CN113911124B - Road adhesion coefficient estimation method based on Bayes and electronic equipment - Google Patents

Road adhesion coefficient estimation method based on Bayes and electronic equipment Download PDF

Info

Publication number
CN113911124B
CN113911124B CN202111140257.3A CN202111140257A CN113911124B CN 113911124 B CN113911124 B CN 113911124B CN 202111140257 A CN202111140257 A CN 202111140257A CN 113911124 B CN113911124 B CN 113911124B
Authority
CN
China
Prior art keywords
road surface
road
bayes
model
real
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.)
Active
Application number
CN202111140257.3A
Other languages
Chinese (zh)
Other versions
CN113911124A (en
Inventor
钟毅
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.)
Wuhan University of Technology WUT
Original Assignee
Wuhan University of Technology WUT
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 Wuhan University of Technology WUT filed Critical Wuhan University of Technology WUT
Priority to CN202111140257.3A priority Critical patent/CN113911124B/en
Publication of CN113911124A publication Critical patent/CN113911124A/en
Application granted granted Critical
Publication of CN113911124B publication Critical patent/CN113911124B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • B60W40/02—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 related to ambient conditions
    • B60W40/06—Road conditions
    • B60W40/064—Degree of grip
    • 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
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • 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
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0043—Signal treatments, identification of variables or parameters, parameter estimation or state estimation
    • 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
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0043—Signal treatments, identification of variables or parameters, parameter estimation or state estimation
    • B60W2050/0052—Filtering, filters

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a road adhesion coefficient estimation method based on Bayes and electronic equipment. The invention divides the whole data point set into a plurality of data subsets, the point set of the road surface with different attachment coefficients in each data subset presents a normal distribution trend, the normal distribution expectations and variances of the point sets of the road surfaces with different attachment coefficients are different, and the road surfaces with different attachment coefficients can be distinguished through the study of the differences; and then, by constructing a normal distribution probability function of the road surface model of each attachment coefficient in each data subset, when a real-time data point arrives, the attachment coefficient of the road surface model can be obtained through combining the traction force and the slip rate of the real-time data point and the normal distribution probability function.

Description

Road adhesion coefficient estimation method based on Bayes and electronic equipment
Technical Field
The invention belongs to the technical field of road adhesion coefficients, and particularly relates to a road adhesion coefficient estimation method based on Bayes and electronic equipment.
Background
Currently, the number of automobiles is increasing, and safety performance becomes the most important consideration part of automobiles. In the field of automatic driving which is more and more mature, the automobile has great requirements on the safety performance of the automobile, and also has strict requirements on parameters such as the road surface roughness, the road surface adhesion coefficient and the like, especially the adhesion coefficient. The braking distances of the automobile under different attached roads are different, and the braking distances under the road surfaces with smaller attachment coefficients are longer, so that when the automobile is driven to switch on the road surfaces with different attachment coefficients, particularly in the automatic driving field, the faster recognition of the switching and the recognition of the road surface with which attachment coefficient is switched are particularly important.
In the existing sensor-based road surface classification recognition and switching system, the direct recognition equipment of the sensor is limited by conditions such as installation environment, use environment, high price and the like, and in some special cases (such as severe weather such as heavy fog, heavy snow and the like), the recognition switching effect on the road surface is not obvious, and the switching speed cannot meet the actual requirement.
And the general algorithm is used for judging the data characteristics of the sampling points in a window mode through a least square method, and the real-time requirement of the high-speed running of the automobile on the road surface switching identification is very high, so that the actual requirement of the actual automobile in running can not be met. Other algorithms, such as fast filtering by a kalman filter, can meet the real-time requirement, and can obtain a result at the first time when the road surface is switched, but only judge that the road surface has been switched, but cannot calculate the attachment coefficient of the road surface with desirable accuracy after the switching, which cannot meet the actual requirement of the automatic driving field with strict requirements on the road surface condition.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a Bayes-based road adhesion coefficient estimation method and electronic equipment, which are used for rapidly identifying road switching and estimating road adhesion coefficients.
In order to achieve the above purpose, the invention provides a road adhesion coefficient estimation method based on Bayes, which comprises the following steps:
Acquiring a set of data points including slip rate and traction force;
Dividing the traction force into a plurality of subintervals for dividing the data point set into a plurality of data subsets; in the same data subset, the slip rates of the road surface models with different attachment coefficients obey different normal distributions;
Calculating the expected and variance of the normal distribution of the road surface model of each attachment coefficient in each data subset, and constructing a normal distribution probability function of the road surface model of each attachment coefficient according to the calculated expected and variance;
Collecting real-time data points, and selecting a corresponding data subset according to traction force of the real-time data points;
Calculating the probability of the real-time data point belonging to the road surface model of each attachment coefficient according to the slip rate of the real-time data point and combining the normal distribution probability function of the road surface model of each attachment coefficient in the selected data subset;
determining a pavement model to which the real-time data point belongs according to the maximum attribution probability;
and calculating the attachment coefficient of the current road surface model through a Kalman filter.
In some alternative embodiments, the slip ratio is the slip ratio on the left or right side of the vehicle during travel of the vehicle.
In some alternative embodiments, the slip ratio calculation formula for the left or right side of the vehicle is as follows:
where S represents the slip ratio of the vehicle on one side, W 1 and W 2 represent the angular speeds of the driving wheel and the driven wheel on the vehicle on one side, respectively, and R 1 and R 2 represent the tire radius of the driving wheel and the driven wheel on the vehicle on one side, respectively, wherein: one side is left or right.
In some alternative embodiments, the subintervals of traction are the same length.
In some alternative embodiments, after determining the road surface model to which the real-time data point belongs, determining whether a switch in the road surface model occurs; if yes, calculating the road surface model of each point in the plurality of data points collected next, counting the number of data points corresponding to each road surface model, and taking the road surface model with the largest number of data points as the current road surface model.
In some alternative embodiments, the plurality of data points is 22.
An electronic device comprising one or more processors and memory;
one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the Bayes-based road adhesion coefficient estimation method described above.
Compared with the prior art, the invention has the following advantages:
The invention determines the attribution of the road model through the collection of the traction force on one side and the data points of the slip rate of the automobile when the automobile runs, and accurately calculates the attachment coefficient of the road model. Instead of taking the whole data point set as a total sample set for learning, the invention divides the whole data point set into a plurality of data subsets, wherein the point set of the road surface with different attachment coefficients in each data subset presents a normal distribution trend, and the normal distribution expectations and variances of the point sets of the road surfaces with different attachment coefficients are different, and the road surfaces with different attachment coefficients can be distinguished through the learning of the differences; by constructing a normal distribution probability function of the road surface model of each attachment coefficient in each data subset, when real-time data points arrive, the time of road surface switching can be calculated by combining the traction force and the slip rate of the real-time data points and the normal distribution probability function to obtain the point set attribution result after the road surface switching, so that attribution of the road model is determined, and the attachment coefficient of the road surface model is obtained.
Drawings
Fig. 1 is a flowchart of a road adhesion coefficient estimation method based on Bayes according to an embodiment of the present invention.
FIG. 2 is a graph of a data point set of slip ratio versus traction provided by an embodiment of the present invention;
FIG. 3 is a time chart of the switching of the automobile from the high adhesion coefficient road to the low adhesion coefficient road according to the embodiment of the invention;
FIG. 4 is a normal distribution chart of slip rates of different subsets of data over a fixed traction range provided by an embodiment of the present invention;
fig. 5 is a graph of kalman calculation results from a road surface with a high adhesion coefficient to a road surface with a low adhesion coefficient according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
The invention mainly relates to how to process a sample set to obtain prior probability, and how to calculate posterior probability and identify road surface switching quickly by Bayes classification when a new data set arrives. How to calculate the road surface adhesion coefficient with a desirable accuracy by controlling the convergence speed of the kalman filter when the road surface switching occurs.
The road adhesion coefficient estimation method based on Bayes, which is disclosed by the embodiment of the invention, is shown in fig. 1, and comprises the following steps of:
S1, acquiring a set of data points comprising slip rate and traction force.
The basis of the Bayes classification of the present invention is that in the data subset, the characterization of the slip ratio of the data set will behave differently on different traction coefficient roadways when the range of traction is fixed. The road slip rates of different adhesion coefficients can be seen from the dot plot of the data set only, and are also because of the great difference between the high adhesion road and the low adhesion road, the working performance of the algorithm can be very good.
The road adhesion coefficient estimation algorithm based on the Bayes classifier depends on a data set of slip rate (S) of left side (XL) or right side (XR) of an automobile and traction force (F f) of the automobile in the running process of the automobile, and the principle of adhesion coefficient estimation is obtained through calculation of friction correlation between the difference of speeds of a driving wheel and a non-driving wheel and the slip.
Taking the front wheel as a driving wheel to calculate the slip rate at the left side as an example: wherein S XL is the left slip ratio, ω FLωRL is the left front wheel angular velocity and the left rear wheel angular velocity, respectively, R FLRRL is the tire radius of the left front wheel and the left rear wheel, respectively, and the calculation formula of S XL is:
The Bayes formula is applied to the data set of the invention as a precondition of Bayes classification, in the calculation of Bayes prior probability, the prior probability learning of subsets of the data set is based on the normal distribution of different adhesion coefficients with different expectations and variances, the expectations and variances of the normal distribution of each adhesion coefficient of each obtained subset are taken as prior probability, and the posterior probability is calculated according to the conditional probability for new data points, so that the attribution of the pavement model of the new data points is obtained.
As shown in fig. 2, a graph of slip ratio and traction force point set of the vehicle running on the road surface with different adhesion coefficients is shown, wherein the vertical axis is traction force (F f), and the horizontal axis is slip ratio S. And the experimental conditions were data point sets of the automobile from a dry asphalt pavement (high adhesion coefficient) to a crushed stone pavement (low adhesion coefficient), "·" represents a dry asphalt pavement and "×" represents a crushed stone pavement. As can be seen from the figure, the road surfaces with different traction coefficients show a significant difference in slip ratio under the same traction range. As the adhesion coefficient difference between two road surfaces is larger, the classification switching can be realized by adopting a Kalman filtering method. However, when the adhesion coefficients of two road surfaces are relatively close to each other, such as a dry asphalt road surface and a water-logging asphalt road surface, the effect of classification switching is poor only by adopting a Kalman filter, and the time for obtaining a data result is greatly increased, so that the requirement of a real system cannot be met.
According to the invention, firstly, the prior probability is obtained by learning the road surfaces with known different attachment coefficients, then the idea of a Bayes classifier is added, the data set coming later is classified, the posterior probability is calculated through the prior probability, so that the model belonging to the road surface which the data set comes belongs to is judged, and then the attachment coefficient of the road surface model can be accurately calculated through a Kalman filter, so that the switching identification of the road surface and the calculation of the attachment coefficient of the road surface model after switching in the running process of the automobile can be realized.
S2, dividing the traction force into a plurality of subintervals for dividing the data point set into a plurality of data subsets; within the same subset of data, the slip rates of road surface models of different adhesion coefficients are subject to different normal distributions.
In the invention, all data sets are not taken as integral learning samples, but are divided into data subsets taking traction force as a certain range, each subset is taken as a learning sample, the expected and variance of the normal distribution of the data sets of the road surfaces with different attachment coefficients in each subset are obviously different, and the prior probability of different data sets caused by the difference is learned.
In this experiment, the traction force range was set at 50 intervals, and the data sets of the road surfaces with different adhesion coefficients were learned in each interval. Through a large number of data experiments, a normal representation bar graph of the road surfaces with different attachment coefficients under the same traction force range is drawn, and when the traction force range is fixed, the slip rate of the road surfaces with different attachment coefficients is obviously characterized and shows a normal distribution trend, as shown in fig. 4. This is also the theoretical basis for the estimation of the road friction coefficient based on the Bayes classification.
S3, in each data subset, calculating expected and variance of normal distribution of the road surface model of each attachment coefficient, and constructing a normal distribution probability function of the road surface model of each attachment coefficient according to the calculated expected and variance;
By way of introduction, the distribution of road surface data sets of different traction coefficients under the same traction range exhibits a normal distribution, so the probability of each point in different road models is actually the probability of knowing the normal distribution of expectations and variances.
Constructing a normal distribution probability function of a pavement model of each attachment coefficient in each data subset:
Where x is the slip ratio s of the real-time data points, the mean and variance are the expected and variance of different traction ranges, and the road surface with each attachment coefficient under the same traction is one.
S4, acquiring real-time data points, and selecting a corresponding data subset according to traction force of the real-time data points.
When a real-time data point arrives, a normal distribution probability function of the road surface model with different attachment coefficients under a corresponding traction range is selected according to the traction of the real-time data point, and then the attribution probability is calculated.
As can be seen from fig. 4, the expected normal distribution and variance of the road surface with different adhesion coefficients are obviously different, and by learning the normal distribution, the road surface model under the fixed traction force range can be obtained, when a new data point arrives, the traction force range interval to which the new data point belongs is determined according to the traction force, and the posterior probability of the new point set under each road model is calculated in the interval. When the probability of belonging to a certain model is maximum, the new point can be considered to belong to the road model. The calculation of the posterior probability has been described in the embodiment, and the probability of each new data point is actually the probability of normal distribution, and the normal distribution expectations and variances of the road surfaces with different attachment coefficients are known, so the calculation of the posterior probability is also reliable.
S5, calculating the probability of the real-time data point belonging to the road surface model of each attachment coefficient according to the slip rate of the real-time data point and combining the normal distribution probability function of the road surface model of each attachment coefficient in the selected data subset;
The prior probability and the posterior probability of the Bayes probability are calculated, firstly, the learning set mentioned in the summary of the invention is divided into a plurality of subsets which are divided into all data sets, all the data sets are point set plane diagrams of automobile traction force (F f) and slip rate (S) on the left side or right side of the automobile, namely, the abscissa of each data point set is slip rate, and the ordinate is traction force.
The new point is designated a, the different traction ranges are designated F 1、F2、F3, the road surfaces with different traction coefficients at the same traction range are designated B 1、B2、B3, the respective probabilities at the traction range F 1 according to Bayes' formula:
P1＝P(A|B1)
P2＝P(A|B2)
P3＝P(A|B3)
Namely, in the subset under the same traction force range, the probability of a new data point under the traction force range can be calculated through Bayes conditional probability, and the road model attribution of the new data point is recorded as C, and then:
C＝max(P1,P2,P3)
In this way, it is possible to obtain which road model the new data point belongs to. In a specific probability calculation, the conditional probability can be obtained by Bayes formula as follows:
In the invention, the probability of the occurrence of roads with different attachment coefficients and new data points are mutually independent, so the formula can be modified into:
Because the normal distribution probability of the road surfaces with different attachment coefficients is different, the attribution probability calculated by the same data point under the road surfaces with different attachment coefficients is different.
S6, determining a pavement model to which the real-time data points belong according to the maximum attribution probability;
While in the present invention it is considered that the probability of occurrence at each traction interval is equal when a new data point arrives. And if the maximum value of the calculated posterior probability exists in one road model under one traction force interval, the data point at the moment is considered to be positioned in the road model.
S7, judging whether the pavement model is switched or not; if yes, calculating the road surface model of each point in the plurality of data points collected next, counting the number of data points corresponding to each road surface model, and taking the road surface model with the largest number of data points as the current road surface model.
After determining to which road model the new data point belongs, it is far from sufficient to rely on only one point to determine if the car has switched to another road, which is a very low fault tolerance. In actual driving of a car, there is also a case where in a road surface with a relatively constant adhesion coefficient, the adhesion coefficient is suddenly changed due to a loss of the road surface or due to some foreign objects (such as liquid being splashed on the road), and if the road surface model is judged by data points when the car passes through the place, erroneous judgment is generated, and the judgment result of the whole system is affected. The solution to this phenomenon is to add a decision logic for the start of the road switching.
When the new data point is calculated by posterior probability, the judgment result is that the road model which exists in the last point is not obtained, namely the road model is possibly switched, the judgment logic for starting the switching of the road surface is activated, and when the road surface is judged to be switched by the new data points, the attachment coefficient of the road model after the switching can be rapidly and accurately calculated by adjusting the Kalman filtering measurement noise covariance matrix.
The invention makes a strict test on the judgment logic for starting switching of the activated pavement, and finally takes 22 new points after the activation of the judgment logic as the basis for judging attribution. The logic is specifically as follows: the probability of 22 consecutive points after the activation of the judgment logic was obtained by Bayes' calculation of the posterior probability, and it was noted that there were a 1 on road surface a, B 1 on road surface B, and C 1 on road surface C (here, 3 road surfaces are illustrated). The probability of belonging to road surface A is P A, the probability of belonging to road surface B is P B, and the probability of belonging to road surface C is P C, and the following can be obtained:
A1+B1+C1＝22
P＝max(P(A1),P(B1),P(C1))
The maximum probability of the road surface with 22 points belonging to various attachment coefficients under the traction force range can be obtained, the automobile is considered to run on the road model at the moment, if the road model calculated at the moment is different from the road model calculated at the moment at the last time, the road surface is considered to be switched, and then the attachment coefficient of the road surface model after the switching can be rapidly and accurately calculated by adjusting a Kalman filtering measurement noise covariance matrix.
As shown in fig. 3, the time performance of the automobile when the automobile is switched from the high adhesion coefficient road to the low adhesion coefficient road is shown, 0 represents that the automobile is on a dry asphalt road, 1 represents that the automobile is on a sand road, the automobile starts to enter the sand road when the 1670 th data point arrives, the derived data point can be considered as the 1690 th data point arrives when the pavement switching is recognized, the speed interval is 20 data points, the period of the system acquisition data point is 10ms, and the time for recognizing the pavement switching is 0.2s, which can meet the requirement of the actual system on time.
S8, calculating the adhesion coefficient of the current road surface model through a Kalman filter.
The Kalman filtering based on the invention is an algorithm for optimally estimating the system state by utilizing a linear system state equation and through system input and output observation data. The optimal estimate can also be seen as a filtering process, since the observed data includes the effects of noise and interference in the system.
The Kalman filtering algorithm of the invention is mainly divided into two steps: prediction and updating.
And (3) predicting: estimating the state of the current moment (moment k) according to the posterior estimated value of the last moment (moment k-1) to obtain the prior estimated value of the moment k;
updating: correcting the estimated value of the prediction stage by using the measured value of the current moment to obtain the posterior estimated value of the current moment.
The kalman filter can be divided into a time update equation and a measurement update equation. The time updating equation (i.e. the prediction stage) calculates a state variable prior estimated value and an error covariance prior estimated value at the current moment according to the state estimated value at the previous moment; the measurement update equation (i.e., the update phase) is responsible for combining the a priori estimates with the new measured variables to construct an improved a posteriori estimate. The time update equation and the measurement update equation are also called a prediction equation and a correction equation. The kalman algorithm is therefore a recursive predictive-corrective approach.
The Kalman filter time update equation is as follows:
The Kalman filter state update equation is as follows:
The meaning of each parameter in the Kalman filter equation is as follows:
1、 and/> The posterior state estimation values respectively representing the k-1 moment and the k moment are one of the filtering results, namely the updated result, and are also called optimal estimation. In this embodiment/>And/>Respectively representing the optimal estimated value of the road surface adhesion coefficient at the last moment and the optimal estimated value of the road surface adhesion coefficient at the current moment.
2、The a priori state estimate at time k is the result of the intermediate calculation of the filtering, i.e. the result at time k predicted from the optimal estimate at the previous time (time k-1), is the result of the prediction equation. In this embodiment/>Represents the optimal estimate/>, of the road adhesion coefficient by the last momentAnd obtaining the predicted value of the road adhesion coefficient at the current moment.
3. P k-1 and P k: the a posteriori estimated covariance (i.e.And/>The covariance of (c) representing the uncertainty of the state), is one of the results of the filtering. In this embodiment, P k-1 and P k represent estimated road surface adhesion coefficient values/>, at the previous timeCovariance of (2) and road adhesion coefficient estimation value at the current time/>Is a covariance of (c).
4、Prior estimated covariance at time k (/ >Is the covariance of the filter). In this embodiment/>The covariance of the predicted road surface adhesion coefficient value at the present time is expressed.
5. H: the method is a conversion matrix from state variables to measurement (observation), and represents the relation between the state and the observation, wherein the Kalman filtering is linear, and the Kalman filtering is responsible for converting the measured value in m dimension into n dimension so as to enable the measured value to conform to the mathematical form of the state variables, and is one of preconditions of the filtering.
In this embodiment the traction and slip ratio approximately satisfy the following linear relationship:
where subscripts l and r represent left and right sides of the vehicle, s (t) represents slip ratio, and u (t) represents normalized traction.
And/>Representing the slopes of the left and right sides of the linear model, respectively, and δ l (t) and δ r (t) representing the intercepts of the left and right sides of the linear model, respectively. The two equations are written in vector form, and the deformation is as follows:
H (t) can be obtained according to the above formula:
6. z k: the measured value (observed value) is the input of the filtering. The measurement values are represented by slip ratios in this embodiment.
7. K k: the filter gain matrix is an intermediate calculation result of the filtering, a Kalman gain, or a Kalman coefficient. By measuring z k (slip ratio) pairs(Predicted road surface adhesion coefficient value) is corrected.
8. A: the state transition matrix is actually a guess model for the state transition of the object. In this embodiment, it is assumed that the road surface adhesion coefficient on the same road surface is substantially unchanged, that is, the road surface adhesion coefficient at the present time is equal to the road surface adhesion coefficient at the last time.
9. Q: process excitation noise covariance (covariance of the system process). This parameter is used to represent the error between the state transition matrix and the actual process, i.e. the noise introduced by the prediction model itself.
10. R: the noise covariance is measured. When the filter is actually implemented, the measurement noise covariance R is generally observed and is a known condition of the filter. In this embodiment, the convergence rate of the kalman filter is controlled by dynamically adjusting the size of R according to the Bayes classification result.
12、The residual errors of actual observation and prediction observation are corrected a priori (predicted) together with the Kalman gain to obtain a posterior. The estimated value of the road surface adhesion coefficient is corrected by this residual.
The Kalman filter can realize accurate calculation of road adhesion coefficients.
As shown in FIG. 5, the graph depicts the Kalman calculation result from a road surface with a high adhesion coefficient to a road surface with a low adhesion coefficient, the former being a dry asphalt road surface, the actual adhesion coefficient being about 0.8, and the latter being a crushed stone road surface, the actual adhesion coefficient being about 0.4. As can be seen from the figure, the road adhesion coefficient is initially stabilized at 200, and after the road switching occurs, the calculated adhesion coefficient of the new road model is 95, where the road adhesion coefficient is obtained by multiplying a certain magnification in the algorithm.
The road friction coefficient estimation algorithm based on Bayes classification has the advantages that the classification effect and the classification time are both dependent on the learning condition of the slip rate under traction in a range, namely, the prior probability is dependent, the finer the road model is, the more the data set is, and the more accurate the posterior probability is obtained through the prior probability. The invention needs to rely on the study of the data sets of the road surfaces with different attachment coefficients in advance, and then calculates the prior probability in each data set for storage.
The invention realizes the switching identification of the road surface meeting the actual real-time requirement, and can realize the identification attribution function with the required precision of the switched road surface. The algorithm is not influenced by factors such as environment, so that the algorithm has good compatibility and workability. And the Bayes classification based on the algorithm has the advantages that the posterior probability calculation result is related to the prior probability, the prior probability sample set is enough, the classification is accurate enough, the result reliability obtained by the Bayes formula is very high, and the advantages have great significance in the automobile driving field, especially in the automatic driving field.
Finally, the invention also provides an electronic device, which is characterized by comprising one or more processors and a memory; one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the Bayes-based road adhesion coefficient estimation method described above.
It should be noted that each step/component described in the present application may be split into more steps/components, or two or more steps/components or part of operations of the steps/components may be combined into new steps/components, according to the implementation needs, to achieve the object of the present application.
It will be readily appreciated by those skilled in the art that the foregoing is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims (7)

1. The road adhesion coefficient estimation method based on Bayes is characterized by comprising the following steps of:
Acquiring a set of data points including slip rate and traction force;
Dividing the traction force into a plurality of subintervals for dividing the data point set into a plurality of data subsets; in the same data subset, the slip rates of the road surface models with different attachment coefficients obey different normal distributions;
Calculating the expected and variance of the normal distribution of the road surface model of each attachment coefficient in each data subset, and constructing a normal distribution probability function of the road surface model of each attachment coefficient according to the calculated expected and variance;
Collecting real-time data points, and selecting a corresponding data subset according to traction force of the real-time data points;
Calculating the probability of the real-time data point belonging to the road surface model of each attachment coefficient according to the slip rate of the real-time data point and combining the normal distribution probability function of the road surface model of each attachment coefficient in the selected data subset;
determining a pavement model to which the real-time data point belongs according to the maximum attribution probability;
Calculating an adhesion coefficient of the current road surface model through a Kalman filter, wherein the traction force and the slip rate meet the following linear relation:
where subscripts l and r represent left and right sides of the vehicle, s (t) represents slip ratio, and u (t) represents normalized traction; And/> Representing the slopes of the left and right sides of the linear model, respectively, and δ l (t) and δ r (t) representing the intercepts of the left and right sides of the linear model, respectively.
2. The Bayes-based road adhesion coefficient estimating method according to claim 1, wherein the slip ratio is a slip ratio of the left side or the right side of the automobile during running of the automobile.
3. The Bayes-based road surface attachment coefficient estimating method according to claim 2, wherein the slip ratio calculation formula of the left side or the right side of the vehicle is as follows:
where S represents the slip ratio of the vehicle on one side, W 1 and W 2 represent the angular speeds of the driving wheel and the driven wheel on the vehicle on one side, respectively, and R 1 and R 2 represent the tire radius of the driving wheel and the driven wheel on the vehicle on one side, respectively, wherein: one side is left or right.
4. The Bayes-based road attachment coefficient estimation method according to claim 1, wherein the subintervals of traction are the same length.
5. The Bayes-based road adhesion coefficient estimating method according to claim 1, wherein after determining the road model to which the real-time data point belongs, judging whether the road model is switched; if yes, calculating the road surface model of each point in the plurality of data points collected next, counting the number of data points corresponding to each road surface model, and taking the road surface model with the largest number of data points as the current road surface model.
6. The Bayes-based road adhesion coefficient estimating method according to claim 5, wherein the plurality of data points is 22.
7. An electronic device comprising one or more processors and memory;
One or more programs stored in memory and configured to be executed by the one or more processors, the one or more programs configured to perform the Bayes-based road adhesion coefficient estimation method of any one of claims 1-6.
CN202111140257.3A 2021-09-28 2021-09-28 Road adhesion coefficient estimation method based on Bayes and electronic equipment Active CN113911124B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111140257.3A CN113911124B (en) 2021-09-28 2021-09-28 Road adhesion coefficient estimation method based on Bayes and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111140257.3A CN113911124B (en) 2021-09-28 2021-09-28 Road adhesion coefficient estimation method based on Bayes and electronic equipment

Publications (2)

Publication Number Publication Date
CN113911124A CN113911124A (en) 2022-01-11
CN113911124B true CN113911124B (en) 2024-04-30

Family

ID=79236593

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111140257.3A Active CN113911124B (en) 2021-09-28 2021-09-28 Road adhesion coefficient estimation method based on Bayes and electronic equipment

Country Status (1)

Country Link
CN (1) CN113911124B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102007060652A1 (en) * 2007-12-15 2009-06-18 Bayerische Motoren Werke Aktiengesellschaft Device for estimating state of roadway, has arithmetic device, which is arranged for determining state of roadway, and is coupled with measuring device, where driving dynamic values are predicted
CN103407451A (en) * 2013-09-03 2013-11-27 东南大学 Method for estimating longitudinal adhesion coefficient of road
CN104182991A (en) * 2014-08-15 2014-12-03 辽宁工业大学 Vehicle running state estimation method and vehicle running state estimation device
CN104325980A (en) * 2014-10-16 2015-02-04 北京汽车股份有限公司 Attachment coefficient estimation method and device
CN111845775A (en) * 2020-07-20 2020-10-30 上海大学 Joint estimation method for driving state and inertia parameters of distributed driving electric automobile
CN112319490A (en) * 2019-07-31 2021-02-05 广州汽车集团股份有限公司 Method for estimating mass center slip angle of automobile, computer equipment and storage medium
CN113183973A (en) * 2021-05-11 2021-07-30 武汉理工大学 Tire pressure monitoring and road surface information intelligent sensing platform and method based on CAN network

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120050451A (en) * 2009-07-17 2012-05-18 콘티넨탈 엔지니어링 서비시스 게엠베하 Laser-based method for the friction coefficient classification of motor vehicles
DE102016208675A1 (en) * 2016-05-19 2017-11-23 Lucas Automotive Gmbh Method for determining a safe speed at a future waypoint

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102007060652A1 (en) * 2007-12-15 2009-06-18 Bayerische Motoren Werke Aktiengesellschaft Device for estimating state of roadway, has arithmetic device, which is arranged for determining state of roadway, and is coupled with measuring device, where driving dynamic values are predicted
CN103407451A (en) * 2013-09-03 2013-11-27 东南大学 Method for estimating longitudinal adhesion coefficient of road
CN104182991A (en) * 2014-08-15 2014-12-03 辽宁工业大学 Vehicle running state estimation method and vehicle running state estimation device
CN104325980A (en) * 2014-10-16 2015-02-04 北京汽车股份有限公司 Attachment coefficient estimation method and device
CN112319490A (en) * 2019-07-31 2021-02-05 广州汽车集团股份有限公司 Method for estimating mass center slip angle of automobile, computer equipment and storage medium
CN111845775A (en) * 2020-07-20 2020-10-30 上海大学 Joint estimation method for driving state and inertia parameters of distributed driving electric automobile
CN113183973A (en) * 2021-05-11 2021-07-30 武汉理工大学 Tire pressure monitoring and road surface information intelligent sensing platform and method based on CAN network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
汽车主动安全关键参数联合估计方法;宋翔;李旭;张为公;蔡凤田;吴明明;;交通运输工程学报;20140215(第01期);第66-73页 *

Also Published As

Publication number Publication date
CN113911124A (en) 2022-01-11

Similar Documents

Publication Publication Date Title
US11608040B2 (en) Traction control based on friction coefficient estimation
CN110095635B (en) Longitudinal speed estimation method of all-wheel drive vehicle
CN111605559B (en) Vehicle mass estimation method, torque control method and device
EP2927066B1 (en) Model-based longitudinal stiffness estimation system and method
EP2927065B1 (en) Road surface friction and surface type estimation system and method
CN104360282B (en) A kind of variable length sliding window recognizes the battery charge state method of estimation of battery parameter
JP5214738B2 (en) Vehicle load estimation method
CN101844561B (en) Road surface condition identification based on statistical pattern recognition
EP2694344B1 (en) Estimation of road inclination
CN110475697B (en) System and method for calibrating vehicle tires
CN110435647B (en) TTC vehicle safety anti-collision control method based on rolling optimization parameters
KR100828778B1 (en) Estimating method for mass of vehicle
CN113544597A (en) Method for optimizing control signals for operating a vehicle
US6711508B2 (en) Method and apparatus for estimating tire air pressure
CN113911124B (en) Road adhesion coefficient estimation method based on Bayes and electronic equipment
CN110626326A (en) Method and system for determining tire-road friction in a vehicle
EP4197821A1 (en) Tire stiffness estimation system and method
CN107719373B (en) Method and system for estimating road friction
JP2009154675A (en) Driver state estimation device and program
CN113619587B (en) Road adhesion coefficient estimation method based on Bayes classifier
Galanis et al. Weather-based road condition estimation in the era of Internet-of-Vehicles (IoV)
CN112406560A (en) Method and device for adjusting generated power of range extender, storage medium and vehicle
CN113954841A (en) Method and system for monitoring wheel end traction force in real time and electronic equipment
CN116691706A (en) Vehicle lateral speed estimation method and device
Lex et al. On-board determination of the friction coefficient between tire and road using standard-application vehicle dynamics sensors

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
GR01 Patent grant
GR01 Patent grant