CN110516311B - Comprehensive compensation strategy construction method for constant error of vehicle acceleration sensor - Google Patents
Comprehensive compensation strategy construction method for constant error of vehicle acceleration sensor Download PDFInfo
- Publication number
- CN110516311B CN110516311B CN201910700562.XA CN201910700562A CN110516311B CN 110516311 B CN110516311 B CN 110516311B CN 201910700562 A CN201910700562 A CN 201910700562A CN 110516311 B CN110516311 B CN 110516311B
- Authority
- CN
- China
- Prior art keywords
- acceleration sensor
- vehicle
- error
- constant
- constant error
- 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
Links
Images
Landscapes
- Steering Control In Accordance With Driving Conditions (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
The invention discloses a comprehensive compensation strategy construction method aiming at constant errors of an acceleration sensor for a vehicle, which comprises the steps of firstly establishing a dynamic expression about lateral and yaw motions of the vehicle, and realizing the observation of the lateral speed of the vehicle; secondly, determining the state quantity and the quantity measurement of an estimator according to an error drift model of the acceleration sensor and the transverse and longitudinal kinematic relations of the vehicle, and estimating the constant error of the acceleration sensor by using a Kalman filter; and finally, completing compensation and updating of the constant error of the acceleration sensor by correcting the tire positive pressure model and the tire cornering stiffness and feeding back the input information of the vehicle lateral speed observer. The method can realize the estimation and iterative update of the constant error of the vehicle acceleration sensor at a higher frequency, does not lose too much real-time property while improving the overall estimation precision of the vehicle motion state, and has the characteristics of simplicity and high efficiency.
Description
Technical Field
The invention belongs to the field of error estimation and compensation of vehicle sensors, and particularly relates to a comprehensive compensation strategy construction method for a constant error of a vehicle acceleration sensor, aiming at further improving the accuracy and the real-time property of vehicle driving state estimation.
Background
With the development of active safety technology of vehicles, the intelligent sensing theory has become one of the important research hotspots. For the vehicle, the anti-lock system needs to judge the slip degree of the wheels according to the output of the wheel speed sensor, and the electronic stability control program determines the motion mode of the vehicle according to the signal of the yaw rate sensor to realize the direct yaw moment control. However, the output signals of these conventional sensors all contain unavoidable noise, and the output of some other sensors (such as an acceleration sensor for a vehicle) may have a phenomenon that an error drifts with time. These noises and drifts amplify the error of the output signal during the time domain integration of the sensor signal.
Disclosure of Invention
In order to solve the problems in the prior art and reduce the precision loss caused by the error drift of the sensor, the invention provides a comprehensive compensation strategy construction method aiming at the constant error of the vehicle acceleration sensor.
The technical purpose is achieved through the following technical scheme.
A comprehensive compensation strategy construction method for constant errors of an acceleration sensor for a vehicle comprises the steps of constructing a lateral speed observer for data preprocessing of constant error estimation of the acceleration sensor, obtaining an observation result of the lateral speed of the vehicle, achieving online estimation of the constant errors of the acceleration sensor according to the observation result and a fused state equation, achieving correction of tire side deflection rigidity through constructing a tire positive pressure error compensation model, and finally feeding the correction result back to the lateral speed observer to complete compensation and updating of the constant errors of the acceleration sensor.
Further, the state equation adopted for estimating the constant value error fuses a vehicle kinematic model and an acceleration sensor error model.
Further, the state equation is:
wherein a is x,mes 、a y,mes As the measurement output of the acceleration sensor, a x,bias 、a y,bias Is a constant error of the acceleration sensor, w 1 、w 2 、w 3 、w 4 Are respectively white Gaussian noise, v x Is the longitudinal speed, v, of the vehicle y Is the lateral velocity of the vehicle and r is the yaw rate of the vehicle.
Further onAnd the quantity of the selected quantity of the state equation is measured
v' x For the longitudinal speed of the driven wheel speed sensor>
Is an observation of lateral velocity.
Further, the cornering stiffness C σ With positive pressure F to which four tyres are subjected z The relationship between them is: c σ ＝a 1 sin[a 2 arctan(a 3 F z )]Wherein a is 1 、a 2 、a 3 Are parameters to be fitted.
Further, the positive pressure is calculated based on a correction result of the acceleration signal, which is determined by a constant error of the acceleration sensor.
