective peak value F p The corresponding peak time point t pThe peak time point t p With the time point corresponding to the peak t p Peak wheel speed v p (wheel speed v) r v r+1 ,...v r+n-1 (One of them) assigns a second time parameter tt2 and a second wheel speed parameter vv2, and sends the second time parameter tt2 and the second wheel speed parameter vv2 to the vehicle speed prediction step S2.
[0056] As mentioned above, update the peak time point t in step L5. p From the currently selected set of time points t r , t r+1 , ...t r+n-1 The selection is made from the available time points. In some embodiments, the time lag caused by executing loop L can be disregarded, and the last time point t can be selected. r+n-1 As the peak time point t p In some embodiments, when considering the time lag factor, this set of time points t can be selected. r , t r+1 , ...t r+n-1 The midpoint of time or the preceding or following time point adjacent to the midpoint is taken as the peak time point t. p Optionally, select this set of time points t. r , t r+1 , ...t r+n-1 It is possible at any point in time. Peak wheel speed v p That is, in measurement step S2 at time point t p The measured wheel speed. For example, in the initial step L1 of cycle L, the state parameter U is 1. Figure 2a and 2b The results show that two valid peak values were detected in two different loops L, and the peak time points were determined as t. p1 and t p2 .
[0057] As described above, the vehicle speed prediction step S2 repeatedly predicts the vehicle speed based on the value of the state parameter U. Before receiving data from the update step L5 of loop L, the state parameter U remains 0, and the vehicle speed V is... T The calculation is always based on formula F1. When the vehicle speed prediction step S2 receives data from the update step L5 of loop L, and at all subsequent target time points T, the state parameter U is 1, and the vehicle speed prediction step S2 will calculate based on formula F2:
[0058] V T=vv2-(vv1-vv2) / (tt2-tt1)×(T-tt2) (F2).
[0059] For the first detection of an effective peak value F p When, refer to Figure 2b The first time parameter tt1 and the first wheel speed parameter vv1 are the initial time point t0 and the corresponding initial wheel speed v0, i.e., position point A0, which are assigned by the special function activation in the start step S1 of this method before executing the peak detection loop L; the second time parameter tt2 and the second wheel speed parameter vv2 are the values assigned in the update step L5, corresponding to the effective peak value F. p The corresponding peak time point t p1 and peak wheel speed v p1 That is, location point A1. Therefore, formula F2 corresponds to... Figure 2b The linear variation law B1. The linear variation law B1 can be understood as connecting the special function activation position (position A0 with time point t0 and wheel speed v0) and the first effective peak value F. p Location (time point t) p1 Wheel speed is v p1 The line connecting point A1 is used to calculate the position at time t. p1 The vehicle speed V at the target time point T afterwards T The target time point T is started from the initial time point.
[0060] After updating the second time parameter tt2 and the second wheel speed parameter vv2 in update step L5, this method returns to test step L1 of loop L via reset step L6. In reset step L6, the loop counter R is set to 1, and the first time parameter tt1 and the first wheel speed parameter vv1 are updated with the second time parameter tt2 and the second wheel speed parameter vv2. After reset step L6, the method returns to test step L1 to redetermine or update the time array and wheel speed array t = [t1 t2 …t a ] and v = [v1 v2 …v a The loop L that detects the peak vehicle speed is started repeatedly.
[0061] The purpose of performing the update operation in reset step L6 is to switch the vehicle speed prediction step S2 from executing formula F1 to executing formula F2 when the first effective peak value is determined, that is, to use the linear change law B1 to predict the time point t. p1The predicted result for the target time point T is as indicated by the extension line Q1 of line segment B1. Subsequently, each time a valid peak is detected, the first time parameter tt1 and the first wheel speed parameter vv1, which are currently executed in the vehicle speed prediction step S2, have been updated in the reset step L6 to the peak time point and peak wheel speed corresponding to the previous valid peak. The second time parameter tt2 and the second wheel speed parameter vv2 have also been updated with the peak time point and peak wheel speed corresponding to the latest valid peak.
[0062] For example, refer to Figure 2c At the time point t corresponding to the detection of the second valid peak, p2 and peak wheel speed v p2 When sent to the vehicle speed prediction step S2, the second time parameter tt2 and the second wheel speed parameter vv2 in formula F2 are the peak time points t corresponding to the second effective peak value. p2 and peak wheel speed v p2 The first time parameter tt1 and the first wheel speed parameter vv1 are the peak time points t corresponding to the first effective peak value, which were assigned in the previous reset step L6. p1 and peak wheel speed v p1 The target time T is at the peak time t. p2 At subsequent time points, formula F2 corresponds to Figure 2c The linear variation law B2, the prediction of vehicle speed at target time point T is based on the peak time points A1 and A2 of the two effective peaks preceding target time point T and the peak wheel speed. That is, it is based on the position point A1(t p1 v p1 ) and location point A2(t p2 v p2 The linear variation law B2 defined for the peak time point t p2 The vehicle speed is then predicted at the target time point T, and the prediction result is shown as Q2, the extension of B2.
[0063] Optionally, but not necessarily, the reset step L6 may further include pausing the execution of the method for a predetermined period of time, or returning to the test step L1 after a predetermined period of time following the completion of the data update operation described above in the reset step L6. The purpose of this is to avoid detecting valid peaks too frequently.
