CN112937587A - Road feel simulation method based on K-Medoids and classification regression tree - Google Patents

Road feel simulation method based on K-Medoids and classification regression tree Download PDF

Info

Publication number
CN112937587A
CN112937587A CN202110213877.9A CN202110213877A CN112937587A CN 112937587 A CN112937587 A CN 112937587A CN 202110213877 A CN202110213877 A CN 202110213877A CN 112937587 A CN112937587 A CN 112937587A
Authority
CN
China
Prior art keywords
vehicle
medoids
road feel
steering wheel
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110213877.9A
Other languages
Chinese (zh)
Inventor
赵蕊
蔡锦康
邓伟文
丁能根
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Jingweida Automobile Technology Co ltd
Original Assignee
Nanjing Jingweida Automobile Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Jingweida Automobile Technology Co ltd filed Critical Nanjing Jingweida Automobile Technology Co ltd
Priority to CN202110213877.9A priority Critical patent/CN112937587A/en
Publication of CN112937587A publication Critical patent/CN112937587A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Automation & Control Theory (AREA)
  • Mathematical Physics (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Steering Control In Accordance With Driving Conditions (AREA)

Abstract

The invention discloses a road feel simulation method based on K-Medoids and classification regression trees, which comprises the steps of carrying out real vehicle tests and collecting data, preprocessing the test data, normalizing the test data clustering, dividing training and testing data sets, training and testing road feel simulation models based on K-Medoids and CART regression trees, judging whether the obtained road feel models meet requirements or not, and carrying out road feel simulation according to the obtained road feel simulation models based on K-Medoids and CART regression trees. The input variables of the CART regression tree model are vehicle longitudinal speed, vehicle lateral acceleration, vehicle yaw rate, vehicle vertical load, steering wheel angle and steering wheel angular speed, and the output variable is steering wheel moment. Experiments prove that the road feel simulation model based on the K-Medoids and CART regression tree obtained by the method is high in precision and easy to implement in the modeling process, and the defects of the prior art are overcome to a certain extent.