Further, the result of the correction of the acceleration signal
Satisfies the following conditions: />
Wherein a is x,mes 、a y,mes Is the measurement output of the acceleration sensor, and>
and the optimal estimation value of the constant error of the acceleration sensor is obtained.
Further, the error compensation model of the positive pressure is as follows:
wherein m is the vehicle body mass, i = { F, R }, F denotes the front wheels, R denotes the rear wheels; j = { L, R }, L representing the left wheel, R representing the right wheel; h is the vertical height of the mass center of the automobile from the ground, and l = l f +l r Is the distance between the front and rear axles of the vehicle, b F 、b R The front and rear wheel tracks of the automobile.
The invention has the beneficial effects that:
1. the method disclosed by the invention integrates the vehicle kinematic model while modeling the sensor error, constructs the lateral speed observer, and measures the lateral speed obtained by the lateral speed observer as the quantity of the integrated model, so that the updating speed of the constant error of the acceleration sensor is increased; the method improves the overall accuracy of the estimation of the motion state of the vehicle, does not lose instantaneity excessively, has the characteristics of simplicity and high efficiency, and can well track the error drift of the acceleration sensor.
2. According to the optimal estimated value of the constant error of the acceleration sensor, the compensation and the correction of the constant error of the acceleration sensor for the vehicle can be realized, so that accurate information is provided for the measurement of the motion state of the vehicle and other vehicle control schemes.
Drawings
Fig. 1 is a flow chart of an error compensation strategy.
Detailed Description
The embodiments of the present invention will be further described with reference to the accompanying drawings, but the scope of the invention is not limited thereto.
As shown in fig. 1, a method for constructing a comprehensive compensation strategy for a constant error of an acceleration sensor for a vehicle includes the following steps:
step (1), observing the lateral speed of the vehicle
The vehicle adopts a front wheel driving mode, the center of mass of the vehicle is specified as the origin of a vehicle body coordinate system, the horizontal forward direction is the positive direction of an x axis, the horizontal leftward direction is the positive direction of a y axis, the vertical horizontal plane upwards is the positive direction of a z axis, and all the revolution angles and moments are positive in the anticlockwise direction in the horizontal plane.
First, the kinetic equations of the vehicle with respect to lateral and yaw motion are determined:
wherein v is y R are the lateral velocity, yaw rate, C of the vehicle, respectively f 、C r Yaw stiffness of the front and rear wheels of the vehicle, delta being the angle of rotation of the front wheel of the vehicle, l f 、l r Respectively the distance of the vehicle mass center from the front axle and the rear axle, m is the vehicle body mass, I z Is the moment of inertia, v, of the vehicle about the z-axis x Is the longitudinal speed of the vehicle, an
R eff Representing the effective radius, omega, of the tyre 3 And ω 4 The wheel speeds of the left rear wheel and the right rear wheel are respectively.
And secondly, designing a Longbeige observer as data preprocessing of constant error estimation of the acceleration sensor to realize observation of the lateral speed of the vehicle.
The real vehicle lateral speed observation model is assumed as follows:
wherein the state quantity X = [ v = y r] T Control quantity U = δ, quantity measurement Y = r,
vehicle dynamics model matrix for lateral and yaw motion
Vehicle dynamics control matrix for lateral and yaw motion
Vehicle dynamics measurement matrix C = [0 1 ] for lateral and yaw motion]。
The following structure of the Roeberg observer is adopted:
wherein, L = [ k = 1 k 2 ] T A pole configuration for vehicle lateral velocity observation.
In order to make the error between the real value and the estimated value of the lateral velocity observation converge to 0 quickly, the equation (2) and the equation (3) are subtracted to obtain the state equation of the error:
the pole configuration parameter k required by the observation of the lateral speed can be determined by making the characteristic value of the matrix A-LC less than zero 1 And k 2 ，k 1 、k 2 The selection of (1) needs to consider the convergence rate of the state error under the condition that the characteristic value of the matrix A-LC is less than zero.