[0064] Now return to comparison step L4. If the peak value F is determined through comparison in comparison step S4... p If the speed is less than a predetermined threshold, then the peak vehicle speed F calculated in step L3 is determined to be... pIf the peak value is invalid, this method will return to step L7 to generate step L2 from the wheel speed difference matrix of loop L to continue executing loop L and continue detecting valid peak values. In step L7, the loop count counter r is incremented by 1. At this time, step L2, which generates the wheel speed difference matrix M2, uses a set of time points and a set of wheel speeds that are still derived from the time array and wheel speed array measured in step L1, the only difference being that the value of r is incremented by 1. For example, if r = 3 before returning to step L7, the time points in the just completed loop L include t3, t4, ... t n+2 After returning to step L7, the process returns to step L2, where the wheel speed difference matrix is generated. The time points used to generate the wheel speed difference matrix M2 include t4, t5, ... t n+3 At this point, r = 4.
[0065] As mentioned above, the vehicle speed prediction technology or method of this application only requires input other than the measured wheel speed, and no other input is needed, and it can produce a relatively accurate vehicle speed prediction value.
[0066] This application also relates to a computer program product comprising a computer-readable program that is permissible on a computer or processor or any other possible hardware, which, when run, causes the computer or processor or any other possible hardware to perform the vehicle prediction method and functions described above in this application.
[0067] This application also relates to a computer-readable medium comprising a computer-readable program that is permissible on a computer or processor or any other possible hardware, which, when run, causes the computer or processor or any other possible hardware to perform the vehicle prediction method and functions described above in this application.
Claims
1. A method for predicting vehicle speed when a special function is activated, comprising: Activate the special function and measure the initial wheel speed (v0) at the initial time point (t0); then simultaneously predict the vehicle speed (V) at the target time point (T). T The vehicle speed prediction step (S2) and the loop (L) for detecting peak vehicle speeds are performed. The vehicle speed prediction step (S2) is performed as follows: before a valid peak is detected in the cycle (L), it is based on the initial wheel speed, a predetermined slope value depending on the vehicle model, and a target time point (T); after a valid peak is detected in the cycle (L), it is performed based on the peak time point and peak wheel speed corresponding to the current valid peak, the peak time point and peak wheel speed corresponding to the previous valid peak, or, when the current valid peak is the first valid peak, based on the initial time point and initial wheel speed, and based on the time interval between the target time point (T) and the peak time point corresponding to the current valid peak or the initial time point. The loop (L) includes: a first step (L1) of repeatedly measuring wheel speeds and recording each wheel speed and its corresponding time point; a second step (L2) of selecting a set of consecutive time points and a corresponding set of wheel speeds and generating a wheel speed difference matrix; a third step (L3) of calculating the vehicle speed peak based on the wheel speed difference matrix, a predetermined peak factor matrix depending on the vehicle model, and a predetermined correction factor matrix depending on environmental factors; a fourth step (L4) of comparing the vehicle speed peak with a predetermined peak threshold depending on the vehicle model to determine whether the vehicle speed peak is a valid peak; and a fifth step: when the vehicle speed peak is determined to be a valid peak, determining the peak time point and peak wheel speed corresponding to the valid peak and providing them to the vehicle speed prediction step (S2), and repeating steps 2 (L2) to 5 of the loop (L) when the vehicle speed peak is determined not to be a valid peak, wherein the set of time points selected in the second step (L2) is the same number of consecutive time points starting from the second time point in the previous set of time points. The element in the i-th row and j-th column of the wheel speed difference matrix represents the difference between the i-th wheel speed and the j-th wheel speed in the set of wheel speeds.
2. The method according to claim 1, wherein, The slope value, the peak factor matrix and the correction factor matrix, the preset threshold, and the number of time points contained in the set of time points are obtained or calibrated through experimentation or empirical methods.
3. The method according to claim 2, wherein, When the special function is activated, the initial time point and initial wheel speed are stored as the first time point and the first wheel speed. After determining that the peak vehicle speed is a valid peak in each loop (L), the peak time point and peak wheel speed corresponding to the valid peak are stored as the second time point and the second wheel speed and sent to the vehicle speed prediction step (S2). Before re-executing the loop (L), the first time point and the first wheel speed are updated using the second time point and the second wheel speed.
4. The method according to claim 3, wherein, Immediately after the update or after waiting for a predetermined period of time, return to the beginning step (L1) of the loop (L) and re-execute the loop (L).
5. The method according to any one of claims 1-4, in, The set of time points and the set of wheel speeds in the cycle (L) are respectively denoted as t1, t2…t n-1 and v1, v2...v n-1 This indicates that n is the number of time points; The peak factor matrix is a 1xn matrix; The correction factor matrix is an nx1 matrix; The calculation of peak vehicle speed includes multiplying the wheel speed difference matrix, the peak factor matrix, and the correction factor matrix to obtain the peak vehicle speed.
6. The method according to claim 5, wherein, The correction factor matrix is a constant matrix based on wheel noise. When the wheel noise level is less than the noise threshold, all elements in the correction factor matrix are 1.
7. The method according to any one of claims 1-6, wherein, When it is determined that the peak vehicle speed is a valid peak, one of the time points is selected from the set of time points as the peak time point and the corresponding wheel speed is selected as the peak wheel speed.
8. The method according to claim 7, wherein, The peak time point is the last time point in the set of time points, or the middle time point in the set of time points, or a time point adjacent to the middle time point.
9. A computer program product comprising a computer-readable program that, when executed by a computer or processor, causes the computer or processor to perform the method according to any one of claims 1-8.