End-to-end tire performance margin identification model modeling, usage method and related equipment

Abstract

The embodiments of the present disclosure provide end-to-end tire performance margin identification model modeling, use methods and devices, computer-readable storage media and electronic devices, and belong to the field of vehicle autonomous driving technology. The modeling method includes obtaining vehicle test data; obtaining the performance margin corresponding to the tire of the vehicle according to the vehicle test data; labeling the vehicle test data with tire performance margin; segmenting the labeled vehicle test data; training the segmented and labeled vehicle test data through an artificial intelligence method to complete the modeling of the end-to-end tire performance margin identification model. The modeling method of the present disclosure can achieve the modeling of the end-to-end tire performance margin identification model.

Description

Technical Field

[0001] The present disclosure relates to the technical field of vehicle autonomous driving, and in particular to end-to-end tire performance margin identification modeling, use methods and devices, computer-readable storage media, and electronic devices. Background Art

[0002] With the development of the automobile industry, the level of vehicle intelligence has gradually improved, and more and more advanced driver assistance systems and active safety systems are installed on vehicles to assist drivers in controlling vehicles and improve vehicle stability and safety. However, as the main connection point between the vehicle and the environment, the working state of the tire will directly affect the changes in vehicle dynamics. As an indicator of the working state of the tire, the tire performance margin directly affects the control efficiency of the vehicle controller.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0004] The embodiments of the present disclosure provide end-to-end tire performance margin identification model modeling, use method and device, computer-readable storage medium and electronic device, which can realize the modeling of the tire performance margin identification model.

[0005] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0006] According to one aspect of the present disclosure, a method for modeling an end-to-end tire performance margin identification model is provided, comprising: acquiring vehicle test data; acquiring performance margins corresponding to tires of the vehicle based on the vehicle test data; labeling the vehicle test data with tire performance margins; segmenting the labeled vehicle test data; and training the segmented and labeled vehicle test data through an artificial intelligence method to complete modeling of an end-to-end tire performance margin identification model.

[0007] In one embodiment, the vehicle experimental data includes one or more of yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle; obtaining the vehicle experimental data includes: obtaining one or more of yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques through sensors of the vehicle; obtaining one or more of expected yaw rate and expected center of mass sideslip angle through a single-track vehicle model.

[0008] In one embodiment, obtaining a performance margin corresponding to a tire of the vehicle according to the vehicle test data includes: obtaining a performance margin in a linear region, a transition region, a saturation region or a slip region corresponding to the tire of the vehicle according to the vehicle test data.

[0009] In one embodiment, segmenting the labeled vehicle test data includes: segmenting the labeled vehicle test data based on a sliding time window.

[0010] In one embodiment, segmenting the labeled vehicle test data based on a sliding time window includes: taking N as a segmentation step, where N is a natural number greater than or equal to 2; supplementing N-1 all-zero data before the labeled vehicle test data; segmenting the vehicle test data based on a sliding time window in units of N; wherein the number of segments into which the labeled vehicle test data is divided based on the sliding time window is equal to the number of vehicle test data.

[0011] In one embodiment, marking the tire performance margin of the vehicle test data includes: taking the performance margin corresponding to the last data or a group of data in each section of the vehicle test data as the performance margin of the section of the vehicle test data.

[0012] In one embodiment, the segmented and labeled vehicle test data is trained through an artificial intelligence method to complete end-to-end tire performance margin identification model building, including: the segmented and labeled vehicle test data is trained through a long short-term memory network model LSTM to complete end-to-end tire performance margin identification model building.

[0013] According to one aspect of the present disclosure, a method for using an end-to-end tire performance margin identification model is provided, characterized in that it includes: acquiring vehicle data; inputting the vehicle data into a trained end-to-end tire performance margin identification model to acquire tire performance margin; wherein the end-to-end tire performance margin identification model is a model obtained by any of the methods in the above embodiments.

[0014] According to one aspect of the present disclosure, there is provided an end-to-end tire performance margin identification model building device, comprising: a first acquisition module, configured to acquire vehicle test data; a second acquisition module, configured to acquire the performance margin corresponding to the tire of the vehicle according to the vehicle test data; a labeling module, configured to label the vehicle test data with tire performance margin; a segmentation module, configured to segment the labeled vehicle test data; and a training module, configured to train the segmented and labeled vehicle test data through an artificial intelligence method to complete the end-to-end tire performance margin identification model building.

[0015] According to one aspect of the present disclosure, an end-to-end tire performance margin identification device is provided, comprising: a third acquisition module, configured to acquire vehicle data; and an identification module, configured to input the vehicle data into a trained end-to-end tire performance margin identification model to acquire the tire performance margin; wherein the end-to-end tire performance margin identification model is a model acquired by any of the methods described in the above embodiments.

[0016] According to one aspect of the present disclosure, an electronic device is provided, comprising: one or more processors; and a storage device configured to store one or more programs, so that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of the above embodiments.

[0017] According to one aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0018] The end-to-end tire performance margin identification modeling method of the present application obtains vehicle test data; obtains the performance margin corresponding to the tire of the vehicle based on the vehicle test data; labels the vehicle test data with tire performance margin; segments the labeled vehicle test data; and trains the segmented and labeled vehicle test data through an artificial intelligence method to complete the end-to-end tire performance margin identification modeling, thereby realizing the end-to-end tire performance margin identification modeling.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 A schematic diagram showing an exemplary system architecture to which the end-to-end tire performance margin identification modeling method according to an embodiment of the present disclosure can be applied;

[0022] Figure 2 is a flow chart of a method for modeling an end-to-end tire performance margin identification model provided by an embodiment of the present disclosure;

[0023] Figure 3 is a flow chart of a method for obtaining vehicle test data provided by an embodiment of the present disclosure;

[0024] Figure 4 is a schematic diagram of a single-rail vehicle model provided by an embodiment of the present disclosure;

[0025] Figure 5 It is a schematic diagram of obtaining an expected yaw rate, an expected center-of-mass sideslip angle, an actual yaw rate and an actual center-of-mass sideslip angle of a single-track vehicle model provided by an embodiment of the present disclosure;

[0026] Figure 6 It is a flow chart of a method for segmenting the labeled vehicle experimental data based on a sliding time window provided by an embodiment of the present disclosure;

[0027] Figure 7 is a schematic diagram of a method for segmenting the labeled vehicle experimental data based on a sliding time window provided by an embodiment of the present disclosure;

[0028] Figure 8 is a schematic diagram of vehicle test data after data classification is completed provided by an embodiment of the present disclosure;

[0029] Fig. 9 It is a schematic diagram of the main network architecture of end-to-end training based on deep learning provided by an embodiment of the present disclosure;

[0030] Fig.10 This is a schematic diagram of the misclassification of the left front wheel test set;

[0031] Fig.11 The three-dimensional schematic diagram of the relationship between the sideslip angle and lateral force under different road friction coefficients for the tire performance margin identification results of the left front wheel test set;

[0032] Fig.12 The three-dimensional schematic diagram of the relationship between slip rate and longitudinal force under different road friction coefficients for the tire performance margin identification results of the left front wheel test set;

[0033] Fig.13 This is a schematic diagram of the relationship between the normalized longitudinal force and the normalized lateral force of the tire performance margin identification result of the left front wheel test set under a road friction coefficient of 0.3;

[0034] Fig.14 This is a schematic diagram of the relationship between the normalized longitudinal force and the normalized lateral force of the tire performance margin identification result of the left front wheel test set under a road friction coefficient of 0.6;

[0035] Fig.15 It is a schematic diagram of the relationship between the normalized longitudinal force and the normalized lateral force of the tire performance margin identification result of the left front wheel test set under the road friction coefficient of 1.0;

[0036] Fig.16 The three-dimensional schematic diagram of the relationship between the total slip rate and normalized force under different road friction coefficients for the tire performance margin identification results of the four wheel test sets;

[0037] Fig.17 To verify the changing rules of the features at the center of mass for joint simulation;

[0038] Fig.18 This is the identification result of the joint simulation verification of the four wheels when the road adhesion coefficient is 1.0;

[0039] Fig.19 The identification results of the four-wheel joint simulation verification when the road adhesion coefficient is 0.6;

[0040] Fig. 20 This is the identification result of the joint simulation verification of the four wheels when the road adhesion coefficient is 0.3;

[0041] Fig.21 is a flow chart of a method for using an end-to-end tire performance margin identification model provided by an embodiment of the present disclosure;

[0042] Fig. 22 It is a structural schematic diagram of an end-to-end tire performance margin identification model building device provided by an embodiment of the present disclosure;

[0043] Fig.23 is a structural schematic diagram of an end-to-end tire performance margin identification device provided by an embodiment of the present disclosure;

[0044] Fig.24 It is a structural diagram of a computer device 2400 provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0045] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more comprehensive and complete and will fully convey the concept of the example embodiments to those skilled in the art.

[0046] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0047] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0048] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0049] In the disclosed embodiments, vehicle test data can be acquired based on technologies such as a single-track vehicle model and a long short-term memory network model; a performance margin corresponding to the tire of the vehicle can be acquired based on the vehicle test data; the vehicle test data is labeled with tire performance margins; the labeled vehicle test data is segmented; the segmented and labeled vehicle test data is trained using an artificial intelligence method to complete end-to-end tire performance margin identification model building.

[0050] First, some terms of the present disclosure are explained below:

[0051] The single-track vehicle model, also known as the two-degree-of-freedom dynamic model, is usually used to characterize the state of the vehicle when it is stable. The output of the model, the expected yaw rate and the expected sideslip angle of the center of mass, reflects the dynamic changes of the vehicle in steady state. As the nonlinearity of the actual vehicle increases, the comparison between the expected value and the actual value can intuitively reflect the changes in the nonlinear characteristics of the vehicle, which is conducive to the extraction of features by the deep learning network.

[0052] The long short-term memory network model (LSTM) can learn information in time series data segments and has good learning ability for long time series data.

[0053] Figure 1 A schematic diagram of an exemplary system architecture 100 is shown to which the end-to-end tire performance margin identification modeling method according to an embodiment of the present disclosure can be applied.

[0054] like Figure 1 As shown, the system architecture 100 may include one or more of terminals 101, 102, 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the terminals 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0055] It should be understood that Figure 1 The number of terminals, networks and servers in the embodiment is only for illustration. Any number of terminals, networks and servers may be provided as required. For example, the server 105 may be a server cluster composed of multiple servers.

[0056] Terminals 101, 102, 103 interact with server 105 via network 104, and can receive or send messages, etc. Terminals 101, 102, 103 can be various electronic devices with display screens, including but not limited to smart phones, tablet computers, portable computers, desktop computers, and the like.

[0057] The server 105 may be a server that provides various services. For example, a staff member uses the terminal 103 (or the terminal 101 or 102) to send a modeling request for the identification model of the tire performance margin to the server 105. The server 105 may obtain vehicle test data; obtain the performance margin corresponding to the tire of the vehicle according to the vehicle test data; annotate the vehicle test data with the tire performance margin; segment the annotated vehicle test data; and train the segmented and annotated vehicle test data through an artificial intelligence method to complete the end-to-end tire performance margin identification modeling. The server 105 may display the trained tire performance margin identification model on the terminal 103, and then the staff member may view the tire performance margin identification model based on the content displayed on the terminal 103.

[0058] The terminal may be a mobile phone (such as terminal 101) or a tablet computer (such as terminal 102), or a desktop computer (such as terminal 101), etc., and is not limited here. The terminal may display an application, which may be an application for modeling an end-to-end tire performance margin identification model. Figure 1 The terminals in the example are only some of the devices. In the present disclosure, the terminals are not limited to the Figure 1 The equipment listed in .

[0059] It is understandable that the terminal mentioned in the embodiments of the present disclosure may be a user device, and the server in the embodiments of the present disclosure includes but is not limited to a server or a cluster of servers. Among them, the terminal mentioned above may be an electronic device, including but not limited to a mobile phone, a tablet computer, an intelligent voice interaction device, a smart home appliance, a vehicle terminal, a desktop computer, a laptop computer, a PDA, a vehicle device, an augmented reality / virtual reality (AR / VR) device, a helmet display, a smart TV, a wearable device, a smart speaker, a digital camera, a camera, and other mobile Internet devices (mobile internet device, MID) with network access capability, or terminal devices in scenarios such as trains, ships, and flights.

[0060] Among them, the servers mentioned above can be cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, vehicle-road collaboration, content delivery networks (Content Delivery Network, CDN), as well as big data and artificial intelligence platforms. They can also be independent physical servers or server clusters or distributed systems composed of multiple physical servers.

[0061] Optionally, the data involved in the embodiments of the present disclosure may be stored in a cloud platform, or the data may be stored based on cloud storage technology or blockchain technology, which is not limited here.

[0062] The tire status cannot be directly obtained through sensors. Usually, it is necessary to estimate the longitudinal speed, tire slip rate and tire force information first and then make further estimates based on this information. In addition, the tire performance margin information is also directly affected by the road friction coefficient. The tire performance margin can be estimated by estimating the road friction coefficient, but the road friction coefficient cannot be directly obtained either.

[0063] Figure 2 is a flow chart of a method for modeling an end-to-end tire performance margin identification model provided by an embodiment of the present disclosure. The method provided by an embodiment of the present disclosure can be Figure 1 The terminal or the server in the embodiment is executed, or the terminal and the server are interactively executed.

[0064] like Figure 2 As shown, the method provided by the embodiment of the present disclosure may include the following steps.

[0065] In step S210, vehicle test data is obtained.

[0066] In this step, the terminal or server obtains vehicle test data.

[0067] The vehicle test data includes one or more of yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle. The four wheel speeds and four wheel torques are data at the tire. The vehicle test data can be obtained, for example, through experiments.

[0068] In step S220, a performance margin corresponding to a tire of the vehicle is acquired according to the vehicle test data.

[0069] In this step, the terminal or the server obtains the performance margin corresponding to the tire of the vehicle according to the vehicle test data.

[0070] In one embodiment, the performance margin of the linear region, transition region, saturation region or slip region corresponding to the tire of the vehicle is obtained according to the vehicle test data. For example, a set of data at a certain moment, such as the performance margin of the linear region, transition region, saturation region or slip region corresponding to the tire corresponding to one or more of the yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle.

[0071] In step S230, the vehicle test data is annotated with tire performance margin.

[0072] In this step, the terminal or server marks the tire performance margin of the vehicle test data.

[0073] In step S240, the labeled vehicle test data is segmented.

[0074] In this step, the terminal or the server segments the labeled vehicle test data.

[0075] In one embodiment, the labeled vehicle experimental data is segmented based on a sliding time window.

[0076] In step S250, the segmented and labeled vehicle test data is trained by an artificial intelligence method to complete the modeling of an end-to-end tire performance margin identification model.

[0077] In this step, the terminal or server trains the segmented and labeled vehicle test data through an artificial intelligence method to complete the end-to-end tire performance margin identification model building.

[0078] In one embodiment, the vehicle test data that has been marked and segmented is trained through a long short-term memory network model LSTM to complete the end-to-end tire performance margin identification model building.

[0079] Figure 2 The invention discloses a method for modeling an end-to-end tire performance margin identification model, which includes acquiring vehicle test data; acquiring the performance margin corresponding to the tire of the vehicle according to the vehicle test data; labeling the vehicle test data with tire performance margin; segmenting the labeled vehicle test data; and training the segmented and labeled vehicle test data through an artificial intelligence method to complete the modeling of the end-to-end tire performance margin identification model. The method can complete the modeling of the end-to-end tire performance margin identification model for identifying the tire performance margin.

[0080] Figure 3 is a flow chart of a method for obtaining vehicle test data provided by an embodiment of the present disclosure. The method provided by an embodiment of the present disclosure can be Figure 1 The terminal or server in the embodiment is executed, or the terminal and the server are interactively executed. Wherein, the vehicle test data includes one or more of yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle.

[0081] like Figure 3 As shown, the method provided by the embodiment of the present disclosure may include the following steps.

[0082] In step S310, one or more of the yaw angular velocity, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, and four wheel torques are acquired through sensors of the vehicle.

[0083] In this step, the terminal or the server obtains one or more of the yaw angular velocity, the steering wheel angle, the longitudinal acceleration, the lateral acceleration, the wheel speeds of the four wheels, and the torques of the four wheels through the sensors of the vehicle.

[0084] Among them, yaw angular velocity, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, and four wheel torques are data that can be directly obtained through vehicle sensors.

[0085] In step S320, one or more of the expected yaw rate and the expected sideslip angle of the center of mass are obtained through the single-track vehicle model.

[0086] In this step, the terminal or the server obtains one or more of the expected yaw rate and the expected sideslip angle of the center of mass through the single-track vehicle model.

[0087] The output of the single-track vehicle model is the expected yaw rate and the expected sideslip angle of the center of mass, which reflect the dynamic changes of the vehicle in steady state. As the nonlinearity of the actual vehicle increases, the comparison between the expected yaw rate and the expected sideslip angle of the center of mass and the actual values ​​can intuitively reflect the changes in the nonlinear characteristics of the vehicle, which is conducive to the deep learning network to extract features. There is an uncertain time lag between the expected yaw rate and the expected sideslip angle of the center of mass and the actual value. It is difficult for the network model to learn the existing uncertain time lag with previous data. The time lag is mainly the hysteresis between the expected yaw rate and the expected sideslip angle of the center of mass and the actual value.

[0088] Figure 4 It is a schematic diagram of a monorail vehicle model provided by an embodiment of the present disclosure.

[0089] refer to Figure 4 , is the single-track vehicle model used based on the single-track vehicle model and the signal of the on-board sensor. The lateral and yaw motions of the model can be expressed as the following formulas (1) and (2):

[0090]

[0091] Where CoG is the center of mass, β is the sideslip angle, γ is the yaw rate, the points on β and γ represent the derivative, m is the sprung mass, V x is the longitudinal speed, F yf, F yr are the front axle lateral force and the rear axle lateral force, F yf =K αf α f , F yr =K αr α r ,ll f is the distance from the center of mass to the front axle, l r is the distance from the center of mass to the rear axle, I z is the moment of inertia at the center of mass; single-track vehicle models usually assume that the vehicle speed is constant and assume that sinδ≈0 and cosδ≈1 based on small angles. The tire slip angle is defined as follows:

[0092]

[0093]

[0094] α f is the front wheel slip angle, α r is the rear wheel slip angle, δ is the wheel turning angle; therefore, the state space equation can be expressed as follows:

[0095]

[0096] Among them, a11 =-2(K αf +K αr / mV x ), a 12 =2(l r K αr +l f K αf / mV x 2 )-1, a 21 =2(l r K αr +l f K αf / I z ), a 22 =-2(l r 2 K αf +l r 2 K αr / I z V x ), d1=2K αf / mV x , d2=2l f K αf / I z , K αf and K αr are the steering stiffness of the front and rear tires respectively. The expected yaw rate and the expected sideslip angle of the center of mass can be obtained through equation (5).

[0097] Figure 5 It is a schematic diagram of obtaining an expected yaw rate, an expected center-of-mass sideslip angle, an actual yaw rate and an actual center-of-mass sideslip angle of a single-track vehicle model provided by an embodiment of the present disclosure.

[0098] refer to Figure 5 In the first row, the dotted line represents the expected yaw rate, and the solid line represents the actual yaw rate; in the second row, the dotted line represents the expected center of mass slip angle, and the solid line represents the actual center of mass slip angle; the third row represents the tire performance margin, 1 represents the linear region, 2 represents the transition region, 3 represents the saturation region, and 4 represents the slip region. Figure 5 In the equation, there is a certain lag between the actual yaw rate and the expected yaw rate; there is a certain lag between the actual center of mass sideslip angle and the expected center of mass sideslip angle.

[0099] Figure 6 is a flow chart of a method for segmenting the labeled vehicle experimental data based on a sliding time window provided by an embodiment of the present disclosure. The method provided by an embodiment of the present disclosure can be performed by Figure 1 The terminal or the server in the embodiment is executed, or the terminal and the server are interactively executed.

[0100] like Figure 6 As shown, the method provided by the embodiment of the present disclosure may include the following steps.

[0101] In step S610, N is used as the segmentation step length, where N is a natural number greater than or equal to 2.

[0102] In this step, the terminal or the server uses N as the segmentation step length, where N is a natural number greater than or equal to 2.

[0103] In step S220, N-1 all-zero data are added to the front of the vehicle test data.

[0104] In this step, the terminal or server adds N-1 all-zero data before the vehicle test data. For example, when the vehicle test data is 1200 or groups of test data collected in 12 seconds, N is taken as 100, and 99 or groups of all-zero data are added before the 1200 or groups of test data.

[0105] In step S430, the vehicle test data is segmented based on the sliding time window in units of N, wherein the number of segments into which the vehicle test data is segmented based on the sliding time window is equal to the number of the vehicle test data. In this step, the terminal or server segments the vehicle test data based on the sliding time window in units of N.

[0106] The number of segments into which the vehicle test data is divided based on the sliding time window is equal to the number of the vehicle test data.

[0107] For example, when the vehicle test data is 1200 or groups of test data collected in 12 seconds, N is taken as 100, and 99 or groups of all-zero data are added before the 1200 or groups of test data. Then the vehicle test data is segmented based on the sliding time window in units of N, and the first segment of data is 99 or groups of all-zero data added before the 1200 or groups of test data plus the first or group of data of the 1200 or groups of test data; the second segment of data is 98 or groups of all-zero data added before the 1200 or groups of test data plus the first or group of data and the second or group of data of the 1200 or groups of test data; and so on, the 1200 or groups of test data are divided into 1200 data segments, and the last or group of data of each data segment is different from each other, so that each or group of data can be trained.

[0108] Figure 7 It is a schematic diagram of a method for segmenting the labeled vehicle experimental data based on a sliding time window provided by an embodiment of the present disclosure.

[0109] In one embodiment, the performance margin corresponding to the last data or a group of data in each segment of the vehicle test data is used as the performance margin of the segment of the vehicle test data.

[0110] Figure 8 It is a schematic diagram of vehicle test data after data classification provided by an embodiment of the present disclosure.

[0111] Reference Figure 8 , the data are classified into four regions: linear region (inverted triangle), transition region (×), saturation region (diamond) and slip region (circle) to label the data.

[0112] Fig. 9 It is a schematic diagram of the main network architecture of end-to-end training based on deep learning provided in an embodiment of the present disclosure.

[0113] See attached Fig. 9 , which is the main network architecture of the end-to-end training module based on deep learning, respectively, the sequence input layer, which is mainly responsible for the input of the data set. In this disclosure, the input features are different signals of the on-board sensor and the expected value of the monorail vehicle model; the LSTM layer is the long short-term memory network layer, which can learn the information in the time series data segment and has a good learning ability for long time series data. See the attached Figure 5 There is a time lag between the expected value and the actual value. This interval changes with the vehicle state and is difficult to express through a formula. The present disclosure learns this time lag by using the learning ability of the LSTM algorithm for long time series data; the batch normalization layer mainly standardizes the small batch data observations of each channel, which can speed up the training speed of the convolutional neural network and reduce the sensitivity of the training to initialization; the fully connected layer acts as a classifier, integrating the learned distributed features together for output; the sigmoid layer is a continuation of the fully connected layer, converting the output of the fully connected layer into a value between 0 and 1, and ensuring that the sum is 1; the classification layer outputs the final category by calculating the differential entropy loss.

[0114] Fig.10 This is a schematic diagram of the misclassification of the left front wheel test set.

[0115] The dataset of the end-to-end training module based on deep learning is divided into training set, validation set and test set at 70%, 15% and 15% respectively. The training set is used for network training to establish an end-to-end tire performance margin identification model. The validation set is used for verification during the training process to prevent overfitting. The test set is used to test the model obtained after training, and to test the model accuracy, precision, recall rate and F1-score coefficients. Fig.10 , which is the misclassification of the left front wheel in the test set. The misclassification in the four areas is less than 6%, and the model has a high accuracy.

[0116] Fig.11 The three-dimensional schematic diagram of the relationship between the sideslip angle and lateral force under different road friction coefficients is the result of the tire performance margin identification of the left front wheel test set.

[0117] Fig.12 The three-dimensional schematic diagram of the relationship between slip rate and longitudinal force under different road friction coefficients for the left front wheel test set tire performance margin identification results.

[0118] Fig.13 Schematic diagram of the relationship between the normalized longitudinal force and the normalized lateral force of the tire performance margin identification results of the left front wheel test set under a road friction coefficient of 0.3.

[0119] Fig.14 Schematic diagram of the relationship between the normalized longitudinal force and the normalized lateral force of the tire performance margin identification results of the left front wheel test set under a road friction coefficient of 0.6.

[0120] Fig.15 Schematic diagram of the relationship between the normalized longitudinal force and the normalized lateral force of the tire performance margin identification results of the left front wheel test set under a road friction coefficient of 1.0.

[0121] Fig.16 The three-dimensional schematic diagram of the relationship between total slip rate and normalized force under different road friction coefficients for the tire performance margin identification results of the four wheel test sets.

[0122] Fig.17 To verify the changing rules of features at the center of mass through joint simulation.

[0123] Fig.18 This is the identification result of the joint simulation verification of the four wheels when the road adhesion coefficient is 1.0.

[0124] Fig.19 The identification results are verified by joint simulation of four wheels when the road adhesion coefficient is 0.6.

[0125] Fig. 20 This is the identification result of the joint simulation verification of the four wheels when the road adhesion coefficient is 0.3.

[0126] See attached Fig.11 and attached Fig.12 , which is a three-dimensional view of the test results of the test set in the data set. Here, the left front wheel is taken as an example, which represents the tire lateral force-side slip angle diagram under three friction coefficients and the tire longitudinal force-slip ratio diagram under pure longitudinal slip conditions. The test results show that the trained model can classify the tire performance margin under pure conditions into four categories with a high accuracy. See the attached Fig.13 , attached Fig.14 and attached Fig.15, respectively, are the normalized lateral force and normalized longitudinal force diagrams of tires under composite working conditions when the road friction coefficient is 0.3, 0.6 and 1.0. The images prove the accuracy of the model under composite working conditions. Fig.16 All the test results of the test set for four tires.

[0127] See attached Figures 17 to 20 , demonstrating the generalization capability of the end-to-end tire performance margin identification coefficient described in this application. Fig.17 shows the signal change at the centroid of the input network, Figures 18 to 20 The model demonstrated its ability to identify conditions that it had never encountered. This test was a joint simulation test and all data was transmitted in real time, proving that it can be used for real-time tire performance margin identification. It showed high identification accuracy under conditions where the road friction coefficient was 1.0, 0.6, and 0.3, and showed a high identification accuracy rate for data that did not appear in the training set, with strong generalization ability.

[0128] Fig.21 The flowchart of a method for using an end-to-end tire performance margin identification model provided by an embodiment of the present disclosure is shown in FIG. The method provided by an embodiment of the present disclosure can be executed by a controller in a vehicle.

[0129] like Fig.21 As shown, the method provided by the embodiment of the present disclosure may include the following steps.

[0130] In step S2110, vehicle data is acquired.

[0131] In this step, a controller in the vehicle acquires vehicle data.

[0132] The vehicle data may include, for example, one or more of yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, desired yaw rate, and desired center of mass slip angle. The desired yaw rate and the desired center of mass slip angle may be obtained, for example, through a single-track vehicle model of a controller in the vehicle. The yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, and four wheel torques may be directly obtained, for example, through on-board sensors.

[0133] In one embodiment, the vehicle data is acquired, for example, via a vehicle bus.

[0134] In step S2120, the vehicle data is input into the trained end-to-end tire performance margin identification model to obtain the tire performance margin.

[0135] Wherein, the end-to-end tire performance margin identification model is a model obtained by any of the methods in the above embodiments. In this step, the controller in the vehicle inputs the vehicle data into the trained end-to-end tire performance margin identification model to obtain the tire performance margin; wherein, the end-to-end tire performance margin identification model is a model obtained by any of the methods in the above embodiments.

[0136] Fig. 22 It is a structural schematic diagram of an end-to-end tire performance margin identification model building device provided in an embodiment of the present disclosure.

[0137] like Fig. 22 As shown, the end-to-end tire performance margin identification model building device 2200 provided by the embodiment of the present disclosure may include:

[0138] A first acquisition module 2210 is configured to acquire vehicle test data;

[0139] A second acquisition module 2220, configured to acquire a performance margin corresponding to a tire of the vehicle according to the vehicle test data;

[0140] A labeling module 2230 is configured to label the vehicle test data with tire performance margin;

[0141] A segmentation module 2240 is configured to segment the labeled vehicle test data;

[0142] The training module 2250 is configured to train the segmented and labeled vehicle test data through an artificial intelligence method to complete the modeling of an end-to-end tire performance margin identification model.

[0143] In one embodiment, the vehicle experimental data includes one or more of yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle; the first acquisition module 2210 is configured to acquire one or more of yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques through sensors of the vehicle; and acquire one or more of expected yaw rate and expected center of mass sideslip angle through a single-track vehicle model.

[0144] In one embodiment, the second acquisition module 2220 is configured to acquire the performance margin of the linear region, transition region, saturation region or slip region corresponding to the tire of the vehicle according to the vehicle test data.

[0145] In one embodiment, the segmentation module 2240 is configured to segment the labeled vehicle experimental data based on a sliding time window.

[0146] In one embodiment, the segmentation module 2240 is configured to use N as the segmentation step, where N is a natural number greater than or equal to 2; add N-1 all-zero data before the labeled vehicle test data; segment the labeled vehicle test data in sequence based on a sliding time window in units of N; wherein the number of segments into which the labeled vehicle test data will be divided based on the sliding time window is equal to the number of vehicle test data.

[0147] In one embodiment, the labeling module 2230 is configured to use the performance margin corresponding to the last data or a group of data in each segment of the vehicle test data as the performance margin of the segment of the vehicle test data.

[0148] In one embodiment, the training module 2250 is configured to train the segmented and labeled vehicle test data through a long short-term memory network model LSTM to complete the end-to-end tire performance margin identification model building.

[0149] The end-to-end tire performance margin identification model building device of the present application is configured to acquire vehicle test data through a first acquisition module; a second acquisition module is configured to acquire the performance margin corresponding to the tire of the vehicle according to the vehicle test data; a labeling module is configured to label the vehicle test data with tire performance margin; a segmentation module is configured to segment the labeled vehicle test data; and a training module is configured to train the segmented and labeled vehicle test data through an artificial intelligence method to complete the end-to-end tire performance margin identification model building, thereby realizing the end-to-end tire performance margin identification model building.

[0150] Fig.23 1 is a schematic diagram of the structure of an end-to-end tire performance margin identification device provided by an embodiment of the present disclosure. The device provided by an embodiment of the present disclosure can be installed in a controller in a vehicle.

[0151] like Fig.23 As shown, the end-to-end tire performance margin identification device 2300 provided by the embodiment of the present disclosure may include:

[0152] A third acquisition module 2310 is configured to acquire vehicle data;

[0153] An identification module 2320 is configured to input the vehicle data into the trained end-to-end tire performance margin identification model to obtain the tire performance margin;

[0154] Wherein, the end-to-end tire performance margin identification model is a model obtained by any method as described in the above embodiments.

[0155] See also Fig.24 , Fig.24 2400 is a schematic diagram of the structure of a computer device 2400 provided in an embodiment of the present disclosure. Fig.24 As shown, the computer device in the embodiment of the present disclosure may include: one or more processors 2401, a memory 2402, and an input-output interface 2403. The processor 2401, the memory 2402, and the input-output interface 2403 are connected via a bus 2404. The memory 2402 is used to store a computer program, which includes program instructions, and the input-output interface 2403 is used to receive data and output data, such as for data interaction between a host machine and a computer device, or for data interaction between various virtual machines in a host machine; the processor 2401 is used to execute the program instructions stored in the memory 2402.

[0156] The processor 2401 may perform the following operations:

[0157] Acquire vehicle test data; acquire performance margins corresponding to tires of the vehicle based on the vehicle test data; label the vehicle test data with tire performance margins; segment the labeled vehicle test data; train the segmented and labeled vehicle test data through an artificial intelligence method to complete end-to-end tire performance margin identification model building.

[0158] Or do the following:

[0159] Acquire vehicle data; and input the vehicle data into the trained end-to-end tire performance margin identification model to acquire the tire performance margin.

[0160] In some feasible implementations, the processor 2401 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0161] The memory 2402 may include a read-only memory and a random access memory, and provide instructions and data to the processor 2401 and the input / output interface 2403. A portion of the memory 2402 may also include a non-volatile random access memory. For example, the memory 2402 may also store information about the device type.

[0162] In a specific implementation, the computer device can execute the implementation methods provided in the various steps in the above embodiments through its built-in functional modules. For details, please refer to the implementation methods provided in the various steps in the above embodiments, which will not be repeated here.

[0163] The embodiment of the present disclosure provides a computer device, including: a processor, an input and output interface, and a memory. The processor obtains a computer program in the memory, executes each step of the method shown in the above embodiment, and performs a transmission operation.

[0164] The present disclosure also provides a computer-readable storage medium, which stores a computer program. The computer program is suitable for being loaded by the processor and executing the methods provided in the various steps of the above embodiments. For details, please refer to the implementation methods provided in the various steps of the above embodiments, which will not be repeated here. In addition, the description of the beneficial effects of adopting the same method will not be repeated. For technical details not disclosed in the computer-readable storage medium embodiment involved in the present disclosure, please refer to the description of the method embodiment of the present disclosure. As an example, the computer program can be deployed to be executed on a computer device, or on multiple computer devices located in one location, or on multiple computer devices distributed in multiple locations and interconnected by a communication network.

[0165] The computer-readable storage medium may be the device provided in any of the aforementioned embodiments or the internal storage unit of the computer device, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (smart media card, SMC), a secure digital (secure digital, SD) card, a flash card (flash card), etc. equipped on the computer device. Further, the computer-readable storage medium may also include both the internal storage unit of the computer device and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0166] The present disclosure also provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in various optional modes in the above embodiments.

[0167] The terms "first", "second", etc. in the specification, claims, and drawings of the embodiments of the present disclosure are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product, or equipment that includes a series of steps or units is not limited to the listed steps or modules, but may optionally include steps or modules that are not listed, or may optionally include other step units inherent to these processes, methods, devices, products, or equipment.

[0168] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in this description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0169] The method and related apparatus provided by the embodiments of the present disclosure are described with reference to the method flow chart and / or structural diagram provided by the embodiments of the present disclosure. Specifically, each process and / or block in the method flow chart and / or structural diagram, as well as the combination of the processes and / or blocks in the flow chart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable transmission device to generate a machine, so that the instructions executed by the processor of the computer or other programmable transmission device generate instructions for implementing the process in the process. Figure 1 Schematic diagram of one or more processes and / or structures Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable transmission device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 Schematic diagram of one or more processes and / or structures Figure 1These computer program instructions can also be loaded onto a computer or other programmable transmission device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A flow or multiple flows and / or structures illustrate the steps of the functions specified in one block or multiple blocks.

[0170] The above disclosure is only a preferred embodiment of the present disclosure, which certainly cannot be used to limit the scope of rights of the present disclosure. Therefore, equivalent changes made according to the claims of the present disclosure are still within the scope of the present disclosure.

Claims

1. A method for modeling an end-to-end tire performance margin identification model, characterized in that: include: Obtain vehicle test data; The vehicle test data includes yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle; Acquiring a performance margin corresponding to a tire of the vehicle according to the vehicle test data; Marking the tire performance margin of the vehicle test data; Segmenting the labeled vehicle test data; The segmented and labeled vehicle experimental data is trained in an end-to-end training manner through a long short-term memory network model LSTM to complete the modeling of an end-to-end tire performance margin identification model; The end-to-end tire performance margin identification model is used to obtain the performance margin of the tire; Among them, obtaining vehicle experimental data includes: The yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, and four wheel torques are directly obtained through the vehicle's onboard sensors; wherein the vehicle's onboard sensors cannot directly obtain tire status; Obtaining an expected yaw rate and an expected sideslip angle of the center of mass through a single-track vehicle model; the expected yaw rate and the expected sideslip angle of the center of mass are used to compare with the yaw rate to reflect changes in nonlinear characteristics of the vehicle, so that the LSTM learns an indefinite time lag; the time lag is the lag between the expected yaw rate and the expected sideslip angle of the center of mass and the actual value; The desired yaw rate and the desired sideslip angle of the center of mass are obtained through the single-track vehicle model, including: Derivative the sideslip angle and yaw rate of the center of mass and determine the state space equations; Based on the state-space equation, obtaining a desired yaw rate and a desired sideslip angle of the center of mass; β is the sideslip angle of the center of mass, γ is the yaw rate, represents the value after the derivative of the center of mass sideslip angle, represents the derivative of the yaw rate, m is the sprung mass, V x is the longitudinal speed, F yf, F yr Respectively represent the front axle lateral force and the rear axle lateral force; ll f is the distance from the center of mass to the front axle, l r is the distance from the center of mass to the rear axle, I z is the moment of inertia at the center of mass, the vehicle speed is preset as a constant in the monorail vehicle model, and sinδ≈0 and cosδ≈1 are preset based on a small angle; Among them, F yf =K αf α f , F yr =K αr a r , α f is the front wheel slip angle, α r is the rear wheel slip angle, δ is the wheel turning angle; The state space equation is: Among them, a 11 =-2(K αf +K ar / mV x ), a 12 =2(l r K ar +l f K αf / mV x 2 )-1, a 21 =2(l r K αr +l f K αf / I z ), a 22 =-2(l r 2 K αf +l r 2 K αr / I z V x ), d1=2K αf / mV x , d2=2l f K af / I z ; K af represents the steering stiffness of the front tire and K ar Indicates the steering stiffness of the rear tires.

2. The method according to claim 1, characterized in that Acquiring a performance margin corresponding to a tire of the vehicle according to the vehicle test data includes: A performance margin in a linear region, a transition region, a saturation region or a slip region corresponding to a tire of the vehicle is obtained according to the vehicle test data.

3. The method according to claim 1, characterized in that Segmenting the labeled vehicle test data includes: The vehicle experimental data that has been labeled is segmented based on a sliding time window.

4. The method according to claim 3, characterized in that Segmenting the labeled vehicle experimental data based on the sliding time window includes: Take N as the segment step length, where N is a natural number greater than or equal to 2; Add N-1 all-zero data before the labeled vehicle test data; Sequentially segmenting the labeled vehicle test data based on a sliding time window in units of N; The number of segments into which the labeled vehicle test data is divided based on the sliding time window is equal to the number of the vehicle test data.

5. The method according to claim 4, characterized in that Marking the tire performance margin of the vehicle test data includes: The performance margin corresponding to the last data or a group of data in each section of the vehicle test data is used as the performance margin of the vehicle test data in that section.

6. A method for using an end-to-end tire performance margin identification model, characterized in that: include: Acquire vehicle data; the vehicle test data includes yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle; Inputting the vehicle data into the trained end-to-end tire performance margin identification model to obtain the tire performance margin; Among them, obtaining vehicle experimental data includes: Acquiring yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, and four wheel torques through sensors of the vehicle; Obtaining an expected yaw rate and an expected sideslip angle of the center of mass through a single-track vehicle model; the expected yaw rate and the expected sideslip angle of the center of mass are used to compare with the yaw rate to reflect changes in the nonlinear characteristics of the vehicle, so that the LSTM learns an indefinite time lag; the time lag is the lag between the expected yaw rate and the expected sideslip angle of the center of mass and the actual value; The desired yaw rate and the desired sideslip angle of the center of mass are obtained through the single-track vehicle model, including: Derivative the sideslip angle and yaw rate of the center of mass and determine the state space equations; Based on the state-space equation, obtaining a desired yaw rate and a desired sideslip angle of the center of mass; β is the sideslip angle of the center of mass, γ is the yaw rate, represents the value after the derivative of the center of mass sideslip angle, represents the derivative of the yaw rate, m is the sprung mass, V x is the longitudinal speed, F yf, F yr Respectively represent the front axle lateral force and the rear axle lateral force; ll f is the distance from the center of mass to the front axle, l r is the distance from the center of mass to the rear axle, I z is the moment of inertia at the center of mass, the vehicle speed is preset as a constant in the monorail vehicle model, and sinδ≈0 and cosδ≈1 are preset based on a small angle; Among them, F yf =K αf α f , F yr =K αr a r , α f is the front wheel slip angle, α r is the rear wheel slip angle, δ is the wheel turning angle; The state space equation is: Among them, a 11 =-2(K αf +K αr / mV x ), a 12 =2(l r K αr +l f K αf / mV x 2 )-1, a 21 =2(l r K αr +l f K αf / I z ), a 22 =-2(l r 2 K αf +l r 2 K αr / I z V x ), d1=2K αf / mV x , d2=2l f K af / I z ; K af represents the steering stiffness of the front tire and K αr Indicates the steering stiffness of the rear tires; Wherein, the end-to-end tire performance margin identification model is a model obtained by any method in claims 1-5.

7. An end-to-end tire performance margin identification model building device, characterized in that: include: A first acquisition module, configured to acquire vehicle test data; The vehicle test data includes yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle; a second acquisition module configured to acquire a performance margin corresponding to a tire of the vehicle according to the vehicle test data; A labeling module configured to label the vehicle test data with tire performance margin; A segmentation module, configured to segment the labeled vehicle test data; A training module is configured to train the segmented and labeled vehicle experimental data through a long short-term memory network model LSTM in an end-to-end training manner to complete end-to-end tire performance margin identification model building; the end-to-end tire performance margin identification model building is used to obtain the performance margin of the tire; The first acquisition module is further configured to directly acquire the yaw angular velocity, the steering wheel angle, the longitudinal acceleration, the lateral acceleration, the four wheel speeds, and the four wheel torques through the vehicle-mounted sensors of the vehicle; wherein the vehicle-mounted sensors of the vehicle cannot directly acquire the tire status; The first acquisition module is further configured to acquire an expected yaw rate and an expected sideslip angle of the center of mass through a single-track vehicle model; the expected yaw rate and the expected sideslip angle of the center of mass are used for comparison with the yaw rate to reflect changes in the nonlinear characteristics of the vehicle, so that the LSTM learns an indefinite time lag; the time lag is a lag between the expected yaw rate and the expected sideslip angle of the center of mass and the actual value; The first acquisition module is further configured to derive the center-of-mass sideslip angle and the yaw rate and determine a state space equation; based on the state space equation, obtain a desired yaw rate and a desired center-of-mass sideslip angle; β is the sideslip angle of the center of mass, γ is the yaw rate, represents the value after the derivative of the center of mass sideslip angle, represents the derivative of the yaw rate, m is the sprung mass, V x is the longitudinal speed, F yf, F yr Represent the lateral force of the front axle and the lateral force of the rear axle respectively; l f is the distance from the center of mass to the front axle, l r is the distance from the center of mass to the rear axle, I z is the moment of inertia at the center of mass, the vehicle speed is preset as a constant in the monorail vehicle model, and sinδ≈0 and cosδ≈1 are preset based on a small angle; Among them, F yf =K αf α f , F yr =K αr a r , α f is the front wheel slip angle, α r is the rear wheel slip angle, δ is the wheel turning angle; The state space equation is: Among them, a 11 =-2(K αf +K αr / mV x ), a 12 =2(l r K αr +l f K αf / mV x 2 )-1, a 21 =2(l r K αr +l f K αf / I z ), a 22 =-2(l r 2 K αf +l r 2 K αr / I z V x ), d1=2K αf / mV x , d2=2l f K αf / I z ; K αf represents the steering stiffness of the front tire and K αr Indicates the steering stiffness of the rear tires.

8. An end-to-end tire performance margin identification device, characterized in that: include: a third acquisition module, configured to acquire vehicle data; the vehicle experimental data includes yaw rate, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, four wheel torques, expected yaw rate and expected center of mass sideslip angle; an identification module configured to input the vehicle data into a trained end-to-end tire performance margin identification model to obtain a tire performance margin; The third acquisition module is further configured to directly acquire the yaw angular velocity, steering wheel angle, longitudinal acceleration, lateral acceleration, four wheel speeds, and four wheel torques through the vehicle's onboard sensors; The third acquisition module is further configured to acquire an expected yaw rate and an expected sideslip angle of the center of mass through a single-track vehicle model; the expected yaw rate and the expected yaw rate are used to compare with the yaw rate to reflect changes in the nonlinear characteristics of the vehicle, so that the LSTM learns an indefinite time lag; the time lag is a lag between the expected yaw rate and the expected sideslip angle of the center of mass and the actual value; The third acquisition module is further configured to derive the center of mass sideslip angle and the yaw rate and determine a state space equation; based on the state space equation, obtain a desired yaw rate and a desired center of mass sideslip angle; β is the sideslip angle of the center of mass, γ is the yaw rate, represents the value after the derivative of the center of mass sideslip angle, represents the derivative of the yaw rate, m is the sprung mass, V x is the longitudinal speed, F yf, F yr Respectively represent the front axle lateral force and the rear axle lateral force; ll f is the distance from the center of mass to the front axle, l r is the distance from the center of mass to the rear axle, I z is the moment of inertia at the center of mass, the vehicle speed is preset as a constant in the monorail vehicle model, and sinδ≈0 and cosδ≈1 are preset based on a small angle; Among them, F yf =K αf α f , F yr =K αr a r , α f is the front wheel slip angle, α r is the rear wheel slip angle, δ is the wheel turning angle; The state space equation is: Among them, a 11 =-2(K αf +K αr / mV x ), a 12 =2(l r K αr +l f K αf / mV x 2 )-1, a 21 =2(l r K αr +l f K αf / I z ), a 22 =-2(l r 2 K αf +l r 2 K αrr / I z V x ), d1=2K αf / mV x , d2=2l f K αf / I z ; K αf represents the steering stiffness of the front tire and K αr Indicates the steering stiffness of the rear tires; Wherein, the end-to-end tire performance margin identification model is a model obtained by any method of claims 1-5.

9. An electronic device, characterized in that: include: one or more processors; A storage device configured to store one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the method according to any one of claims 1 to 6.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images