Step (2), establishing an error model of the acceleration sensor
Wherein, a x,mes 、a y,mes As the measurement output of the acceleration sensor, a x 、a y Is the longitudinal and lateral acceleration of the vehicle, a x,bias 、a y,bias Is a constant error of the acceleration sensor, w 1 、w 2 、w 3 、w 4 Respectively gaussian white noise. In this embodiment, the acceleration sensor error model is considered to be a random walk model approximately, that is:
and (3) estimating the constant error of the acceleration sensor by using a Kalman filter
Establishing a kinematic model about the longitudinal direction and the lateral direction of the vehicle, namely:
the error model of the acceleration sensor and the kinematic model of the vehicle are fused (hereinafter referred to as a fusion model), that is, the error model of the sensor is substituted into the kinematic model of the vehicle about the longitudinal direction and the lateral direction, so as to obtain the following state equation:
selecting a state quantity x = [ v ] x a x,bias v y a y,bias ] T Control amount u = [ a ] x,mes a y,mes ] T Measurement of quantity
Wherein->
Is observed as lateral velocity and has longitudinal velocity v 'of driven wheel speed sensor' x ＝v x The state equation and the metrology equation can be further expressed as:
wherein model state matrices are fused
Fusion model control matrix>
Measurement matrix for fusion model>
White gaussian noise->
The vehicle acceleration constant error estimation based on the Kalman filtering algorithm comprises the following steps:
1) Time update
And (3) carrying out one-step prediction on the state quantities such as the longitudinal and lateral speeds of the vehicle and the constant deviation of the longitudinal and lateral accelerations of the vehicle:
x k|k-1 ＝Φ k-1 x k-1|k-1 +Gu k-1 (10)
wherein: x is the number of k|k-1 For the fusion model, the state at time k-1 is transferred to the predicted state at time k using the state equation k-1 The control quantity at the moment of fusing the model k-1 is obtained;
calculating an error covariance prediction value of the state quantity:
P k|k-1 ＝Φ k-1 P k-1|k-1 Φ k-1 T +Q (11)
where Q is the noise covariance of the estimation process, P k-1|k-1 For the optimal covariance, P, at the moment of fusing the model state quantities k-1 k|k-1 The prediction covariance of the model state quantity at the moment k is fused, and a state transition matrix is provided:
Φ k-1 ＝exp(F k-1 T t )≈I+F k-1 T t (12)
wherein, T t For updating the step size of the Kalman Filter Algorithm, F k-1 The model state matrix is fused at the moment of k-1, I is a unit matrix, and the dimensionality is consistent with the dimensionality of the state quantity x.
2) Measurement update
Determining the kalman gain of the state quantity estimation:
K k ＝P k|k-1 H k T (H k P k|k-1 H k T +R) -1 (13)
wherein H k A measurement matrix at the moment k is obtained, and R is the covariance of measurement noise;
and (3) updating the state quantities such as the longitudinal and lateral speeds of the vehicle, and the constant deviation of the longitudinal and lateral accelerations of the vehicle:
x k|k ＝x k|k-1 +K k (y k -H k x k|k-1 ) (14)
wherein, y k And the measured value of the state at the moment k comprises a longitudinal speed and a lateral speed, wherein the longitudinal speed is provided by a wheel speed sensor, and the lateral speed is provided by a Roeberg observer.
Updating the error covariance of the state quantities:
P k|k ＝(I-K k H k )P k|k-1 (I-K k H k ) T +K k RK k T (15)
considering that the processing of the noise characteristics by the kalman filter is limited to the gaussian distribution, it is necessary to comprehensively refer to the accuracy of the established model (equation (8)) in which the information constituting the quantity measurement is derived from the longitudinal velocity v 'of the driven wheel speed sensor and the accuracy level of the sensor when selecting the initial value of the noise covariance' x And observed value of lateral velocity
The Kalman filtering is used for continuously predicting and correcting constant error state quantities based on an error model of the sensor and longitudinal and lateral kinematics models of the vehicle, so that accurate constant errors of the acceleration sensor are obtained finally.
Step (4) determining the correction result of the acceleration signal according to the estimated constant error of the acceleration sensor
Wherein, the first and the second end of the pipe are connected with each other,
and obtaining the optimal estimation value of the constant error of the acceleration sensor by the fusion model estimation.
Step (5), according to the accelerationThe correction result of the signals calculates the positive pressure F received by the four tires zij
According to the formula (17), the compensated tire positive pressure error F zij,bias The compensation model is as follows:
wherein g is the acceleration of gravity, i = { F, R }, F represents the front wheel, R represents the rear wheel; j = { L, R }, L representing the left wheel, R representing the right wheel; h is the vertical height of the mass center of the automobile from the ground, and l = l f +l r Distance between front and rear axles of the vehicle, b F 、b R The front and rear wheel tracks of the automobile.