Description

Road feel simulation method based on K-Medoids and classification regression tree
Technical Field
The invention relates to the field of vehicles, in particular to a road feel simulation method based on K-Medoids and a classification regression tree.
Background
The steering road feel, also called steering force feel or simply called road feel, refers to the condition that a driver feels the motion and stress of the whole vehicle and tires through the feedback torque of a steering wheel. The steering force sense can enable a driver to obtain necessary vehicle running state and running environment information to a certain extent, so that the driver can make a driving decision in a mode most suitable for the current running working condition to ensure the running safety. For most wheeled vehicles traveling on a road surface, the steering system of the vehicle is used as a direct mechanical connection between the steering wheel and the steered wheels, and the reaction force of the road surface to the wheels can be transmitted to the steering wheel through the steered wheels and the steering system to form a 'real road feel'. However, there is no such "real road feel" for vehicle driving simulators and vehicles using steer-by-wire systems. Therefore, an indispensable function is to provide a more realistic road feeling or "simulated road feeling" so as to ensure that the driver can obtain necessary vehicle driving state and driving environment information through the simulated road feeling, so as to enable the driver to drive rationally and ensure driving safety (for a vehicle employing a steer-by-wire system) or to make the driver's behavior more realistic (for a driving simulator).
How to simulate the more realistic road feel is called road feel modeling. At present, the main road feel modeling method is a modeling method using a mechanism. The method needs more setting parameters and has low simulation precision compared with the actually measured steering road feel. Chinese patent with patent publication No. CN110606121A entitled "a steer-by-wire road feel simulation control method" discloses a control system for steering wheel feedback force. The patent proposes to construct a steering load model and calculate the steering resisting moment through dynamics, and the modeling method belongs to mechanism modeling.
Disclosure of Invention
The invention aims to provide a road feel simulation method based on K-Medoids and classification regression trees, which is used for modeling by using real vehicle test data, a K-Medoids clustering algorithm and a CART regression tree algorithm to obtain a road feel simulation model based on the K-Medoids and the CART regression tree and solves the problems of complex model structure, low precision and the like in the traditional mechanism modeling.
In order to achieve the above purpose, the invention provides a road feel simulation method based on K-Medoids and classification regression trees, comprising the following steps:
step one, carrying out a real vehicle test and acquiring data: the method comprises the following steps that a driver drives a vehicle to run on a test road, and collected test data comprise vehicle longitudinal speed, vehicle transverse acceleration, vehicle yaw velocity, vehicle vertical load, steering wheel corners, steering wheel angular speed and steering wheel moment;
step two, test data preprocessing: carrying out normalization processing on the test data after removing abnormal points to obtain a normalized test data set;
step three, normalization test data clustering: clustering the normalized test data by using a K-Medoids clustering algorithm, and obtaining data classes with the same number as that of the clustered communities after clustering;
step four, dividing training and testing data sets: dividing the normalized test data set into a training data set and a test data set to obtain a clustered training data set and a clustered test data set;
step five, training and testing the road feel model: using the clustered training data set and a CART regression tree algorithm, and when training a model, inputting variables of the model comprise vehicle longitudinal speed, vehicle transverse acceleration, vehicle yaw velocity, vehicle vertical load, steering wheel turning angle and steering wheel angular velocity; the output variable is the torque of a steering wheel, and a road feel simulation model which has the same number as the data class and is based on K-Medoids and CART regression trees is obtained through training; testing the obtained road feel simulation model based on the K-Medoids and CART regression tree by using a test data set;
step six, judging whether the obtained road feel model meets the requirements: judging whether the model meets the requirements or not according to the precision of the obtained road feel model;
and seventhly, carrying out road feel simulation according to the obtained road feel simulation model based on the K-Medoids and the CART regression tree.
Further, the road surface types involved in the real vehicle test include expressways, urban roads, rural roads and suburban roads; the vehicle running conditions involved in the actual vehicle test comprise straight running, reversing, turning, pivot steering, uphill slope and downhill slope.
Further, in step two, the removed abnormal points include data points beyond the normal value range, data points with severely deviated distribution, and data points with a variation range beyond the normal range.
The data points that exceed the normal value range are defined as: a point that is clearly not within the normal range. For example, in a certain test, the highest vehicle longitudinal speed is only 50km/h, and in the data collected in the test, the data points with the vehicle longitudinal speed more than 50km/h are all data points beyond the normal range.
The definition of the data points where the distribution deviates severely is: the data are distributed at points outside the a-fold standard deviation range of the relevant variable data. That is, the standard deviation of each variable data acquired in a certain test is calculated, and if one or more variable values of a certain data point are greater than 3.5 times or less than minus 3.5 times of the standard deviation of the relevant variable, the data point is called a point where the distribution is seriously deviated.
The data points with the variation amplitude exceeding the normal range are defined as follows: the maximum instantaneous change amplitude of each variable under the normal condition is preset, and if the absolute value of the difference value of one or more variable values of a certain data point relative to the corresponding variable value of the previous data point in the actual test data set is larger than the maximum instantaneous change amplitude of the related variable, the maximum instantaneous change amplitude of each variable exceeds the normal range. If the expert confirms that the maximum instantaneous change amplitude of the steering wheel torque is 0.5N when a high-speed driving test is performed by using a small passenger car, the data points in which the absolute value of the difference between the steering wheel torque value and the previous data point is greater than 0.5N are all points with the change amplitude exceeding the normal range.
Further, in the second step, the test data is normalized according to the following formula to obtain normalized test data:
Figure BDA0002953248310000031
wherein i is a data number, j is a variable number, and xi,jRepresenting the i-th group without normalizationThe j variable, X, in the datajAnd representing a set consisting of variable data values corresponding to all j, min representing the minimum value of the related variable in the test data after the abnormal point is removed, and max representing the maximum value of the related variable in the test data after the abnormal point is removed.
Preferably, in step three, when clustering is performed by using a K-Medoids clustering algorithm, variables participating in clustering include, but are not limited to, a vehicle longitudinal speed, a vehicle lateral acceleration, a vehicle yaw rate, a vehicle vertical load, a steering wheel angle and a steering wheel angular speed, and the number K of clusters is 4. And after clustering is finished, the coordinates of the central point of 4 clusters can be obtained. These coordinates are normalized values of the relevant variables.
The clustering steps of the K-Medoids algorithm are as follows:
(1) and determining the number k of the required communities. In a specific embodiment, the number k of the communities is 4;
(2) randomly selecting k data points in a data set to be clustered as central points of k clusters;
(3) calculating Euclidean distances from all non-central points to k central points determined in the previous step, wherein a community corresponding to the central point closest to the data point is a community to which the data point belongs;
(4) sequentially selecting one point in each community, calculating the sum of Euclidean distances between the point and all other points in the community where the point is located at present, and taking the point with the minimum sum of the Euclidean distances as a new central point of the community;
(5) repeating the steps (2) and (3) until the central point of each cluster is not changed.
Preferably, in step four, when the training and testing data sets are divided, a certain number of data points are randomly selected from the normalized test data set as the training data set, and the others are all used as the testing data set. In a preferred embodiment, the ratio is 70%, i.e. the ratio of the number of data points in the training data set to the number of data points in the test data set is 7: 3.
Preferably, in the fifth step, modeling is performed by using the training data set after clustering and the CART regression tree algorithm, and K road feel simulation models based on K-Medoids and CART regression trees, which are the same as the number of data classes (also the number of clustered communities), are obtained through training. When the model is trained, input variables of the CART regression tree model comprise vehicle longitudinal speed, vehicle transverse acceleration, vehicle yaw velocity, vehicle vertical load, steering wheel turning angle and steering wheel angular speed; the output variable is the steering wheel torque. And during model training, training by using training models belonging to different communities to obtain a force sense model corresponding to the related community, wherein the model can only predict the points of the related community, otherwise, the prediction precision is difficult to ensure. The model obtained by training the same type of training data points is related to the type of the data points, namely the road feel simulation model corresponding to a certain type of training data points can only be used for predicting the steering wheel torque of the type of data points. In the embodiment of the present invention, after training, the training data points with k-4 types will obtain the corresponding road feel simulation models with k-4 types.
When a road sense simulation model based on K-Medoids and CART regression trees is trained, the specific steps are as follows:
the CART regression tree model is represented as:
Figure BDA0002953248310000041
wherein, f (x) is a CART regression tree function, m is a positive integer greater than 1, I is a unit matrix, and x is an input variable; the data space is divided into R1～RmCells, each cell having a fixed output value cm；
Calculating the error of the model output value and the actual value:
Figure BDA0002953248310000042
wherein x isiFor input of the i-th data of variable x, yiIs the actual output value; i is a positive integer greater than 1;
suppose that the jth input variable x is selectedjFor segmentation variables, the input variables being vehicle longitudinal speed, vehicleAny one of the lateral acceleration of the vehicle, the yaw angular velocity of the vehicle, the yaw angular acceleration of the vehicle, the vertical load of the vehicle, the turning angle of the steering wheel and the angular velocity of the steering wheel is a segmentation variable, and j is a variable number; taking the value s of the segmentation variable as a segmentation point to obtain two regions R1，R2：
R1(j,s)＝{x|x(f)≤s}；R2(j,s)＝{x|x(f)＞s}
When j and s are fixed, find the representative value c of the two regions1，c2The squared difference over the respective intervals is minimized, i.e.:
Figure BDA0002953248310000043
in the formula c1，c2Is the average over the interval, i.e.:
Figure BDA0002953248310000044
the working steps for training the CART regression tree model using the training data set are as follows:
1) inputting: a training data set D;
2) and (3) outputting: regression tree f (x);
3) recursively dividing each region into two sub-regions in an input space where the training data set is located, and determining an output value of each sub-region; constructing a binary decision tree, comprising the steps of:
selecting an optimal segmentation variable j and a segmentation point s, and solving:
Figure BDA0002953248310000051
secondly, traversing the variable j, scanning a segmentation point s for the fixed segmentation variable j, and selecting a pair (j, s) which enables the above formula to reach the minimum value;
-dividing the area by the selected pair (j, s) and determining the corresponding output value:
R1(j,s)＝{x|x(f)≤s}；R2(j,s)＝{x|x(f)＞s}
Figure BDA0002953248310000052
in the formula, NmIs the total number of data points in space;
fourthly, continuously calling the steps (1) and (2) for the two subregions until the cycle number reaches an upper limit value;
divide the input space into M regions R1,R2,...,RMAnd generating a decision tree.
In one embodiment of the present invention, when the CART regression tree model is trained using the training data set, the upper limit of the number of cycles is set to 20, i.e., the maximum tree depth is 20.
Further, when testing road feel simulation models based on K-Medoids and CART regression trees, the Mean Square Error (MSE) value can be used, but is not limited to being used, as the evaluation criterion of the model quality. When the road feel simulation model based on the K-Medoids and the CART regression tree is tested by using a test data set, the steps are as follows:
1) sequentially taking out test data points in the test data set, taking the longitudinal speed of the vehicle, the lateral acceleration of the vehicle, the yaw velocity of the vehicle, the vertical load of the vehicle, the turning angle of the steering wheel and the angular velocity of the steering wheel corresponding to the test data points as input variables, and inputting a road feel model based on K-Medoids and CART regression trees corresponding to the data class to which the test data points belong to obtain a predicted steering wheel moment value;
2) calculating the whole of the test data set, and predicting the MSE value between the obtained steering wheel moment data set and the real steering wheel moment data set;
and when judging whether the obtained road feel model meets the requirements: and if the MSE value is larger than the threshold value alpha, the established road feel simulation model based on the K-Medoids and the CART regression tree is considered to be acceptable, otherwise, the road feel simulation model is not acceptable. The threshold value alpha is determined empirically by the expert and in a preferred embodiment is set to 0.2.
And after the modeling is finished, the method also comprises a model application step, and road feel simulation is carried out according to the obtained multiple road feel simulation models based on the K-Medoids and the CART regression tree. And acquiring real-time driving data of the vehicle as new data, wherein the new data comprises the longitudinal speed of the vehicle, the lateral acceleration of the vehicle, the yaw velocity of the vehicle, the vertical load of the vehicle, the steering wheel angle and the steering wheel angular velocity, and according to the driving data, calculating Euclidean distances from k clustering center coordinates, wherein the data class to which the clustering center with the minimum Euclidean distance belongs is the new data class group. Then, the driving data is input into a road feel simulation model which corresponds to the belonged class and is based on the K-Medoids and the CART regression tree, a predicted steering wheel moment value is obtained through model calculation, and the steering wheel is controlled according to the steering wheel moment value, so that vivid road feel is simulated.
Due to the adoption of the technical scheme, the invention achieves the following technical effects: the method is based on real vehicle road acquisition data, adopts a K-Medoids clustering algorithm for clustering, and carries out modeling based on a CART regression tree algorithm, so that the data acquisition is convenient, the modeling process is easy to implement, and the obtained model has higher precision; the road feel simulation model obtained according to the invention can obtain vivid steering road feel by performing road feel simulation, and solves the problems of low precision, difficult real-time property guarantee and the like of the traditional mechanism modeling method to a certain extent.
Drawings
Fig. 1 is a flow chart of modeling steps in a road feel simulation method based on K-Medoids and classification regression trees according to the invention.
Fig. 2 is a (partial) steering wheel angle curve collected in an embodiment according to the invention in a low speed condition.
FIG. 3 is model test data (partial) in an embodiment in accordance with the invention.
Detailed Description
In order to make the technical solution of the embodiments of the present invention better understood, the technical solution of the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by equivalent changes and modifications by one skilled in the art based on the embodiments of the present invention, shall fall within the scope of the present invention.
Referring to fig. 1 to fig. 3, the present embodiment provides a road feel simulation method based on K-media and classification regression tree, which includes modeling steps S1-S7, and a model application step. Steps S1-S7 of the modeling process are described in detail below in conjunction with FIG. 1.
S1, carrying out a real vehicle test and acquiring data:
selecting a driver to carry out a real vehicle test, wherein the vehicle runs in a test road, and the test road types include but are not limited to an expressway, an urban road, a rural road and a suburban road; the related vehicle running conditions comprise straight running, reverse running, turning, pivot steering, uphill slope and downhill slope.
The collected test data includes vehicle longitudinal speed, vehicle lateral acceleration, vehicle yaw rate, vehicle vertical load, steering wheel angle, steering wheel angular velocity, steering wheel moment, and the like. Steering wheel angle, steering wheel angular velocity, steering wheel torque are measured using a steering angle torque sensor model KISTLER MSW DTI sensors. Other data such as the longitudinal speed of the vehicle, the lateral acceleration of the vehicle, the yaw rate of the vehicle and the like are measured by an inertial navigation system, and the model is OxTs RT 3002. The data acquisition frequency in this example is 100 Hz. The total mileage of the vehicle is 128 kilometers, and the total test time is about 3 hours.
As shown in fig. 2, the steering wheel angle curve (local) at the low speed condition collected in the experiment of the present embodiment is represented by an actual steering wheel angle (°) -time(s) curve, and the steering wheel angle amplitude is-500 ° to 500 ° during 140 s.
S2, test data preprocessing:
processing the test data includes removing outliers and normalizing the data. The removed abnormal points include data points outside the normal value range, data points with severely deviated distribution and data points with the variation amplitude exceeding the normal range. The abnormal points of the test data can be removed manually or filtered by a low-pass filter.
In this embodiment, the collected test data is normalized according to the following formula, so as to obtain normalized test data. The normalization formula may take the following formula, but is not limited to it:
Figure BDA0002953248310000071
wherein i is a data number, j is a variable number, and xi,jDenotes the j variable, X, in the non-normalized i group of datajAnd representing a set consisting of variable data values corresponding to all j, min representing the minimum value of the related variable in the test data after the abnormal point is removed, and max representing the maximum value of the related variable in the test data after the abnormal point is removed.
And obtaining a normalized test data set after pretreatment.
S3, clustering normalized test data
In this embodiment, the normalized test data is clustered by using a K-Medoids clustering algorithm, the number of clustering groups is set to K equal to 4, and 4 clustering center coordinates and 4 data classes corresponding to the K equal to 4 clustering center coordinates are obtained after clustering. These cluster center coordinates are normalized values of the correlation variables.
Variables participating in clustering include, but are not limited to, vehicle longitudinal velocity, vehicle lateral acceleration, vehicle yaw rate, vehicle vertical load, steering wheel angle, steering wheel angular velocity. In this embodiment, when clustering is performed on the normalized test data by using K-Medoids, it is necessary to set the number K of clustering communities to be 4, and after clustering is completed, coordinates of 4 clustering center points can be obtained. The clustering steps of the K-Medoids algorithm are as follows:
(1) and determining the number k of the required communities. In a specific embodiment, the number k of the communities is 4;
(2) randomly selecting k data points in a data set to be clustered as central points of k clusters;
(3) calculating Euclidean distances from all non-central points to k central points determined in the previous step, wherein a community corresponding to the central point closest to the data point is a community to which the data point belongs;
(4) sequentially selecting one point in each community, calculating the sum of Euclidean distances between the point and all other points in the community where the point is located at present, and taking the point with the minimum sum of the Euclidean distances as a new central point of the community;
(5) repeating the steps (2) and (3) until the central point of each cluster is not changed.
S4, dividing training data set test data set
When the training data set and the test data set are divided, data points in a certain proportion are randomly selected from the normalized test data set to serve as the training data set, and the other data sets serve as the test data set. In this embodiment the ratio is 70%, i.e. the ratio of the number of data points in the training data set to the number of data points in the test data set is 7: 3. The clustered training data set is referred to as clustered training data, and the clustered test data set is referred to as a clustered test data set.
S5, training a road feel model based on K-Medoids and CART regression trees:
modeling is performed by using a training data set and a CART regression tree algorithm, a road feel simulation model based on K-Medoids and CART regression trees with the same number as that of data classes is obtained through training, and 4 road feel simulation models corresponding to 4 clustering centers (namely 4 data classes) are obtained in the embodiment. When the model is trained, input variables of the CART regression tree model comprise vehicle longitudinal speed, vehicle transverse acceleration, vehicle yaw velocity, vehicle vertical load, steering wheel turning angle and steering wheel angular speed; the output variable is the steering wheel torque. The model obtained by training the same type of training data points is related to the type of the data points, namely the road feel simulation model corresponding to a certain type of training data points can only be used for predicting the steering wheel torque of the type of data points. In this embodiment, 4 types of training data points are trained to obtain 4 corresponding road feel simulation models.
When a road sense simulation model based on K-Medoids and CART regression trees is trained, the specific steps are as follows:
the CART regression tree model is represented as:
Figure BDA0002953248310000081
wherein, f (x) is a CART regression tree function, m is a positive integer greater than 1, I is a unit matrix, and x is an input variable; the data space is divided into R1～RmCells, each cell having a fixed output value cm；
Calculating the error of the model output value and the actual value:
Figure BDA0002953248310000082
wherein x isiFor input of the i-th data of variable x, yiIs the actual output value; i is a positive integer greater than 1;
suppose that the jth input variable x is selectedjThe input variable is any one of a longitudinal speed of a vehicle, a lateral acceleration of the vehicle, a yaw velocity of the vehicle, a yaw acceleration of the vehicle, a vertical load of the vehicle, a steering wheel angle and an angular velocity of the steering wheel, and j is a variable number; taking the value s of the segmentation variable as a segmentation point to obtain two regions R1，R2：
R1(j,s)＝{x|x(f)≤s}；R2(j,s)＝{x|x(f)＞s}
When j and s are fixed, find the representative value c of the two regions1，c2The squared difference over the respective intervals is minimized, i.e.:
Figure BDA0002953248310000091
in the formula c1，c2Is the average over the interval, i.e.:
Figure BDA0002953248310000092
the working steps for training the CART regression tree model using the training data set are as follows:
1) inputting: a training data set D;
2) and (3) outputting: regression tree f (x);
3) recursively dividing each region into two sub-regions in an input space where the training data set is located, and determining an output value of each sub-region; constructing a binary decision tree, comprising the steps of:
selecting an optimal segmentation variable j and a segmentation point s, and solving:
Figure BDA0002953248310000093
secondly, traversing the variable j, scanning a segmentation point s for the fixed segmentation variable j, and selecting a pair (j, s) which enables the above formula to reach the minimum value;
-dividing the area by the selected pair (j, s) and determining the corresponding output value:
R1(j,s)＝{x|x(f)≤s}；R2(j,s)＝{x|x(f)＞s}
Figure BDA0002953248310000094
in the formula, NmIs the total number of data points in space;
fourthly, continuously calling the steps (1) and (2) for the two subregions until the cycle number reaches an upper limit value; in the present embodiment, the upper limit value of the number of cycles is set to 20, that is, the maximum tree depth is 20.
Divide the input space into M regions R1,R2,...,RMAnd generating a decision tree.
The example was trained using the Hewlett packard Z1G6 workstation, and the total training time of 4 road feel models was 5 hours and 32 minutes.
S6, testing a road feel model based on K-Medoids and CART regression trees:
the steps of using the test data set to test the obtained road feel simulation model based on the K-Medoids and CART regression tree are as follows:
1) sequentially taking out test data points in the test data set, taking the longitudinal speed of the vehicle, the lateral acceleration of the vehicle, the yaw velocity of the vehicle, the vertical load of the vehicle, the turning angle of the steering wheel and the angular velocity of the steering wheel corresponding to the test data points as input variables, and inputting a road feel model based on K-Medoids and CART regression trees corresponding to the data class to which the test data points belong to obtain a predicted steering wheel moment value;
2) calculating the whole of the test data set, and predicting the Mean Square Error (MSE) between the obtained steering wheel moment data set and the real steering wheel moment data set;
judging whether the obtained road feel model meets the requirements: and if the MSE value is larger than the threshold value alpha, the established road feel simulation model based on the K-Medoids and the CART regression tree is considered to be acceptable, otherwise, the road feel simulation model is not acceptable.
In the present embodiment, as shown in fig. 3, which represents a model test curve (local), it can be seen from the graph that in a time period of 0-200s, a simulated steering wheel moment-time curve (sim, solid line) substantially coincides with an actual steering wheel moment-time curve (real, dashed line), and the MSE value is 0.11645.
S7, judging whether the obtained road feel model meets the requirements or not
And (3) testing to obtain an MSE (mean square error) value 0.11645 which is far smaller than a threshold value alpha preset by an expert as 0.2, wherein the obtained model meets the precision requirement, is acceptable and does not need a supplementary road mining test.
The model application step:
after modeling is completed, the road feel simulation method further comprises the following model application steps: and performing road feel simulation according to the obtained road feel simulation model based on the K-Medoids and the CART regression tree. Inputting the obtained 4 road feel simulation models based on K-Medoids and CART regression trees into a driving simulator, acquiring running state parameters such as vehicle longitudinal speed, vehicle transverse acceleration, vehicle yaw velocity, vehicle vertical load, steering wheel turning angle, steering wheel angular velocity and the like of a simulated vehicle in real time when a simulated driving test is carried out on the driving simulator, and calculating Euclidean distances between the simulated vehicle and K which is 4 clustering center coordinates according to the running data, wherein the data class to which the clustering center with the minimum Euclidean distance belongs is the new data class to which the new data belongs. And then inputting the corresponding variable of the driving state parameter as an input variable into a road feel simulation model corresponding to the belonging class, calculating to obtain a steering wheel torque value through a road feel simulation model based on K-Medoids and a CART regression tree, and controlling a steering wheel in real time according to the steering wheel torque value, thereby simulating more realistic road feel. Tests prove that the road feel simulation model established by the method has stable performance, high precision and high operation speed, and overcomes the defects of the prior art to a certain extent.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention; also, the above description should be understood as being readily apparent to those skilled in the relevant art and can be implemented, and therefore, other equivalent changes and modifications without departing from the concept disclosed herein are intended to be included within the scope of the present invention.

Claims (10)

1. A road feel simulation method based on K-Medoids and classification regression trees is characterized by comprising the following steps:
step one, carrying out a real vehicle test and acquiring data: the method comprises the following steps that a driver drives a vehicle to run on a test road, and collected test data comprise vehicle longitudinal speed, vehicle transverse acceleration, vehicle yaw velocity, vehicle vertical load, steering wheel corners, steering wheel angular speed and steering wheel moment;
step two, test data preprocessing: carrying out normalization processing on the test data after removing abnormal points to obtain a normalized test data set;
step three, normalization test data clustering: clustering the normalized test data by using a K-Medoids clustering algorithm, and obtaining data classes with the same number as that of the clustered communities after clustering;
step four, dividing training and testing data sets: dividing the normalized test data set into a training data set and a test data set to obtain a clustered training data set and a clustered test data set;
step five, training and testing the road feel model: using the clustered training data set and a CART regression tree algorithm, and when training a model, inputting variables of the model comprise vehicle longitudinal speed, vehicle transverse acceleration, vehicle yaw velocity, vehicle vertical load, steering wheel turning angle and steering wheel angular velocity; the output variable is the torque of a steering wheel, and a road feel simulation model which has the same number as the data class and is based on K-Medoids and CART regression trees is obtained through training; testing the obtained road feel simulation model based on the K-Medoids and CART regression tree by using a test data set;
step six, judging whether the obtained road feel model meets the requirements: judging whether the model meets the requirements or not according to the precision of the obtained road feel model;
and seventhly, carrying out road feel simulation according to the obtained road feel simulation model based on the K-Medoids and the CART regression tree.
2. The road feel simulation method based on K-Medoids and classification regression trees according to claim 1, characterized in that in the real vehicle test of step one: the test road types comprise an expressway, an urban road, a rural road and a suburban road; the running conditions of the vehicle comprise straight running, reversing, turning, pivot steering, uphill slope and downhill slope.
3. The road feel simulation method based on K-Medoids and classification regression trees as claimed in claim 1, wherein in step two, the removed abnormal points include data points out of the normal range, data points with a severely deviated distribution, and data points with a variation amplitude out of the normal range.
4. The road feel simulation method based on K-Medoids and classification regression trees according to claim 1, wherein in step two, the test data is normalized according to the following formula to obtain normalized test data:
Figure FDA0002953248300000021
wherein i is a data number, j is a variable number, and xi,jDenotes the j variable, X, in the non-normalized i group of datajAnd representing a set consisting of variable data values corresponding to all j, min representing the minimum value of the related variable in the test data after the abnormal point is removed, and max representing the maximum value of the related variable in the test data after the abnormal point is removed.
5. The road feel simulation method based on K-Medoids and classification regression trees as claimed in claim 1, wherein in step three, when clustering is performed by using a K-Medoids clustering algorithm, variables participating in clustering include vehicle longitudinal speed, vehicle lateral acceleration, vehicle yaw rate, vehicle vertical load, steering wheel angle and steering wheel angular speed, and the number of clusters is 4.
6. The road feel simulation method based on K-Medoids and classification regression trees as claimed in claim 1, wherein in step four, when the training and testing data sets are divided, a certain proportion of data points are randomly selected from the normalized test data set as the training data set, and other data points are all used as the testing data set.
7. The road feel simulation method based on K-Medoids and classification regression trees according to any one of claims 1-6, wherein in the fifth step, when training the road feel simulation model based on K-Medoids and CART regression trees, the specific steps are as follows:
the CART regression tree model is represented as:
Figure FDA0002953248300000022
wherein, f (x) is a CART regression tree function, m is a positive integer greater than 1, I is a unit matrix, and x is an input variable; the data space is divided into R1～RmCells, each cell having a fixed output value cm；
Calculating the error of the model output value and the actual value:
Figure FDA0002953248300000023
wherein x isiFor input of the i-th data of variable x, yiIs the actual output value; i is a positive integer greater than 1;
suppose that the jth input variable x is selectedjThe input variable is any one of a longitudinal speed of a vehicle, a lateral acceleration of the vehicle, a yaw velocity of the vehicle, a yaw acceleration of the vehicle, a vertical load of the vehicle, a steering wheel angle and an angular velocity of the steering wheel, and j is a variable number; taking the value s of the segmentation variable as a segmentation point to obtain two regions R1，R2：
R1(j,s)＝{x|x(f)≤s}；R2(j,s)＝{x|x(f)＞s}
When j and s are fixed, find the representative value c of the two regions1，c2The squared difference over the respective intervals is minimized, i.e.:
Figure FDA0002953248300000031
in the formula c1，c2Is the average over the interval, i.e.:
Figure FDA0002953248300000032
the working steps for training the CART regression tree model using the training data set are as follows:
1) inputting: a training data set D;
2) and (3) outputting: regression tree f (x);
3) recursively dividing each region into two sub-regions in an input space where the training data set is located, and determining an output value of each sub-region; constructing a binary decision tree, comprising the steps of:
selecting an optimal segmentation variable j and a segmentation point s, and solving:
Figure FDA0002953248300000033
secondly, traversing the variable j, scanning a segmentation point s for the fixed segmentation variable j, and selecting a pair (j, s) which enables the above formula to reach the minimum value;
-dividing the area by the selected pair (j, s) and determining the corresponding output value:
R1(j,s)＝{x|x(f)≤s}；R2(j,s)＝{x|x(f)＞s}
Figure FDA0002953248300000034
in the formula, NmIs the total number of data points in space;
fourthly, continuously calling the steps (1) and (2) for the two subregions until the cycle number reaches an upper limit value;
divide the input space into M regions R1,R2,...,RMAnd generating a decision tree.
8. The K-Medoids and classification regression tree based road feel simulation method of claim 1, wherein the upper limit value of the number of cycles is set to 20 when training the CART regression tree model using the training data set.
9. The road feel simulation method based on K-Medoids and classification regression trees as claimed in claim 1, wherein the specific steps of testing the obtained road feel simulation model based on K-Medoids and CART regression trees and judging whether the obtained road feel model meets the requirements according to the test result are as follows:
1) sequentially taking out test data points in the test data set, taking the longitudinal speed of the vehicle, the lateral acceleration of the vehicle, the yaw velocity of the vehicle, the vertical load of the vehicle, the turning angle of the steering wheel and the angular velocity of the steering wheel corresponding to the test data points as input variables, and inputting a road feel model based on K-Medoids and CART regression trees corresponding to the data class to which the test data points belong to obtain a predicted steering wheel moment value;
2) calculating the whole of the test data set, and predicting the MSE value between the obtained steering wheel moment data set and the real steering wheel moment data set;
judging whether the obtained road feel model meets the requirements: and if the MSE value is larger than the threshold value alpha, the established road feel simulation model based on the K-Medoids and the CART regression tree is considered to be acceptable, otherwise, the road feel simulation model is not acceptable.
10. The K-Medoids and CART regression tree based road feel simulation method according to claim 9, wherein the threshold α is 0.2.
CN202110213877.9A 2021-02-26 2021-02-26 Road feel simulation method based on K-Medoids and classification regression tree Pending CN112937587A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110213877.9A CN112937587A (en) 2021-02-26 2021-02-26 Road feel simulation method based on K-Medoids and classification regression tree

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110213877.9A CN112937587A (en) 2021-02-26 2021-02-26 Road feel simulation method based on K-Medoids and classification regression tree

Publications (1)

Publication Number Publication Date
CN112937587A true CN112937587A (en) 2021-06-11

Family

ID=76246313

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110213877.9A Pending CN112937587A (en) 2021-02-26 2021-02-26 Road feel simulation method based on K-Medoids and classification regression tree

Country Status (1)

Country Link
CN (1) CN112937587A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116465647A (en) * 2023-04-18 2023-07-21 日照朝力信息科技有限公司 Automobile performance testing method and system based on virtual reality technology

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105083373A (en) * 2015-06-15 2015-11-25 南京航空航天大学 Steering-by-wire feeling device based on parameter estimation and control method thereof
CN109885883A (en) * 2019-01-21 2019-06-14 江苏大学 A kind of control method of the unmanned vehicle transverse movement based on GK clustering algorithm model prediction
CN110606121A (en) * 2019-08-28 2019-12-24 中国第一汽车股份有限公司 Drive-by-wire steering road feel simulation control method
US20200023846A1 (en) * 2018-07-23 2020-01-23 SparkCognition, Inc. Artificial intelligence-based systems and methods for vehicle operation
CN112550450A (en) * 2020-12-26 2021-03-26 浙江天行健智能科技有限公司 Road feel simulation method based on K-Means and CART regression tree

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105083373A (en) * 2015-06-15 2015-11-25 南京航空航天大学 Steering-by-wire feeling device based on parameter estimation and control method thereof
US20200023846A1 (en) * 2018-07-23 2020-01-23 SparkCognition, Inc. Artificial intelligence-based systems and methods for vehicle operation
CN109885883A (en) * 2019-01-21 2019-06-14 江苏大学 A kind of control method of the unmanned vehicle transverse movement based on GK clustering algorithm model prediction
CN110606121A (en) * 2019-08-28 2019-12-24 中国第一汽车股份有限公司 Drive-by-wire steering road feel simulation control method
CN112550450A (en) * 2020-12-26 2021-03-26 浙江天行健智能科技有限公司 Road feel simulation method based on K-Means and CART regression tree

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116465647A (en) * 2023-04-18 2023-07-21 日照朝力信息科技有限公司 Automobile performance testing method and system based on virtual reality technology
CN116465647B (en) * 2023-04-18 2024-03-26 日照朝力信息科技有限公司 Automobile performance testing method and system based on virtual reality technology

Similar Documents

Publication Publication Date Title
CN112801143B (en) Steering road feel simulation method based on K-Means and Gaussian process regression
CN113010967B (en) Intelligent automobile in-loop simulation test method based on mixed traffic flow model
CN112632706B (en) Road feel simulation method based on GMM and CART regression tree
CN113239986B (en) Training method and device for vehicle track evaluation network model and storage medium
CN112550450B (en) Road feel simulation method based on K-Means and CART regression tree
CN112749508B (en) Road feel simulation method based on GMM and BP neural network
CN112632705B (en) Road feel simulation method based on GMM and Gaussian process regression
CN112580149A (en) Vehicle following model generation method based on generation of countermeasure network and driving duration
CN112937587A (en) Road feel simulation method based on K-Medoids and classification regression tree
CN113642114A (en) Modeling method for humanoid random car following driving behavior capable of making mistakes
CN112528568B (en) Road feel simulation method based on K-Means and BP neural network
CN113033902A (en) Automatic driving track-changing planning method based on improved deep learning
CN116975671A (en) L3-level automatic driving vehicle driver trust evaluation method and system
CN116946183A (en) Commercial vehicle driving behavior prediction method considering driving capability and vehicle equipment
CN116620327A (en) Lane changing decision method for realizing automatic driving high-speed scene based on PPO and Lattice
CN112597597A (en) Road feel simulation method based on K-Medoids and BP neural network
CN114148349B (en) Vehicle personalized following control method based on generation of countermeasure imitation study
CN112651134A (en) Steering road feel simulation method based on Gaussian process regression
Chen et al. DCFS-based deep learning supervisory control for modeling lane keeping of expert drivers
CN112572606B (en) Working condition fusion road sense simulation method based on SVM (support vector machine) and BP (back propagation) neural network
CN112836722B (en) Road feel simulation method based on data driving
CN112685956B (en) Road feel simulation method based on CART regression tree
Zhao et al. An improved data-driven method for steering feedback torque of driving simulator
CN112949173A (en) Road feel simulation method based on K-Medoids and Gaussian process regression
CN112308171A (en) Vehicle position prediction modeling method based on simulated driver

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination