Vehicle steering non-linearity estimation method and device

Abstract

The application provides a vehicle steering nonlinearity estimation method and device, the method comprising: obtaining vehicle steering related data to be predicted; inputting the vehicle steering related data to be predicted into a steering nonlinearity estimation model to obtain a nonlinearity estimation result output by the steering nonlinearity estimation model; wherein the steering nonlinearity estimation model is obtained by performing parameter identification on a pre-constructed original steering nonlinearity estimation model based on pre-obtained vehicle steering related test data and in combination with a gradient descent method. The application is based on vehicle steering related test data and utilizes the gradient descent method to perform parameter identification on the pre-constructed original steering nonlinearity estimation model to obtain a vehicle steering nonlinearity parameter directly representing a degree of vehicle steering nonlinearity, thereby preventing a situation of deterioration of automatic driving control effect caused by serious vehicle steering nonlinearity in advance, further improving control effect of an intelligent truck, improving control precision, and ensuring driving safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, in particular to a vehicle steering nonlinearity estimation method and device. BACKGROUND

[0002] The appearance of the automobile has changed the way people travel, greatly improving travel efficiency and making people's lifestyle more convenient. With the progress of science and the rapid development of technology, especially the rapid development of computers, automobiles have gradually become intelligent. Intelligent vehicles have broad market prospects and are constantly becoming new hotspots in the industry. As an important means of transportation on the logistics transportation line, intelligent trucks will solve the problems of fatigue driving, increasing labor costs and other problems in the process of manual driving, while also reducing the incidence of traffic accidents, and have great commercial value.

[0003] Generally, an intelligent vehicle is composed of three parts. The first part is perception, which collects data in real time through sensors such as laser radar, cameras, and millimeter wave radar to perceive the surrounding environment. The second part is decision planning, which decides the current driving behavior according to the perception result and plans the motion trajectory according to certain rules. The third part is control execution, which calculates the executable throttle, brake, steering, etc. signals according to the planned motion trajectory and issues corresponding driving operations.

[0004] In the control execution layer, lateral control is one of the important links, which needs to consider the planned motion trajectory in combination with the performance of the vehicle's steering system, and the corresponding steering control amount is calculated for control. The steering system of a truck is more complex, with stronger nonlinearity and coupling characteristics, which will lead to a decline in control effect and may even cause traffic accidents. SUMMARY

[0005] The present application provides a vehicle steering nonlinearity estimation method and device to solve the problem of poor automatic driving control effect caused by nonlinearity in the prior art, to quickly and accurately evaluate the severity of truck nonlinearity and improve the accuracy of automatic driving control.

[0006] The present application provides a vehicle steering nonlinearity estimation method, comprising: obtaining vehicle steering related data to be predicted; inputting the vehicle steering related data to be predicted into a steering nonlinearity estimation model to obtain a nonlinearity estimation result output by the steering nonlinearity estimation model; wherein the steering nonlinearity estimation model is obtained by parameter identification of a pre-constructed original steering nonlinearity estimation model based on pre-acquired vehicle steering related test data and in combination with a gradient descent method.

[0007] The application provides a vehicle steering nonlinearity estimation method, a steering nonlinearity estimation model is obtained by parameter identification of a previously constructed original steering nonlinearity estimation model based on previously obtained vehicle steering related test data and in combination with a gradient descent method, and the method comprises the following steps: obtaining vehicle steering related test data, wherein the vehicle steering related test data comprises a steering wheel steering angle and a vehicle yaw rate corresponding to the steering wheel steering angle; performing parameter identification on the previously constructed original steering nonlinearity estimation model based on the vehicle steering related test data by using the gradient descent method, obtaining identification parameters, and updating the original steering nonlinearity estimation model by using the identification parameters to obtain a plurality of initial steering nonlinearity estimation models; inputting the steering wheel steering angle in the vehicle steering related test data into the initial steering nonlinearity estimation model to obtain a model estimated yaw rate; obtaining an evaluation standard corresponding to each initial steering nonlinearity estimation model based on a preset standard evaluation rule according to the model estimated yaw rate and the vehicle yaw rate corresponding to the steering wheel steering angle; and selecting the initial steering nonlinearity estimation model according to the evaluation standard to obtain a steering nonlinearity estimation model.

[0008] The application provides a vehicle steering nonlinearity estimation method, and the parameter identification on the previously constructed original steering nonlinearity estimation model based on the vehicle steering related test data by using the gradient descent method comprises the following steps: classifying the vehicle steering related test data to obtain test data classification information, wherein the test data classification information comprises vehicle steering related test data corresponding to different categories; and obtaining identification parameters corresponding to vehicle steering related test data of each category by performing parameter identification on the previously constructed original steering nonlinearity estimation model based on the test data classification information by using the gradient descent method, wherein the identification parameters comprise a yaw rate response degree and a vehicle nonlinearity.

[0009] The application provides a vehicle steering nonlinearity estimation method, and the vehicle steering related test data further comprises a vehicle speed corresponding to the steering wheel steering angle, and the step of classifying the vehicle steering related test data to obtain test data classification information comprises the following steps: converting the vehicle steering related test data to a frequency domain based on Fourier transform; and finding an amplitude of the steering wheel steering angle converted to the frequency domain and an amplitude of the vehicle yaw rate corresponding to the same frequency, and classifying the test data according to the vehicle speed to obtain the test data classification information.

[0010] According to the vehicle steering nonlinearity estimation method provided by the application, after the test data classification information is obtained, the test data classification information is screened according to a preset screening rule, and the preset screening rule includes that the main frequency of the steering wheel steering angle converted into the frequency domain meets a first preset frequency range, the main frequency of the vehicle yaw rate meets a second preset frequency range, and the proportion of the amplitude of the main frequency in the amplitude of the entire frequency domain exceeds a preset threshold.

[0011] According to the vehicle steering nonlinearity estimation method provided by the application, the vehicle steering related test data is converted into the frequency domain based on the Fourier transform, and the vehicle steering related test data is segmented according to the signal input of the steering wheel steering angle.

[0012] According to the vehicle steering nonlinearity estimation method provided by the application, the vehicle steering related test data is converted into the frequency domain based on the Fourier transform, and the vehicle steering related test data is segmented according to the signal input of the steering wheel steering angle.

[0013] The application further provides a vehicle steering nonlinearity estimation device, which comprises a data acquisition module that acquires vehicle steering related data to be predicted, and a result estimation module that inputs the vehicle steering related data to be predicted into a steering nonlinearity estimation model to obtain a nonlinearity estimation result output by the steering nonlinearity estimation model.

[0014] The application further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the vehicle steering nonlinearity estimation method according to any one of the above-described methods when executing the program.

[0015] The application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is executable on a processor to implement the steps of the vehicle steering nonlinearity estimation method according to any one of the above-described methods.

[0016] The application further provides a computer program product comprising a computer program, wherein the computer program is executable on a processor to implement the steps of the vehicle steering nonlinearity estimation method according to any one of the above-described methods.

[0017] The vehicle steering nonlinearity estimation method and device provided by the application are based on vehicle steering related test data, and a gradient descent method is used to perform parameter identification on a pre-constructed original steering nonlinearity estimation model to obtain a vehicle steering nonlinearity parameter directly representing the degree of vehicle steering nonlinearity. According to the characteristic that the greater the vehicle steering nonlinearity parameter is, the more serious the vehicle steering nonlinearity is, the nonlinearity of the vehicle as a whole is taken as the object, so that the nonlinearity severity of the truck can be quickly and accurately evaluated, the situation that the automatic driving control effect is deteriorated due to the serious vehicle steering nonlinearity can be prevented in advance, the control effect of the intelligent truck is further improved, the control precision is improved, and the driving safety is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0019] Figure 1 is a flowchart of the vehicle steering nonlinearity estimation method provided by the application;

[0020] Figure 2 is a structural schematic diagram of the vehicle steering nonlinearity estimation device provided by the application;

[0021] Figure 3 is a structural schematic diagram of the electronic device provided by the application. DETAILED DESCRIPTION

[0022] In order to make the objects, technical solutions and advantages of the application clearer, the technical solutions in the application will be described clearly and completely in combination with the drawings in the application. Obviously, the described embodiments are some embodiments of the application, not all embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor belong to the protection scope of the application.

[0023] Figure 1 shows a flowchart of a vehicle steering nonlinearity estimation method provided by the application. The method comprises:

[0024] S11, obtaining vehicle steering related data to be predicted;

[0025] S12, input the vehicle steering related data to be predicted into a steering nonlinearity estimation model to obtain a nonlinearity estimation result output by the steering nonlinearity estimation model; wherein the steering nonlinearity estimation model is obtained by performing parameter identification on a pre-constructed original steering nonlinearity estimation model based on pre-acquired vehicle steering related test data and in combination with a gradient descent method.

[0026] It should be noted that before the vehicle steering related data to be predicted is input into the steering nonlinearity estimation model, the following is included: performing model parameter identification based on pre-acquired vehicle steering related test data and in combination with a gradient descent method to obtain the steering nonlinearity estimation model.

[0027] Specifically, performing parameter identification on a pre-constructed original steering nonlinearity estimation model based on pre-acquired vehicle steering related test data and in combination with a gradient descent method to obtain the steering nonlinearity estimation model includes:

[0028] Sa, acquiring vehicle steering related test data, the vehicle steering related test data including a steering wheel steering angle and a vehicle yaw rate corresponding to the steering wheel steering angle;

[0029] Sb, performing parameter identification on a pre-constructed original steering nonlinearity estimation model based on the vehicle steering related test data and in combination with a gradient descent method to obtain identification parameters, and updating the original steering nonlinearity estimation model with the identification parameters to obtain a plurality of initial steering nonlinearity estimation models;

[0030] Sc, inputting the steering wheel steering angle in the vehicle steering related test data into the initial steering nonlinearity estimation model to obtain a model estimated yaw rate;

[0031] Sd, obtaining an evaluation standard corresponding to each initial steering nonlinearity estimation model based on a preset standard evaluation rule according to the model estimated yaw rate and the vehicle yaw rate corresponding to the steering wheel steering angle;

[0032] Se, selecting an initial steering nonlinearity estimation model according to the evaluation standard to obtain the steering nonlinearity estimation model.

[0033] It should be noted that Sa-Sf in the present specification do not represent the order of obtaining the steering nonlinearity estimation model, and the steps of obtaining the steering nonlinearity estimation model are described in detail below.

[0034] Step Sa, acquiring vehicle steering related test data, the vehicle steering related test data including a steering wheel steering angle and a vehicle yaw rate corresponding to the steering wheel steering angle.

[0035] In this embodiment, the vehicle steering related test data further includes vehicle speed corresponding to the steering wheel steering angle, and a same steering wheel steering angle corresponds to multiple different vehicle speeds. The vehicle steering related test data is obtained by: obtaining the steering wheel steering angle corresponding to a preset amplitude range, a preset amplitude variation interval, a third preset frequency range and a preset frequency interval; and obtaining the vehicle yaw rate corresponding to the steering wheel steering angle according to the preset vehicle speed.

[0036] In an optional embodiment, the preset amplitude range can be [0.1 rad, 1.5 rad], the preset amplitude variation interval can be 0.1 rad, the third preset frequency range can be [0.1 Hz, 1 Hz], and the preset frequency interval can be 0.05 Hz. It should be noted that, when the steering wheel steering angle corresponding to the preset amplitude range, the preset amplitude variation interval, the third preset frequency range and the preset frequency interval is obtained, the signal form is a sine wave, and the number of times of inputting the steering wheel steering angle corresponding to each amplitude frequency point is not less than a preset number of times. In this embodiment, the preset number of times can be set according to actual design requirements or prior experience, for example, 5 times, which is not limited further herein.

[0037] In addition, the preset vehicle speed can be set according to a preset vehicle speed range and a preset speed interval, for example, the preset vehicle speed range is 0-100 km / h, and the speed interval is 5 km / h. The preset vehicle speed is adjusted according to the above setting, and the steering wheel steering angle corresponding to each vehicle speed is obtained according to the preset amplitude range, the preset amplitude variation interval, the third preset frequency range and the preset frequency interval.

[0038] For example, in the working condition that the preset vehicle speed is 30 km / h, the input amplitude of the steering wheel steering angle is 0.5 rad, and the frequency is 0.2 Hz, the vehicle is kept in a straight-line cruise at 30 km / h, the steering wheel is first turned to the left by 0.5 rad, and then turned to the right by 0.5 rad, the whole process is 5 s, to simulate the input of the steering wheel sine signal, and the process is repeated 5 times to obtain the corresponding steering wheel steering angle and the corresponding vehicle speed and vehicle yaw rate.

[0039] In step Sb, based on the vehicle steering related test data, the gradient descent method is used to perform parameter identification on the pre-constructed original steering nonlinear prediction model to obtain identification parameters, and the original steering nonlinear prediction model is updated by using the identification parameters to obtain multiple initial steering nonlinear prediction models.

[0040] In the embodiment, based on the vehicle steering related test data, the gradient descent method is used to identify the parameters of the pre-constructed original steering nonlinear estimation model, and the identification parameters are obtained, including: classifying the vehicle steering related test data to obtain test data classification information, and the test data classification information includes vehicle steering related test data corresponding to different categories; according to the test data classification information, the gradient descent method is used to identify the parameters of the pre-constructed original steering nonlinear estimation model, and the identification parameters corresponding to the vehicle steering related test data of each category are obtained, including the yaw rate response and the vehicle nonlinear degree.

[0041] Specifically, the vehicle steering related test data is classified to obtain test data classification information, including: based on Fourier transform, the vehicle steering related test data is converted to the frequency domain; according to the same frequency, the amplitude of the steering wheel steering angle and the amplitude of the vehicle yaw rate corresponding to the converted frequency domain are found, and classified according to the vehicle speed to obtain the test data classification information.

[0042] Further, based on the Fourier transform, the vehicle steering related test data is converted to the frequency domain, including: the vehicle steering related test data is segmented according to the signal input of the steering wheel steering angle; the segmented vehicle steering related test data is subjected to Fourier transform to convert the vehicle steering related test data to the frequency domain. It should be noted that when the vehicle steering related test data is segmented according to the signal input of the steering wheel steering angle, it is necessary to note that each test signal input is a segment, and the continuous sinusoidal signal input at the same frequency is classified into the same segment. In addition, the test data classification information can be represented in tabular or non-tabular form, such as a vehicle speed-steering wheel steering angle-yaw rate table.

[0043] In addition, before the gradient descent method is used to identify the parameters of the pre-constructed original steering nonlinear estimation model according to the test data classification information, it includes: according to the preset screening rule, the test data classification information is screened, and the preset screening rule includes that the main frequency of the steering wheel steering angle converted to the frequency domain conforms to the first preset frequency range, the main frequency of the vehicle yaw rate conforms to the second preset frequency range, and the amplitude of the main frequency accounts for more than a preset threshold in the entire frequency domain amplitude. It should be noted that the first preset frequency range can be set according to the actual setting requirement or prior experience of the steering wheel steering angle, and the second preset frequency range can be set according to the actual design requirement or prior experience of the vehicle yaw rate. The first preset frequency range can be the same as the second preset frequency range, such as both being within 1 Hz, which is not limited further here. In addition, the preset threshold can be set according to the actual design requirement or prior experience, such as 50%.

[0044] For example, assuming there is vehicle steering related test data, in which the vehicle speed is 40 km / h, the steering wheel steering angle input amplitude is 0.3 rad. The frequency is 0.5 Hz, and the corresponding frequency domain transformation is that the steering wheel steering angle signal main frequency is 0.57 Hz, the amplitude is 0.24 rad, the yaw rate main frequency is 0.54 Hz, and the amplitude is 0.013 rad, which meets the above preset screening rule and is reserved.

[0045] It should be noted that before the gradient descent method is used to identify the parameters of the pre-constructed original steering nonlinear estimation model based on the vehicle steering related test data, a steering nonlinear estimation model is also constructed. In one possible implementation, the original steering nonlinear estimation model is updated by identifying parameters to obtain a plurality of initial steering nonlinear estimation models, and the initial steering nonlinear estimation model is represented as:

[0046]

[0047] Wherein, ω represents the vehicle yaw rate, represents the vehicle steering wheel steering angle, v represents the vehicle longitudinal speed, r represents the vehicle steering wheel to wheel end transmission ratio, k1 represents the yaw rate response, k2 represents the vehicle nonlinearity, and L represents the vehicle wheelbase.

[0048] It should be noted that since the steering system of the truck is relatively complex and has strong nonlinearity, the gradient descent method is used to identify the parameters of the pre-constructed original steering nonlinear estimation model to obtain the vehicle steering nonlinearity parameter directly representing the nonlinearity degree of the truck steering. According to the characteristic that the larger the vehicle steering nonlinearity parameter is, the more serious the vehicle steering nonlinearity is, the nonlinearity of the truck as a whole is taken as the object, so that the nonlinearity severity of the truck can be quickly and accurately evaluated, the control effect of the intelligent truck is improved, the control precision is improved, and the safety of driving is ensured.

[0049] Step Sc, inputting the steering wheel steering angle in the vehicle steering related test data into the initial steering nonlinear estimation model to obtain the model estimated yaw rate.

[0050] Step Sd, according to the model estimated yaw rate and the vehicle yaw rate corresponding to the steering wheel steering angle, the evaluation standard corresponding to each initial steering nonlinear estimation model is obtained based on the preset standard evaluation rule.

[0051] In this embodiment, the preset standard evaluation rule is represented as:

[0052]

[0053] Wherein, J represents the evaluation standard, N represents the number of data groups participating in evaluation, N dω represents the number of data in a single data set model ω represents the model estimated yaw rate real ω represents the vehicle yaw rate corresponding to the steering wheel steering angle.

[0054] It should be noted that the evaluation criterion J represents the structure of parameter optimization, the smaller J is, the smaller the difference between the model estimated yaw rate and the vehicle yaw rate corresponding to the steering wheel steering angle corresponding to the identification parameter, the higher the accuracy of the model, and the higher the reliability of the vehicle nonlinear estimation structure.

[0055] Step Se, selecting an initial steering nonlinear estimation model according to the evaluation criterion, obtaining a steering nonlinear estimation model. By evaluating the criterion, the corresponding initial steering nonlinear estimation model is selected as the model for predicting the nonlinear estimation of the vehicle steering related data, which contains the vehicle nonlinear degree k2, and the larger k2 is, the more serious the steering nonlinearity is.

[0056] In summary, the embodiment of the application obtains the vehicle steering nonlinear degree parameter directly representing the degree of vehicle steering nonlinearity by performing parameter identification on the originally constructed steering nonlinear estimation model based on the vehicle steering related test data and using the gradient descent method, and according to the characteristics that the larger the vehicle steering nonlinear degree parameter is, the more serious the vehicle steering nonlinearity is, the vehicle nonlinear is taken as the object, so as to quickly and accurately evaluate the nonlinearity of the truck, to prevent the situation that the automatic driving control effect is deteriorated due to the serious vehicle steering nonlinearity, and to further improve the control effect of the intelligent truck, improve the control precision, and ensure the driving safety.

[0057] The vehicle steering nonlinear degree estimation device provided by the application is described below, and the vehicle steering nonlinear degree estimation device described below can be referred to each other corresponding to the vehicle steering nonlinear degree estimation method described above.

[0058] Figure 2 A structural schematic diagram of a vehicle steering nonlinear degree estimation device is shown, the device comprises:

[0059] The data acquisition module 21 acquires the vehicle steering related data to be predicted.

[0060] The result estimation module 22 inputs the vehicle steering related data to be predicted into the steering nonlinear estimation model to obtain the nonlinear estimation result output by the steering nonlinear estimation model; wherein the steering nonlinear estimation model is obtained by performing parameter identification on the originally constructed steering nonlinear estimation model based on the vehicle steering related test data obtained in advance and combining the gradient descent method.

[0061] In an optional embodiment, the device further comprises a model obtaining module, which, before the vehicle steering related data to be predicted is input into the steering non-linear estimation model, performs parameter identification on a pre-constructed original steering non-linear estimation model based on pre-obtained vehicle steering related test data and in combination with a gradient descent method to obtain the steering non-linear estimation model.

[0062] Specifically, the model obtaining module comprises: a data obtaining unit configured to obtain vehicle steering related test data, the vehicle steering related test data comprising a steering wheel steering angle and a vehicle yaw rate corresponding to the steering wheel steering angle; an initial model obtaining unit configured to perform parameter identification on a pre-constructed original steering non-linear estimation model based on the vehicle steering related test data and in combination with a gradient descent method to obtain identification parameters, and update the original steering non-linear estimation model with the identification parameters to obtain a plurality of initial steering non-linear estimation models; an estimation unit configured to input the steering wheel steering angle in the vehicle steering related test data into the initial steering non-linear estimation model to obtain a model estimated yaw rate; an evaluation unit configured to obtain an evaluation criterion corresponding to each initial steering non-linear estimation model based on a preset standard evaluation criterion according to the model estimated yaw rate and the vehicle yaw rate corresponding to the steering wheel steering angle; and a model determining unit configured to select the initial steering non-linear estimation model according to the evaluation criterion to obtain the steering non-linear estimation model.

[0063] In this embodiment, the data obtaining unit comprises: a first data obtaining sub-unit configured to obtain corresponding steering wheel steering angles according to a preset amplitude range, a preset amplitude variation interval, a third preset frequency range and a preset frequency interval; and a second data obtaining sub-unit configured to obtain corresponding vehicle yaw rates according to the steering wheel steering angles and at a preset vehicle speed.

[0064] The initial model obtaining unit comprises: a classification sub-unit configured to classify the vehicle steering related test data to obtain test data classification information, the test data classification information comprising vehicle steering related test data corresponding to different categories; and an identification sub-unit configured to perform parameter identification on a pre-constructed original steering non-linear estimation model based on the test data classification information and in combination with a gradient descent method to obtain identification parameters corresponding to vehicle steering related test data of each category, the identification parameters comprising a yaw rate response degree and a vehicle non-linearity.

[0065] More specifically, the classification sub-unit comprises: a frequency domain conversion grand sub-unit configured to convert the vehicle steering related test data to a frequency domain based on Fourier transform; and a classification grand sub-unit configured to find amplitudes of the steering wheel steering angles and amplitudes of the vehicle yaw rates corresponding to the steering wheel steering angles converted to the frequency domain according to a same frequency, and classify according to vehicle speed to obtain the test data classification information.

[0066] The frequency domain conversion grandson unit comprises: a segmentation great-grandson unit, which segments vehicle steering related test data according to a steering wheel steering angle signal input; and a Fourier transform great-grandson unit, which performs Fourier transform on the segmented vehicle steering related test data to convert the vehicle steering related test data to the frequency domain.

[0067] In addition, the initial model acquisition unit further comprises a screening subunit that screens the test data classification information according to a preset screening rule, and the preset screening rule comprises that a main frequency of the steering wheel steering angle converted to the frequency domain meets a first preset frequency range, a main frequency of the vehicle yaw rate meets a second preset frequency range, and an amplitude of the main frequency accounts for more than a preset threshold in an overall frequency domain amplitude.

[0068] In an optional embodiment, the model acquisition module further comprises a model construction unit that constructs an original steering nonlinear estimation model before performing parameter identification on the original steering nonlinear estimation model constructed in advance by using a gradient descent method based on the vehicle steering related test data.

[0069] In summary, the embodiment of the present application obtains a vehicle steering nonlinear degree parameter directly representing a vehicle steering nonlinear degree by using a gradient descent method to perform parameter identification on an original steering nonlinear estimation model constructed in advance based on vehicle steering related test data through the result estimation module, and according to the characteristics that the greater the vehicle steering nonlinear degree parameter is, the more serious the vehicle steering nonlinearity is, takes the nonlinearity of the vehicle as a whole as an object, thereby quickly and accurately evaluating the nonlinearity severity of the truck, preventing the situation that the automatic driving control effect is deteriorated due to the serious vehicle steering nonlinearity in advance, further improving the control effect of the intelligent truck, improving the control precision, and ensuring driving safety.

[0070] Figure 3 An example of an electronic device is shown in the physical structure diagram of the electronic device as shown in Figure 3 As shown, the electronic device can include a processor 31, a communications interface 32, a memory 33, and a communications bus 34, wherein the processor 31, the communications interface 32, and the memory 33 complete mutual communication through the communications bus 34. The processor 31 can invoke a logical instruction in the memory 33 to execute a vehicle steering nonlinear degree estimation method, which comprises: acquiring vehicle steering related data to be predicted; inputting the vehicle steering related data to be predicted into a steering nonlinear estimation model to obtain a nonlinear degree estimation result output by the steering nonlinear estimation model; wherein the steering nonlinear estimation model is obtained by performing parameter identification on an original steering nonlinear estimation model constructed in advance based on pre-acquired vehicle steering related test data and in combination with a gradient descent method.

[0071] In addition, the logic instructions in the memory 33 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions 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 making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, 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 disk, and various media that can store program codes.

[0072] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the vehicle steering nonlinearity estimation method provided by the above-mentioned methods. The method comprises: obtaining vehicle steering related data to be predicted; inputting the vehicle steering related data to be predicted into a steering nonlinearity estimation model to obtain a nonlinearity estimation result output by the steering nonlinearity estimation model; wherein the steering nonlinearity estimation model is obtained by parameter identification of a pre-constructed original steering nonlinearity estimation model based on pre-acquired vehicle steering related test data and in combination with a gradient descent method.

[0073] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the vehicle steering nonlinearity estimation method provided by the above-mentioned methods. The method comprises: obtaining vehicle steering related data to be predicted; inputting the vehicle steering related data to be predicted into a steering nonlinearity estimation model to obtain a nonlinearity estimation result output by the steering nonlinearity estimation model; wherein the steering nonlinearity estimation model is obtained by parameter identification of a pre-constructed original steering nonlinearity estimation model based on pre-acquired vehicle steering related test data and in combination with a gradient descent method.

[0074] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0075] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0076] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part 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 application.

Claims

1. A method for estimating the nonlinearity of vehicle steering, characterized in that, include: Obtain relevant data on the vehicle's steering to be predicted; The steering-related data of the vehicle to be predicted is input into the steering nonlinear prediction model to obtain the nonlinearity prediction result output by the steering nonlinear prediction model; wherein, the steering nonlinear prediction model is obtained based on the pre-acquired vehicle steering-related test data and combined with the gradient descent method to identify the parameters of the pre-constructed original steering nonlinear prediction model. Before inputting the steering-related data of the vehicle to be predicted into the steering nonlinear prediction model, the following steps are included: Based on pre-acquired vehicle steering-related test data, and combined with gradient descent method for model parameter identification, a steering nonlinear prediction model is obtained. Based on pre-acquired vehicle steering-related test data and using the gradient descent method to identify model parameters, a steering nonlinear prediction model is obtained, including: Acquire vehicle steering-related test data, including steering wheel angle and the corresponding vehicle yaw rate; Based on the vehicle steering-related test data, the gradient descent method is used to identify the parameters of the pre-built original steering nonlinear prediction model to obtain the identified parameters. The identified parameters are then used to update the original steering nonlinear prediction model to obtain multiple initial steering nonlinear prediction models. The steering wheel angle from the vehicle steering-related test data is input into the initial steering nonlinear prediction model to obtain the model-predicted yaw rate. Based on the yaw rate predicted by the model and the vehicle yaw rate corresponding to the steering wheel angle, the evaluation criteria for each of the initial steering nonlinear prediction models are obtained based on the preset standard evaluation rules. Based on the evaluation criteria, the initial steering nonlinear prediction model is selected to obtain the steering nonlinear prediction model; The initial steering nonlinear prediction model is expressed as follows: Where ω represents the vehicle's yaw rate, φ represents the vehicle's steering wheel angle, v represents the vehicle's longitudinal speed, r represents the vehicle's steering wheel-to-wheel transmission ratio, k1 represents the yaw rate response, k2 represents the vehicle's nonlinearity, and L represents the vehicle's wheelbase.

2. The vehicle steering nonlinearity estimation method according to claim 1, characterized in that, Based on the vehicle steering-related test data, the gradient descent method is used to identify the parameters of the pre-built original steering nonlinear prediction model, resulting in identified parameters, including: The vehicle steering-related test data is classified to obtain test data classification information, which includes vehicle steering-related test data corresponding to different categories. Based on the test data classification information, the gradient descent method is used to identify the parameters of the pre-constructed original steering nonlinearity prediction model, and the identification parameters of the steering-related test data of each category of vehicle are obtained. The identification parameters include yaw rate response and vehicle nonlinearity.

3. The vehicle steering nonlinearity estimation method according to claim 2, characterized in that, The vehicle steering-related test data also includes the vehicle speed corresponding to the steering wheel angle. The classification of the vehicle steering-related test data to obtain test data classification information includes: Based on Fourier transform, the vehicle steering-related test data are converted to the frequency domain; Based on the same frequency, find the amplitude of the steering wheel angle and the amplitude of the vehicle yaw rate converted to the frequency domain, and classify them according to vehicle speed to obtain test data classification information.

4. The vehicle steering nonlinearity estimation method according to claim 3, characterized in that, After obtaining the test data classification information, the following is included: According to preset filtering rules, the test data classification information is filtered. The preset filtering rules include that the main frequency of the steering wheel angle after conversion to the frequency domain conforms to a first preset frequency range, the main frequency of the vehicle yaw rate conforms to a second preset frequency range, and the amplitude of the main frequency accounts for more than a preset threshold in the entire frequency domain amplitude.

5. The vehicle steering nonlinearity estimation method according to claim 3, characterized in that, The step of converting the vehicle steering-related test data to the frequency domain based on Fourier transform includes: The vehicle steering-related test data is segmented according to the steering wheel angle signal input. The segmented vehicle steering-related test data is subjected to Fourier transform to convert the vehicle steering-related test data to the frequency domain.

6. The vehicle steering nonlinearity estimation method according to claim 1, characterized in that, The acquisition of vehicle steering-related test data includes: The corresponding steering wheel angle is obtained based on the preset amplitude range, preset amplitude change interval, third preset frequency range, and preset frequency interval; Based on the steering wheel angle and the preset vehicle speed, the corresponding vehicle yaw rate is obtained.

7. A vehicle steering nonlinearity estimation device, characterized in that, include: The data acquisition module acquires relevant data on the vehicle's steering to be predicted. The result prediction module inputs the vehicle steering-related data to be predicted into the steering nonlinear prediction model to obtain the nonlinearity prediction result output by the steering nonlinear prediction model; wherein, the steering nonlinear prediction model is obtained based on the pre-acquired vehicle steering-related test data and combined with the gradient descent method to identify the parameters of the pre-constructed original steering nonlinear prediction model. The device further includes: Before inputting the vehicle steering-related data to be predicted into the steering nonlinear prediction model, the model acquisition module identifies the model parameters based on the pre-acquired vehicle steering-related test data and combines the gradient descent method to obtain the steering nonlinear prediction model. The model acquisition module includes: The data acquisition unit acquires vehicle steering-related test data, including the steering wheel angle and the corresponding vehicle yaw rate. The initial model acquisition unit, based on the vehicle steering-related test data, uses the gradient descent method to identify the parameters of the pre-built original steering nonlinear prediction model, obtains the identified parameters, and uses the identified parameters to update the original steering nonlinear prediction model, thereby obtaining multiple initial steering nonlinear prediction models. The prediction unit inputs the steering wheel angle from the vehicle steering-related test data into the initial steering nonlinear prediction model to obtain the model-predicted yaw rate. The evaluation unit, based on the yaw rate predicted by the model and the vehicle yaw rate corresponding to the steering wheel angle, obtains the evaluation criteria for each of the initial steering nonlinear prediction models according to the preset standard evaluation rules. The model determination unit selects the initial steering nonlinear prediction model according to the evaluation criteria to obtain the steering nonlinear prediction model; The initial steering nonlinear prediction model is expressed as follows: Where ω represents the vehicle's yaw rate. The steering wheel angle is represented by v, the longitudinal speed of the vehicle is represented by r, the transmission ratio from the steering wheel to the wheel end is represented by k1, the yaw rate response is represented by k2, the nonlinearity of the vehicle is represented by L, and the wheelbase of the vehicle is represented by L.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the vehicle steering nonlinearity estimation method as described in any one of claims 1 to 6.

9. 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, it implements the steps of the vehicle steering nonlinearity estimation method as described in any one of claims 1 to 6.

Images