Step (6), calculating the cornering stiffness of the front and rear wheels
Since the load of the wheel is time-varying during driving, the cornering stiffness of the corresponding tire is also varied, the cornering stiffness C of the tire being σ And a positive pressure F z The relationship between them is as follows:
C σ ＝a 1 sin[a 2 arctan(a 3 F z )] (19)
wherein, a 1 、a 2 、a 3 As the parameters to be fitted, a Tester module in CarSim software can be used for data fitting;
combined formula (17) for determining cornering stiffness C of front and rear wheels in a vehicle model f 、C r ：
Wherein, C σFL 、C σFR 、C σRL 、C σRR The cornering stiffnesses of the front left, front right, rear left and rear right tires of the vehicle, respectively.
And (7) feeding the calculated tire cornering stiffness back to the Roeberg observer in the step (1), wherein the Roeberg observer is a vehicle dynamic model about lateral and yaw motion, the model comprises tire parameters, namely cornering stiffness, the corrected tire cornering stiffness is used as the tire cornering stiffness required by the Roeberg observer by estimating a constant error of acceleration, and the compensation and updating of the constant error of the acceleration sensor are completed, so that the observation of the lateral speed of the vehicle are updated.
And mutual iteration is performed between the lateral speed observation based on the LongBege observer and the constant error estimation of the acceleration sensor based on Kalman filtering to form a complementary closed loop structure, and finally the estimation precision of the motion state of the vehicle is gradually improved.
The invention has been described in detail, and it is within the scope of the invention to adopt the concept and working method of the invention to make simple modifications, or to make improvements and decorations without changing the principle of the main concept of the invention.
Claims (6)
1. A comprehensive compensation strategy construction method for constant errors of an acceleration sensor for a vehicle is characterized in that a lateral speed observer is constructed to be used as data preprocessing of constant error estimation of the acceleration sensor, an observation result of the lateral speed of the vehicle is obtained, online estimation of the constant errors of the acceleration sensor is achieved according to the observation result and a fused state equation, correction of tire lateral deflection rigidity is achieved through construction of a tire positive pressure error compensation model, and finally the correction result is fed back to the lateral speed observer to complete compensation updating of the constant errors of the acceleration sensor;
the state equation is:
wherein a is x,mes 、a y,mes As the measurement output of the acceleration sensor, a x,bias 、a y,bias Is a constant error of the acceleration sensor, w 1 、w 2 、w 3 、w 4 Are respectively white Gaussian noise, v x Is the longitudinal speed, v, of the vehicle y Is the lateral velocity of the vehicle, r is the yaw rate of the vehicle;
said tire positive pressure error F zij,bias The compensation model is as follows:
wherein m is the vehicle body mass, i = { F, R }, F denotes the front wheels, R denotes the rear wheels; j = { L, R }, L representing the left wheel, R representing the right wheel; h is the vertical height of the mass center of the automobile from the ground, and l = l f +l r Distance between front and rear axles of the vehicle, b F 、b R Front and rear wheel tracks of the vehicle, l f 、l r Respectively the distance between the center of mass of the automobile and the front axle and the distance between the center of mass of the automobile and the rear axle.
2. The method for constructing the comprehensive compensation strategy for the constant error of the vehicular acceleration sensor according to the claim 1, characterized in that the estimation of the constant error adopts a state equation which fuses a vehicle kinematic model and an acceleration sensor error model.
3. The method for constructing the comprehensive compensation strategy for the constant error of the vehicular acceleration sensor according to claim 1, characterized in that the state equation selection quantity measurement method
v′ x The longitudinal speed of the driven wheel speed sensor is,
is an observation of lateral velocity.
4. The method for constructing the comprehensive compensation strategy for the constant error of the vehicular acceleration sensor according to claim 1, characterized in that the cornering stiffness C is σ With positive pressure F to which four tyres are subjected z The relationship between them is: c σ ＝a 1 sin[a 2 arctan(a 3 F z )]Wherein a is 1 、a 2 、a 3 Are parameters to be fitted.
5. The method for constructing the comprehensive compensation strategy for the constant error of the acceleration sensor for the vehicle as claimed in claim 4, wherein the positive pressure is calculated according to the correction result of the acceleration signal, and the correction result of the acceleration signal is determined by the constant error of the acceleration sensor.
6. The method for constructing the comprehensive compensation strategy for the constant error of the vehicular acceleration sensor according to claim 5, characterized in that the correction result of the acceleration signal
Satisfies the following conditions:
wherein a is x,mes 、a y,mes Is the measurement output of the acceleration sensor,
and the optimal estimated value of the constant error of the acceleration sensor is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910700562.XA CN110516311B (en) | 2019-07-31 | 2019-07-31 | Comprehensive compensation strategy construction method for constant error of vehicle acceleration sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910700562.XA CN110516311B (en) | 2019-07-31 | 2019-07-31 | Comprehensive compensation strategy construction method for constant error of vehicle acceleration sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110516311A CN110516311A (en) | 2019-11-29 |
| CN110516311B true CN110516311B (en) | 2023-03-24 |
Family
ID=68625056
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910700562.XA Active CN110516311B (en) | 2019-07-31 | 2019-07-31 | Comprehensive compensation strategy construction method for constant error of vehicle acceleration sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110516311B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112950812B (en) * | 2021-02-04 | 2022-07-26 | 南京航空航天大学 | Vehicle state fault-tolerant estimation method based on long-time and short-time memory neural network |
| CN114684159A (en) * | 2022-03-21 | 2022-07-01 | 潍柴动力股份有限公司 | Vehicle mass estimation method and device, electronic equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108594652B (en) * | 2018-03-19 | 2021-05-25 | 江苏大学 | Observer information iteration-based vehicle state fusion estimation method |
| CN109606378B (en) * | 2018-11-19 | 2020-06-09 | 江苏大学 | Vehicle running state estimation method for non-Gaussian noise environment |
-
2019
- 2019-07-31 CN CN201910700562.XA patent/CN110516311B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110516311A (en) | 2019-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8886400B2 (en) | Method for determining the drawbar length of a trailer of a tractor vehicle | |
| CN111007722B (en) | Transverse robust fault-tolerant control system and method for four-wheel steering automatic driving automobile | |
| CN112758097B (en) | State prediction and estimation method for unmanned vehicle | |
| CN111645699B (en) | Model self-adaptive lateral speed estimation method based on multi-sensor information fusion | |
| CN110516311B (en) | Comprehensive compensation strategy construction method for constant error of vehicle acceleration sensor | |
| CN108482379A (en) | Wheel-hub motor driven vehicle coefficient of road adhesion and road gradient synchronize real-time estimation system and method | |
| CN109606378A (en) | Vehicle running state estimation method towards non-Gaussian noise environment | |
| CN110979026B (en) | Distributed driving bus torque distribution method based on real-time road conditions | |
| CN111688715B (en) | Centroid slip angle observation method of four-wheel drive electric vehicle based on fusion technology | |
| KR20200088396A (en) | Vehicle, vehicle motion state estimation device and vehicle motion state estimation method | |
| CN113183950A (en) | Self-adaptive control method for steering of active front wheel of electric automobile | |
| CN116552550A (en) | Vehicle track tracking control system based on parameter uncertainty and yaw stability | |
| CN113247004A (en) | Joint estimation method for vehicle mass and road transverse gradient | |
| CN113811472B (en) | Ground contact load estimating device, control device, and ground contact load estimating method | |
| CN114620052A (en) | Vehicle signal processing method and device and driving equipment | |
| CN114435371B (en) | Road gradient estimation method and device | |
| CN113650621B (en) | Distributed driving electric vehicle state parameter estimation method facing complex working conditions | |
| CN113978476B (en) | Wire-controlled automobile tire lateral force estimation method considering sensor data loss | |
| JP6426698B2 (en) | Method for reducing the overturning risk of a motor vehicle having a controllable suspension system | |
| CN114044003B (en) | Tracking control method for front-rear double-axle steering vehicle | |
| CN113721609A (en) | 4WID high-clearance sprayer trajectory tracking control method under sideslip condition | |
| JP3271955B2 (en) | Road surface friction coefficient estimation device for vehicles | |
| CN113853329A (en) | Touchdown load estimation device, control device, and touchdown load estimation method | |
| CN115571156B (en) | Front vehicle transverse and longitudinal motion state joint estimation method based on sensor fusion | |
| CN117807703A (en) | Method for estimating key parameters of scooter chassis vehicle with mutually corrected dynamic and static parameters |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230302 Address after: 213032 No. 666 Liaohe Road, Xinbei District, Changzhou City, Jiangsu Province Applicant after: CHANGZHOU INSTITUTE OF TECHNOLOGY Address before: Zhenjiang City, Jiangsu Province, 212013 Jingkou District Road No. 301 Applicant before: JIANGSU University |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |