Method and device for updating vehicle dynamic prediction model

Abstract

The present invention provides a method and device for updating a vehicle dynamic prediction model. The method includes: obtaining the actual steering wheel angle and the actual yaw rate within a target time window; performing normalization processing on the actual steering wheel angle within the target time window; inputting the normalized actual steering wheel angle into the vehicle dynamic prediction model, and the vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; updating the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model. The present invention realizes the online automatic update of the vehicle dynamic model, improves the accuracy of model prediction, and further facilitates subsequent control according to the model prediction results, meeting the accuracy requirements of control.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and particularly to a method and device for updating a vehicle dynamic prediction model. Background Art

[0002] With the popularization of autonomous driving vehicles, autonomous driving vehicles can be used as taxis or public transportation. When passengers use autonomous driving vehicles, they need to input a destination. The autonomous driving vehicle generates a driving route based on the current location and the destination, and travels according to the generated driving route. Precise control of vehicle parameters is a prerequisite for traveling according to the driving route.

[0003] Currently, most static modeling and parameter calibration of vehicles and the environment are performed, and each vehicle is independently calibrated and modeled. However, due to uncertainties such as flexible trailers, cross-slope cross-wind, etc., a single static model is difficult to cover the parameter changes of vehicles and cannot adapt to all operating scenarios. In addition, after large-scale operation, fine parameter adjustment for each vehicle makes the cost high. At the same time, the vehicle consistency of some types of vehicles is poor, and parameter drift is likely to occur during operation. For example, for the steering zero position, the standard for passenger cars off the production line is usually within 1°, while the standard for commercial vehicles is 15°. With the accumulation of operating mileage, under the influence of various factors such as tie rods and tires, the steering zero position will deviate to varying degrees. Summary of the Invention

[0004] The present invention provides a method and device for updating a vehicle dynamic prediction model to solve the defect that the existing static model in the prior art is difficult to cover the parameter changes of vehicles and cannot adapt to all operating scenarios. By adopting the model dynamic update technology, the model is updated online in real time to improve the accuracy of model prediction, and then it is convenient to perform control according to the model prediction result to meet the accuracy requirements of control.

[0005] The present invention provides a method for updating a vehicle dynamic prediction model, including: obtaining the actual steering wheel angle and the actual yaw rate within a target time window; normalizing the actual steering wheel angle within the target time window; inputting the normalized actual steering wheel angle into a vehicle dynamic prediction model, and the vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; updating the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

[0006] A method for updating a vehicle dynamic prediction model provided by the present invention updates the vehicle dynamic prediction model to obtain a vehicle dynamic updated model, including: updating the vehicle dynamic prediction model to obtain a vehicle dynamic updated model, including: obtaining a yaw rate error according to the predicted yaw rate and the actual yaw rate within the target time window; updating the vehicle dynamic prediction model according to the actually normalized steering wheel angle, a first preset value, and the yaw rate error; determining a response strength of the updated vehicle dynamic prediction model, where the response strength is the sum of all weight values of the updated vehicle dynamic prediction model; determining whether the response strength is within a preset interval, and determining a vehicle dynamic updated model according to the determination result.

[0007] A method for updating a vehicle dynamic prediction model provided by the present invention, where determining the vehicle dynamic updated model according to the determination result includes: based on the response strength being outside the preset interval, adjusting the first preset value; using the adjusted first preset value, combining the actually normalized steering wheel angle and the yaw rate error, and re-updating the vehicle dynamic prediction model; according to the re-updated vehicle dynamic prediction model, re-determining the response strength; re-determining whether the re-determined response strength is within the preset interval, so as to re-determine the vehicle dynamic updated model according to the re-determined result.

[0008] A method for updating a vehicle dynamic prediction model provided by the present invention, where based on the response strength being outside the preset interval, adjusting the first preset value includes: based on the response strength being greater than the maximum boundary value of the preset interval, reducing the first preset value; based on the response strength being less than the minimum boundary value of the preset interval, increasing the first preset value.

[0009] A method for updating a vehicle dynamic prediction model provided by the present invention, where determining the vehicle dynamic updated model according to the determination result further includes: based on the response strength being within the preset interval, determining the currently updated vehicle dynamic prediction model as the vehicle dynamic updated model.

[0010] A method for updating a vehicle dynamic prediction model provided by the present invention, where normalizing the actually steering wheel angle within the target time window includes: obtaining the torque corresponding to the actually steering wheel angle within the target time window; obtaining the median steering wheel torque according to the actually steering wheel angle within the target time window, the torque corresponding to the actually steering wheel angle, and a pre-established impulse response model; obtaining the actually normalized steering wheel angle according to the median steering wheel torque and the corresponding actually steering wheel angle.

[0011] A method for updating a vehicle dynamic prediction model provided by the present invention, after obtaining the actual steering wheel angle and the actual yaw rate within a target time window, includes: obtaining the corresponding average steering wheel angle and average yaw rate according to the actual steering wheel angle and the actual yaw rate within the target time window; respectively reducing each actual steering wheel angle within the target time window according to the average steering wheel angle; and respectively reducing each actual yaw rate within the target time window according to the average yaw rate.

[0012] The present invention also provides a vehicle dynamic prediction model updating device, including: a data acquisition module, which acquires the actual steering wheel angle and the actual yaw rate within a target time window; a normalization processing module, which performs normalization processing on the actual steering wheel angle within the target time window; an angle prediction module, which inputs the normalized actual steering wheel angle into the vehicle dynamic prediction model, and the vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; and a model updating module, which updates the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the steps of the vehicle dynamic prediction model updating method as described in any one of the above are implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the vehicle dynamic prediction model updating method as described in any one of the above are implemented.

[0015] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the vehicle dynamic prediction model updating method as described in any one of the above are implemented.

[0016] The vehicle dynamic prediction model updating method and device provided by the present invention obtain the actual steering wheel angle and the actual yaw rate in real time through a target time window, so as to update the vehicle dynamic prediction model by using the actually obtained actual steering wheel angle and actual yaw rate in real time; by normalizing the obtained actual steering wheel angle, the actual steering wheel angle affected by the steering wheel dead zone is obtained, which is convenient for improving the accuracy of subsequent vehicle control; the predicted yaw rate, the preprocessed actual yaw rate and steering wheel angle are used to update the vehicle dynamic prediction model, so that it does not depend on the vehicle specifications, the mass and mass distribution of the hanging box, and the adhesion coefficient of the ground, etc. Only based on the steering wheel response of the vehicle, the vehicle dynamic model can be updated automatically online, the prediction accuracy of the model can be improved, and then it is convenient to perform control according to the model prediction result subsequently to meet the accuracy requirements of control, avoid the situation of adjusting parameters for each vehicle, adapt to various driving environments, and is suitable for large-scale mass production deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 is one of the flow diagrams of the vehicle dynamic prediction model updating method provided by the present invention;

[0019] Figure 2 is the second flow diagram of the vehicle dynamic prediction model updating method provided by the present invention;

[0020] Figure 3 is the structural diagram of the vehicle dynamic prediction model updating device provided by the present invention;

[0021] Figure 4 is the structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0023] Figure 1The flowchart of a method for updating a vehicle dynamic prediction model according to the present invention is shown. The method includes:

[0024] S11, obtaining the actual steering wheel angle and the actual yaw rate within a target time window;

[0025] S12, performing normalization processing on the actual steering wheel angle within the target time window;

[0026] S13, inputting the normalized actual steering wheel angle into the vehicle dynamic prediction model. The vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate;

[0027] S14, updating the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

[0028] It should be noted that S1N in this specification does not represent the sequence of the vehicle dynamic prediction model update method. The following specifically describes the vehicle dynamic prediction model update method of the present invention in combination with Figure 2 Describe the vehicle dynamic prediction model update method of the present invention.

[0029] Step S11, obtaining the actual steering wheel angle and the actual yaw rate within a target time window.

[0030] In this embodiment, the actual steering wheel angle and the actual yaw rate obtained within the target time window include: obtaining the trajectory points within the target time window; according to each trajectory point, obtaining the corresponding actual steering wheel angle and actual yaw rate of each trajectory point.

[0031] It should be noted that the length of the target time window can be set according to actual design requirements, such as 3s, 6s, 10s, etc., and no further limitation is made here. For example, if the length of the target time window is 3 seconds, then the actual steering wheel angle and the actual yaw rate within the historical 3 seconds are obtained based on the target time window. If the system sampling frequency is 50hz, then the actual steering wheel angle and the actual yaw rate of 150 trajectory points are taken as the model input. Among them, the actual steering wheel angle and the actual yaw rate of the first trajectory point are the model input corresponding to the current time, and the actual steering wheel angle and the actual yaw rate of the 2-150th trajectory points are the model input within the historical 0.02 - historical 3 seconds, so as to obtain the actual steering wheel angle and the actual yaw rate online and in real time according to the target time window, and then facilitate the subsequent update of the vehicle dynamic prediction model using the real-time obtained data.

[0032] In an alternative embodiment, since there are zero-offset and non-linear situations for some types of vehicles, such as trucks, after obtaining the actual steering wheel angle and actual yaw rate within the target time window, it includes: preprocessing the actual steering wheel angle and actual yaw rate within the target time window. Specifically, it includes: obtaining the corresponding average steering wheel angle and average yaw rate according to the actual steering wheel angle and actual yaw rate within the target time window; reducing each actual steering wheel angle within the target time window according to the average steering wheel angle; reducing each actual yaw rate within the target time window according to the average yaw rate.

[0033] It should be noted that reducing each actual steering wheel angle within the target time window according to the average steering wheel angle means subtracting the average steering wheel angle from each actual steering wheel angle within the target time window respectively, so as to preprocess the actual steering wheel angle obtained within the target time window.

[0034] Similarly, reducing each actual yaw rate within the target time window according to the average yaw rate means subtracting the average yaw rate from each actual yaw rate within the target time window respectively.

[0035] It should be noted that the preprocessing of the actual steering wheel angle and actual yaw rate within the target time window can be performed before or after the normalization processing of the actual steering wheel angle within the target time window, and no further limitation is made here.

[0036] It should be noted that if the preprocessing of the actual steering wheel angle and actual yaw rate within the target time window is performed before the normalization processing of the actual steering wheel angle within the target time window, then when normalizing the actual steering wheel angle within the target time window, the preprocessed actual steering wheel angle is normalized. Similarly, if the preprocessing of the actual steering wheel angle and actual yaw rate within the target time window is performed after the normalization processing of the actual steering wheel angle within the target time window, then when preprocessing the actual steering wheel angle within the target time window, the normalized actual steering wheel angle needs to be preprocessed.

[0037] Step S12, normalizing the actual steering wheel angle within the target time window.

[0038] In this embodiment, normalizing the actual steering wheel angle within the target time window includes: obtaining the torque corresponding to the actual steering wheel angle within the target time window; obtaining the median steering wheel torque according to the actual steering wheel angle within the target time window, the torque corresponding to the actual steering wheel angle, and the pre-established impulse response model; obtaining the actual steering wheel angle after normalization processing according to the median steering wheel torque and the corresponding actual steering wheel angle.

[0039] It should be noted that the impulse response model is expressed as:

[0040] y1 = h1 T *P + b

[0041] Wherein, y1 represents the output of the impulse response model, that is, the actual steering wheel angle; P represents the input of the impulse response model, that is, the torque corresponding to the actual steering wheel angle within the target time window, expressed as a row vector; h1 represents the unit impulse response of the impulse response model, and h1 T represents the transpose of h1. When the torque is 0, b correspondingly represents the steering wheel torque median. It should be noted that since h1 is a column vector, for the convenience of calculation, h1 needs to be transposed into the corresponding row vector h1 T .

[0042] It should be noted that the impulse response model can adopt an online-updated FIR model. During the vehicle driving process, both h1 and b will be updated according to the input actual steering wheel angle and the torque corresponding to the actual steering wheel angle. For the specific update method, reference can be made to the vehicle dynamic prediction model described below, and it will not be described here.

[0043] In addition, since there is a dead zone in the steering wheel, that is, within a certain angle range in the center of the steering wheel, the rotation of the steering wheel will not cause vehicle movement. Therefore, it is necessary to first determine the steering wheel torque median of the vehicle, and then determine the influence of the dead zone on the steering wheel angle according to the distance between the actual steering wheel angle and the steering wheel torque median, so as to obtain the steering wheel angle affected by the dead zone, that is, the actual steering wheel angle after the above normalization processing.

[0044] Furthermore, according to the steering wheel torque median and the corresponding actual steering wheel angle, the actual steering wheel angle after normalization processing is obtained, including: obtaining the influence degree of the dead zone on the steering wheel angle according to the steering wheel torque median and the corresponding actual steering wheel angle, combined with the vehicle weight and the preset dead zone sensitivity; obtaining the actual steering wheel angle after normalization processing according to the preset dead zone width, the actual steering wheel angle and the influence degree of the dead zone on the steering wheel angle.

[0045] In this embodiment, the actual steering wheel angle after normalization processing is expressed as:

[0046]

[0047] Wherein, y2 represents the actual steering wheel angle after normalization processing, u1 represents the actual steering wheel angle obtained within the target time window, tanh represents the function, m represents the vehicle weight, k1 represents the dead zone width, k2 represents the dead zone sensitivity, represents the influence degree of the dead zone on the steering wheel angle. For example, there is When y2 = y3 = d, the corresponding u1 and u2 are obtained. k1 represents the distance between u1 and u2, and k2 represents the steepness of the y2 curve at the point (c, 0). It should be noted that the tanh function changes rapidly near the point (c, 0) and approaches plus or minus 1 when the input is relatively large. Therefore, in the case of a large angle, a value of k1 * 1 is subtracted. At this time, the angle change is not within the dead zone; in the case of a small angle, the value obtained by tanh is relatively close to the original small angle, and after subtraction, it is close to 0, thus connecting the dead zone and the steering wheel torque median.

[0048] Step S13: Input the normalized actual steering wheel angle into the vehicle dynamic prediction model. The vehicle dynamic prediction model convolves the input actual steering wheel angle within the target time window with the unit impulse response of the vehicle dynamic prediction model to obtain the predicted yaw rate.

[0049] In this embodiment, the predicted yaw rate is expressed as:

[0050]

[0051] Wherein, represents the predicted yaw rate at the nth moment, M represents the number of trajectory points within the target time window, h(k) represents the unit impulse response of the vehicle dynamic prediction model, and u(n - k) represents the corresponding preprocessed actual steering wheel angle. It should be noted that the vehicle dynamic prediction model can adopt an existing FIR model or a model customized according to requirements for predicting the yaw rate to perform discrete convolution on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain the predicted yaw rate.

[0052] Step S14: Update the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain the vehicle dynamic updated model.

[0053] In this embodiment, updating the vehicle dynamic prediction model to obtain the vehicle dynamic updated model includes: obtaining the yaw rate error according to the predicted yaw rate and the actual yaw rate within the target time window; updating the vehicle dynamic prediction model according to the normalized actual steering wheel angle, the first preset value, and the yaw rate error; determining the response intensity of the updated vehicle dynamic prediction model, where the response intensity is the sum of all weight values of the updated vehicle dynamic prediction model; determining whether the response intensity is within the preset interval, and determining the vehicle dynamic updated model according to the determination result. Specifically:

[0054] First, based on the predicted yaw rate and the actual yaw rate within the target time window, the yaw rate error is obtained. In this embodiment, the yaw rate error is expressed as: where y represents the actual yaw rate after preprocessing, represents the predicted yaw rate.

[0055] Secondly, based on the actual steering wheel angle after normalization, the first preset value, and the yaw rate error, the vehicle dynamic prediction model is updated. More specifically, based on the actual steering wheel angle after preprocessing, the first preset value, and the yaw rate error, the unit impulse response of the vehicle dynamic prediction model is updated, thereby realizing the update of the vehicle dynamic prediction model. Among them, the unit impulse response of the updated vehicle dynamic prediction model is expressed as:

[0056] h new (k) = h(k) + Δh

[0057]

[0058] where h new (k) represents the unit impulse response of the updated vehicle dynamic prediction model, h(k) represents the unit impulse response of the vehicle dynamic prediction model before update, k3 represents the first preset value, δ represents the second preset value, e represents the yaw rate error, and u represents the steering wheel angle within the target time window after preprocessing and normalization. It should be noted that the first preset value and the second preset value can be set according to actual design requirements or prior experience, and no further limitation is made here.

[0059] For example, if the number of trajectory points within the target time window is 150, then correspondingly, u is represented as a 150*1 matrix, and the unit impulse response corresponding to the updated vehicle dynamic prediction model is also represented as a 150*1 matrix.

[0060] Subsequently, the response intensity of the updated vehicle dynamic prediction model is determined. The response intensity is the sum of all weight values of the updated vehicle dynamic prediction model. It should be noted that the response intensity of the updated vehicle dynamic prediction model can be determined according to all weight values of the updated vehicle dynamic prediction model, so as to determine the steady-state response situation of the corresponding vehicle dynamic prediction model according to the response intensity.

[0061] Finally, determine whether the response intensity is within a preset interval, and determine the vehicle dynamic update model according to the determination result. In this embodiment, determining the vehicle dynamic update model according to the determination result includes: based on the response intensity being outside the preset interval, adjusting a first preset value; using the adjusted first preset value, combining the actual steering wheel angle and yaw rate error after normalization processing, and updating the vehicle dynamic prediction model again; according to the newly updated vehicle dynamic prediction model, determining the response intensity again; re-determining whether the re-determined response intensity is within the preset interval, so as to re-determine the vehicle dynamic update model according to the determination result.

[0062] It should be noted that, regardless of the vehicle specifications, the mass and mass distribution of the hanging box, and the ground adhesion coefficient, etc., only based on the steering wheel response of the vehicle, the vehicle dynamic model can be automatically updated, avoiding the situation of adjusting parameters for each vehicle, adapting to various driving environments, and being suitable for large-scale mass production deployment.

[0063] Furthermore, based on the response intensity being outside the preset interval, adjusting the first preset value includes: based on the response intensity being greater than the maximum boundary value of the preset interval, decreasing the first preset value; based on the response intensity being less than the minimum boundary value of the preset interval, increasing the first preset value. By adjusting the size of the first preset value, the response intensity of the updated vehicle dynamic prediction model is made to be within the preset interval.

[0064] In an alternative embodiment, determining the vehicle dynamic update model according to the determination result further includes: based on the response intensity being within the preset interval, determining the currently updated vehicle dynamic prediction model as the vehicle dynamic update model.

[0065] In summary, the embodiment of the present invention obtains the actual steering wheel angle and actual yaw rate in real time through the target time window, so as to update the vehicle dynamic prediction model by using the actually obtained steering wheel angle and yaw rate in real time later; by normalizing the obtained actual steering wheel angle, the actual steering wheel angle affected by the steering wheel dead zone is obtained, which is convenient for improving the accuracy of subsequent vehicle control; using the predicted yaw rate and the preprocessed actual yaw rate and steering wheel angle to update the vehicle dynamic prediction model, so that regardless of the vehicle specifications, the mass and mass distribution of the hanging box, and the ground adhesion coefficient, etc., only based on the steering wheel response of the vehicle, the vehicle dynamic model can be updated online automatically, improving the accuracy of model prediction, and then being convenient for subsequent control according to the model prediction result to meet the accuracy requirements of control, avoiding the situation of adjusting parameters for each vehicle, adapting to various driving environments, and being suitable for mass production deployment.

[0066] The vehicle dynamic prediction model updating device provided by the present invention will be described below. The vehicle dynamic prediction model updating device described below can be correspondingly referred to the vehicle dynamic prediction model updating method described above.

[0067] Figure 3 FIG. shows a schematic structural diagram of a vehicle dynamic prediction model updating device, which includes:

[0068] A data acquisition module 31, which acquires the actual steering wheel angle and the actual yaw rate within a target time window;

[0069] A normalization processing module 32, which performs normalization processing on the actual steering wheel angle within the target time window;

[0070] An angle prediction module 33, which inputs the normalized actual steering wheel angle into the vehicle dynamic prediction model. The vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate;

[0071] A model updating module 34, which updates the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

[0072] In this embodiment, the data acquisition module 31 includes: a trajectory acquisition unit, which acquires trajectory points within the target time window; a data acquisition unit, which acquires the corresponding actual steering wheel angle and actual yaw rate for each trajectory point according to each trajectory point.

[0073] In an optional embodiment, since some types of vehicles, such as trucks, have zero bias and non-linearity, therefore, the device further includes: a preprocessing module, which performs preprocessing on the actual steering wheel angle and the actual yaw rate within the acquired target time window. Specifically, the preprocessing module includes: a mean value acquisition unit, which obtains the corresponding average steering wheel angle and average yaw rate according to the actual steering wheel angle and the actual yaw rate within the target time window; a first preprocessing unit, which respectively reduces each actual steering wheel angle within the target time window according to the average steering wheel angle; a second preprocessing unit, which respectively reduces each actual yaw rate within the target time window according to the average yaw rate.

[0074] It should be noted that the preprocessing module can perform preprocessing on the actual steering wheel angle and the actual yaw rate within the target time window before or after the normalization processing module 32 performs normalization processing on the actual steering wheel angle within the target time window, and no further limitation is made here.

[0075] The normalization processing module 32 includes: a torque acquisition unit that acquires the torque corresponding to the actual steering wheel angle within a target time window; a median acquisition unit that obtains the median of the steering wheel torque based on the actual steering wheel angle within the target time window, the torque corresponding to the actual steering wheel angle, and a pre-established impulse response model; and a normalization processing unit that obtains the normalized actual steering wheel angle based on the median of the steering wheel torque and the corresponding actual steering wheel angle.

[0076] Furthermore, the normalization processing unit includes: an influence degree determination subunit that obtains the influence degree of the dead zone on the steering wheel angle based on the median of the steering wheel torque and the corresponding actual steering wheel angle, in combination with the vehicle weight and a preset dead zone sensitivity; and an angle acquisition subunit that obtains the normalized actual steering wheel angle based on the preset dead zone width, the actual steering wheel angle, and the influence degree of the dead zone on the steering wheel angle.

[0077] In an optional embodiment, the normalization processing module 32 further includes an impulse response model update unit for updating the impulse response model. It should be noted that the impulse response model update unit can refer to the model update module 34 below and will not be further described here.

[0078] In an optional embodiment, the normalization processing module 32 further includes an impulse response model update unit for updating the impulse response model. It should be noted that the impulse response model update unit can refer to the model update module 34 below and will not be further described here.

[0079] The angle prediction module 33 includes: a data input unit that inputs the normalized actual steering wheel angle into a vehicle dynamic prediction model; and an angle prediction unit that convolves the vehicle dynamic prediction model based on the input actual steering wheel angle within the target time window with the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate.

[0080] The model update module 34 includes: an error acquisition unit that obtains a yaw rate error based on the predicted yaw rate and the actual yaw rate within the target time window; a model update unit that updates the vehicle dynamic prediction model based on the normalized actual steering wheel angle, a first preset value, and the yaw rate error; a response intensity determination unit that determines the response intensity of the updated vehicle dynamic prediction model, where the response intensity is the sum of all weight values of the updated vehicle dynamic prediction model; and a judgment unit that judges whether the response intensity is within a preset interval and determines the vehicle dynamic update model according to the judgment result.

[0081] Furthermore, the judgment unit includes: an adjustment subunit, which adjusts a first preset value based on the response intensity being outside a preset interval; a model re-update subunit, which re-updates the vehicle dynamic prediction model by using the adjusted first preset value and combining the actual steering wheel angle and yaw rate error after normalization processing; an intensity update subunit, which re-determines the response intensity according to the re-updated vehicle dynamic prediction model; and a secondary judgment unit, which re-judges whether the re-determined response intensity is within the preset interval, so as to re-determine the vehicle dynamic update model according to the judgment result.

[0082] The adjustment subunit includes: a first adjustment grandson subunit, which decreases the first preset value based on the response intensity being greater than the maximum boundary value of the preset interval; and a second adjustment grandson subunit, which increases the first preset value based on the response intensity being less than the minimum boundary value of the preset interval. By adjusting the size of the first preset value, the response intensity of the updated vehicle dynamic prediction model is made to be within the preset interval.

[0083] In an alternative embodiment, the judgment unit further includes: a model determination subunit, which determines the currently updated vehicle dynamic prediction model as the vehicle dynamic update model based on the response intensity being within the preset interval.

[0084] In summary, in the embodiment of the present invention, the actual steering wheel angle and the actual yaw rate are obtained in real time through the target time window, so as to subsequently update the vehicle dynamic prediction model by using the actually obtained steering wheel angle and yaw rate in real time; the obtained actual steering wheel angle is normalized by the normalization processing module to obtain the actual steering wheel angle affected by the steering wheel dead zone, which is convenient for improving the accuracy of subsequent vehicle control; the model update module updates the vehicle dynamic prediction model according to the predicted yaw rate and the preprocessed actual yaw rate and steering wheel angle, so that it does not depend on the vehicle specifications, does not depend on the mass and mass distribution of the hanging box, and does not depend on the ground adhesion coefficient, etc. Only based on the steering wheel response of the vehicle, the vehicle dynamic model is updated online automatically, the accuracy of model prediction is improved, and then it is convenient to perform control according to the model prediction result subsequently to meet the accuracy requirements of control, avoid the situation of parameter adjustment for each vehicle, adapt to various driving environments, and is suitable for large-scale mass production deployment.

[0085] Figure 4 An entity structure diagram of an electronic device is illustrated, as Figure 4As shown in the figure, the electronic device may include: a processor 41, a communications interface 42, a memory 43, and a communication bus 44. Among them, the processor 41, the communications interface 42, and the memory 43 complete communication with each other through the communication bus 44. The processor 41 may call the logical instructions in the memory 43 to execute the vehicle dynamic prediction model update method, which includes: obtaining the actual steering wheel angle and the actual yaw rate within a target time window; performing normalization processing on the actual steering wheel angle within the target time window; inputting the normalized actual steering wheel angle into the vehicle dynamic prediction model, and the vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; updating the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

[0086] In addition, when the logical instructions in the foregoing memory 43 can be implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0087] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the vehicle dynamic prediction model update method provided by each of the above methods. The method includes: obtaining the actual steering wheel angle and the actual yaw rate within a target time window; performing normalization processing on the actual steering wheel angle within the target time window; inputting the normalized actual steering wheel angle into the vehicle dynamic prediction model, and the vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; updating the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

[0088] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the vehicle dynamic prediction model update method provided by each of the above methods. The method includes: obtaining the actual steering wheel angle and the actual yaw rate within a target time window; performing normalization processing on the actual steering wheel angle within the target time window; inputting the normalized actual steering wheel angle into the vehicle dynamic prediction model, and the vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; updating the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

[0089] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0090] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for updating a vehicle dynamic prediction model, characterized in that, Including: Obtain the actual steering wheel angle and actual yaw rate within a target time window; Perform normalization processing on the actual steering wheel angle within the target time window; Input the normalized actual steering wheel angle into a vehicle dynamic prediction model, and the vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; Update the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

2. The method for updating a vehicle dynamic prediction model according to claim 1, wherein Updating the vehicle dynamic prediction model to obtain a vehicle dynamic updated model includes: Obtain a yaw rate error according to the predicted yaw rate and the actual yaw rate within the target time window; Update the vehicle dynamic prediction model according to the normalized actual steering wheel angle, a first preset value, and the yaw rate error; Determine the response strength of the updated vehicle dynamic prediction model, where the response strength is the sum of all weight values of the updated vehicle dynamic prediction model; Judge whether the response strength is within a preset interval and determine the vehicle dynamic updated model according to the judgment result.

3. The method for updating the vehicle dynamic prediction model according to claim 2, wherein The determining the vehicle dynamic updated model according to the judgment result includes: Based on the response strength being outside the preset interval, adjust the first preset value; Use the adjusted first preset value, combined with the normalized actual steering wheel angle and the yaw rate error, to re-update the vehicle dynamic prediction model; According to the re-updated vehicle dynamic prediction model, re-determine the response strength; Re-judge whether the re-determined response strength is within the preset interval to re-determine the vehicle dynamic updated model according to the re-judgment result.

4. The method for updating the vehicle dynamic prediction model according to claim 3, wherein The adjusting the first preset value based on the response strength being outside the preset interval includes: Based on the response strength being greater than the maximum boundary value of the preset interval, decrease the first preset value; Based on the response strength being less than the minimum boundary value of the preset interval, increase the first preset value.

5. The method for updating a vehicle dynamic prediction model according to claim 2, wherein The determining the vehicle dynamic updated model according to the judgment result further includes: Based on the response strength being within the preset interval, determine the currently updated vehicle dynamic prediction model as the vehicle dynamic updated model.

6. The method for updating a vehicle dynamic prediction model according to claim 1, wherein Performing normalization processing on the actual steering wheel angle within the target time window includes: Obtain the torque corresponding to the actual steering wheel angle within the target time window; Obtain the steering wheel torque median according to the actual steering wheel angle within the target time window, the torque corresponding to the actual steering wheel angle, and a pre-established impulse response model; Obtain the normalized actual steering wheel angle according to the steering wheel torque median and the corresponding actual steering wheel angle.

7. The method for updating a vehicle dynamic prediction model according to claim 1, wherein After obtaining the actual steering wheel angle and actual yaw rate within the target time window, it includes: Obtain the corresponding average steering wheel angle and average yaw rate according to the actual steering wheel angle and actual yaw rate within the target time window; Reduce each actual steering wheel angle within the target time window respectively according to the average value of the steering wheel angles. Reduce each actual yaw rate within the target time window respectively according to the average value of the yaw rates.

8. A vehicle dynamic prediction model updating device, characterized in that, Comprising: A data acquisition module, which acquires the actual steering wheel angle and the actual yaw rate within the target time window; A normalization processing module, which performs normalization processing on the actual steering wheel angle within the target time window; An angle prediction module, which inputs the normalized actual steering wheel angle into a vehicle dynamic prediction model. The vehicle dynamic prediction model performs convolution based on the input actual steering wheel angle within the target time window and the unit impulse response of the vehicle dynamic prediction model to obtain a predicted yaw rate; A model update module, which updates the vehicle dynamic prediction model according to the predicted yaw rate, the normalized actual steering wheel angle, and the actual yaw rate within the target time window to obtain a vehicle dynamic updated model.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, the steps of the vehicle dynamic prediction model update method according to any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the vehicle dynamic prediction model update method according to any one of claims 1 to 7 are implemented.

Images