CN113978473A - Vehicle mass and road gradient estimation method - Google Patents

Vehicle mass and road gradient estimation method Download PDF

Info

Publication number
CN113978473A
CN113978473A CN202111417841.9A CN202111417841A CN113978473A CN 113978473 A CN113978473 A CN 113978473A CN 202111417841 A CN202111417841 A CN 202111417841A CN 113978473 A CN113978473 A CN 113978473A
Authority
CN
China
Prior art keywords
gradient
vehicle
mass
road gradient
road
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.)
Granted
Application number
CN202111417841.9A
Other languages
Chinese (zh)
Other versions
CN113978473B (en
Inventor
熊会元
杨子超
张辉
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.)
Sun Yat Sen University
Institute of Dongguan of Sun Yat Sen University
Original Assignee
Sun Yat Sen University
Institute of Dongguan of Sun Yat Sen University
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 Sun Yat Sen University, Institute of Dongguan of Sun Yat Sen University filed Critical Sun Yat Sen University
Priority to CN202111417841.9A priority Critical patent/CN113978473B/en
Publication of CN113978473A publication Critical patent/CN113978473A/en
Application granted granted Critical
Publication of CN113978473B publication Critical patent/CN113978473B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • 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
    • B60W40/076—Slope angle of the road
    • 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
    • B60W40/12—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 parameters of the vehicle itself, e.g. tyre models
    • B60W40/13—Load or weight
    • 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
    • B60W2520/00—Input parameters relating to overall vehicle dynamics

Landscapes

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

Abstract

The invention relates to a vehicle mass and road gradient estimation method, which comprises the following steps: the method comprises the following steps: acquiring vehicle parameters; step two: establishing a longitudinal dynamics model related to the mass of the whole vehicle and the road gradient and a longitudinal kinematics model related to the road gradient; step three: building a gradient and quality estimator based on a longitudinal dynamics model, and estimating the gradient of the road and the vehicle; building a gradient estimator based on a longitudinal kinematics model to estimate the road gradient; step four: the gradient is fused with the results of the mass estimator and the gradient estimator, and a vehicle mass estimate and a road gradient estimate are output. The method integrates the longitudinal dynamics model of the whole vehicle mass and the road gradient and the longitudinal kinematics model related to the road gradient, can simultaneously integrate the two conditions for estimation when predicting the whole vehicle mass estimation value and the road gradient estimation value, has stronger adaptability, high estimation precision and good stability when facing the variation of the vehicle running condition.

Description

Vehicle mass and road gradient estimation method
Technical Field
The invention relates to the field of parameter estimation, in particular to a method for estimating vehicle mass and road gradient.
Background
Various active safety systems are gradually increasing during vehicle modernization, but the performance of these systems depends on precise vehicle and environmental parameters. The mass of the heavy commercial vehicle can be changed frequently in the whole running process, and the mass change can reach more than 400% due to the large difference between the unloaded mass and the fully loaded mass. Meanwhile, the quality has great influence on the braking performance, the operation stability and the like of the whole vehicle. If the accurate whole vehicle mass can be acquired in real time, the control effect of the vehicle can be obviously improved. During the running process of the vehicle, the road surface gradient also influences the acceleration performance of the vehicle, and if the gradient of the road can be estimated, the braking and acceleration effects of the vehicle can be improved.
The existing method for estimating the automobile mass and the road gradient by considering the parameter coupling relation comprises the steps of firstly, acquiring vehicle state data and vehicle intrinsic parameters, and calculating to obtain the acceleration a and the transmission ratio i at the kth moment; step two, establishing an automobile mass model and a road gradient model; and step three, constructing a least square quality estimation model and a Kalman filtering slope estimation model based on the automobile quality model and the road slope model respectively, and step four, performing combined estimation on the automobile quality and the road slope by adopting nested loop iteration. The invention can provide real-time road gradient information and automobile load conditions for an automobile intelligent system, and provides important basis for automatic driving aid decision, green driving and automatic transmission gear-shifting control, thereby realizing safe, economic and comfortable driving.
However, in the existing scheme, the road gradient model is constructed only according to a vehicle longitudinal dynamic model, the requirement is high, the used acceleration is obtained by speed differentiation, and a large error is easy to generate. The actual working conditions which are difficult to be suitable for the vehicle are complicated and changeable, and the estimation accuracy is insufficient.
Disclosure of Invention
The invention aims to overcome the problem of insufficient accuracy of vehicle mass and road slope estimation in the prior art, and provides a vehicle mass and road slope estimation method.
In order to solve the technical problems, the invention adopts the technical scheme that: a vehicle mass and road grade estimation method comprising the steps of:
the method comprises the following steps: acquiring vehicle parameters;
step two: establishing a longitudinal dynamics model related to the mass of the whole vehicle and the road gradient and a longitudinal kinematics model related to the road gradient according to the parameters in the step one;
step three: building a gradient and quality estimator based on a longitudinal dynamics model, and estimating the gradient of the road and the vehicle; building a gradient estimator based on a longitudinal kinematics model to estimate the road gradient;
step four: the gradient is fused with the results of the mass estimator and the gradient estimator, and a vehicle mass estimate and a road gradient estimate are output.
Preferably, in the second step, the longitudinal dynamics model of mass and road gradient is specifically:
Figure BDA0003375794000000021
assuming cos α ≈ 1 and sin α ≈ tan α ═ i, simplification yields:
Figure BDA0003375794000000022
in the formula: t iseIs the drive torque, igIs the transmission ratio of the variator, i0Is the main reducer transmission ratio, etaTIs the mechanical efficiency of the drive chain, r is the rolling radius of the wheel, CDIs the air resistance coefficient, A is the windward area, ρ is the air density, v is the vehicle speed, m is the vehicle mass, g is the gravitational acceleration, i is the slope, α is the toe, f is the rolling resistance coefficient,
Figure BDA0003375794000000023
is the vehicle acceleration;
further modifications yield:
Figure BDA0003375794000000024
the selected state quantity is as follows:
x1＝[v i]T
further can be expressed as:
Figure BDA0003375794000000025
in the formula: Δ t is the sampling time, W1Is process noise.
Selecting the observed quantity as follows:
y1＝[v]
the observation equation was established as follows:
Figure BDA0003375794000000031
in the formula: Δ t is the sampling time, V1Is the measurement noise.
The resulting state space expression for longitudinal dynamics is shown below:
Figure BDA0003375794000000032
in the formula: h is1Is an observation matrix; f. of1Is a state transfer function.
Preferably, in the second step, the longitudinal kinematic model of the road gradient is specifically:
Figure BDA0003375794000000033
in the formula: a isxAcceleration values measured by a longitudinal acceleration sensor;
the selected state quantity is as follows:
x2＝[v i]T
further can be expressed as:
Figure BDA0003375794000000034
in the formula: Δ t is the sampling time, W2Is process noise;
selecting the observed quantity as follows:
y2＝[v]
the observation equation was established as follows:
Figure BDA0003375794000000041
in the formula: Δ t is the sampling time, V2Is process noise;
the resulting state space expression for longitudinal kinematics is as follows:
Figure BDA0003375794000000042
in the formula: h is2Is an observation matrix.
Preferably, in the third step, a dual extended kalman filter based on longitudinal dynamics is established as the gradient and quality estimator, and the specific process is as follows:
s3.1: predicting parameters;
s3.2: predicting the state;
s3.3: correcting the state;
s3.4: and (6) parameter estimation.
Preferably, a dual extended kalman filter or a dual unscented kalman filter of the gradient and quality estimator based on longitudinal dynamics is set up, and the specific flow is as follows:
s3.1: parameter prediction, specifically:
Figure BDA0003375794000000043
Figure BDA0003375794000000044
s3.2: the state prediction specifically includes:
Figure BDA0003375794000000045
Figure BDA0003375794000000046
s3.3: and (3) state correction, specifically:
Figure BDA0003375794000000047
Figure BDA0003375794000000048
Figure BDA0003375794000000049
s3.4: parameter estimation, specifically:
Figure BDA00033757940000000410
Figure BDA00033757940000000411
Figure BDA00033757940000000412
in the formula, xp1＝[m]；xs1＝[v i]；Ps1、Pp1Respectively a covariance matrix; qp1、Qs1Respectively, process noise covariance matrices; rs1、Rp1Respectively measuring a noise covariance matrix; jacobian matrix
Figure BDA0003375794000000051
Observation matrix Hs1＝h1＝[1 0](ii) a Observation matrix
Figure BDA0003375794000000052
Figure BDA0003375794000000053
Is a priori an estimate of the state variable,
Figure BDA0003375794000000054
is a priori estimate of the covariance matrix, Ks1、Kp1Respectively kalman gain.
Preferably, in step three, the slope estimator includes two procedures of prediction and correction based on the extended kalman filter of the longitudinal kinematics.
Preferably, the gradient estimator is adapted to, in the gradient estimator,
the prediction is specifically as follows:
Figure BDA0003375794000000055
Figure BDA0003375794000000056
the correction is specifically as follows:
Figure BDA0003375794000000057
Figure BDA0003375794000000058
Figure BDA0003375794000000059
in the formula, xs2＝[v i]T；Ps2Is a covariance matrix; qs2Is a process noise covariance matrix; rs2Is a measurement noise covariance matrix;
Figure BDA00033757940000000510
Hs2＝h2＝[1 0]。
preferably, the step four comprises the following specific processes:
s4.1: state interaction;
state estimation value x obtained according to k-1 times1(k-1) and xs2(k-1), covariance estimation value Ps1(k-1) and Ps2(k-1) and model mixture probability μ1(k) And mu2(k) To obtain a mixed state estimation value and a covariance estimation value;
predicted probability c of model jj：
Figure BDA00033757940000000511
Mixed probability mu of model i to model jij：
μ11(k-1)＝p11μ1(k-1)/c1
μ12(k-1)＝p12μ1(k-1)/c2
μ21(k-1)＝p21μ2(k-1)/c1
μ22(k-1)＝p22μ2(k-1)/c2
Hybrid estimation of states xs1i、xs2i：
Figure BDA0003375794000000061
Mixed estimation of covariance Ps1i、Ps2i：
Figure BDA0003375794000000062
S4.2: filtering; inputting the mixed state estimation value and covariance estimation value obtained in S4.1 into an estimator to enable the estimator to obtain a state estimation value and a covariance estimation value
Figure BDA0003375794000000063
Figure BDA0003375794000000064
After filtering, obtaining:
Figure BDA0003375794000000065
s4.3: updating the model probability, specifically:
respectively calculating innovation e of k timei：
Figure BDA0003375794000000066
Figure BDA0003375794000000067
Innovation covariance matrix calculation Si：
Figure BDA0003375794000000068
Figure BDA0003375794000000071
Updating the model probability by adopting a likelihood function, and normalizing:
Figure BDA0003375794000000072
Figure BDA0003375794000000073
Figure BDA0003375794000000074
Figure BDA0003375794000000075
in the formula, Λ1、Λ2Respectively, the likelihood of the model.
Preferably, the method further comprises a fifth step, wherein the fifth step specifically comprises the following steps:
parameters to be used for calculating the estimated mass of the entire vehicle and the estimated road gradient and other sensor parameters At1As characteristic input neural network, the vehicle mass estimated value and road gradient estimated value A in the fourth stept2Also used as characteristic input neural network, and the quality truth value and road gradient truth value Y of the experimental vehicletInput to the neural network as a tag. Training the neural network to learn the relationship between the features and the labels. For a trained network, At1And At2Inputting the network to obtain the vehicle mass and road gradient Y predicted by the networkt. Using vehicle mass estimate and road slope estimate At2As training data for neural networks, such that the predicted vehicle mass and road grade Y of the networktAnd is more accurate.
Preferably, in the step one, the structural parameters of the whole vehicle are acquired, and the parameters of the vehicle in the running process are acquired by using a sensor.
Compared with the prior art, the invention has the beneficial effects that: the method integrates the longitudinal dynamics model of the whole vehicle mass and the road gradient and the longitudinal kinematics model related to the road gradient, can simultaneously integrate the two conditions for estimation when predicting the whole vehicle mass estimation value and the road gradient estimation value, and has stronger adaptability, high estimation precision and good stability when facing the complexity and the variability of the vehicle running condition.
Drawings
FIG. 1 is a flow chart of a vehicle mass and road grade estimation method of the present invention;
FIG. 2 is a flowchart of the calculation of step four of the present invention;
FIG. 3 is a schematic diagram of the dual extended Kalman filter of the present invention.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent; for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted. The positional relationships depicted in the drawings are for illustrative purposes only and are not to be construed as limiting the present patent.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "long", "short", etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the drawings, it is only for convenience of description and simplicity of description, but does not indicate or imply that the device or element referred to must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationships in the drawings are only used for illustrative purposes and are not to be construed as limitations of the present patent, and specific meanings of the terms may be understood by those skilled in the art according to specific situations.
In the drawings of the embodiments of the present invention, for convenience of reading and understanding, a front plate, a rear plate and a top plate in a cabinet structure are all shown.
The technical scheme of the invention is further described in detail by the following specific embodiments in combination with the attached drawings:
example 1
FIG. 1 shows an embodiment 1 of a vehicle mass and road grade estimation method, comprising the steps of:
the method comprises the following steps: acquiring vehicle parameters; acquiring structural parameters of the whole vehicle and acquiring parameters of the vehicle in the running process by using a sensor.
Step two: establishing a longitudinal dynamics model related to the mass of the whole vehicle and the road gradient and a longitudinal kinematics model related to the road gradient according to the parameters in the step one;
step three: building a gradient and quality estimator based on a longitudinal dynamics model, and estimating the gradient of the road and the vehicle; building a gradient estimator based on a longitudinal kinematics model to estimate the road gradient;
step four: the gradient is fused with the results of the mass estimator and the gradient estimator, and a vehicle mass estimate and a road gradient estimate are output.
The beneficial effects of this embodiment: the method integrates the longitudinal dynamics model of the whole vehicle mass and the road gradient and the longitudinal kinematics model related to the road gradient, can simultaneously integrate the two conditions for estimation when predicting the whole vehicle mass estimation value and the road gradient estimation value, and has stronger adaptability, high estimation precision and good stability when facing the complexity and the variability of the vehicle running condition.
Example 2
Embodiment 2 of a vehicle mass and road grade estimation method, comprising the steps of:
the method comprises the following steps: acquiring vehicle parameters;
step two: establishing a longitudinal dynamics model related to the mass of the whole vehicle and the road gradient and a longitudinal kinematics model related to the road gradient according to the parameters in the step one;
the longitudinal dynamics model of the mass and the road gradient is specifically as follows:
the longitudinal kinematic model of the road gradient is specifically as follows:
Figure BDA0003375794000000091
in the formula: t iseIs the drive torque, igIs the transmission ratio of the variator, i0Is the main reducer transmission ratio, etaTIs the mechanical efficiency of the drive chain, r is the rolling radius of the wheel, CDIs the air resistance coefficient, A is the windward area, ρ is the air density, v is the vehicle speed, m is the vehicle mass, g is the gravitational acceleration, i is the slope, α is the toe, f is the rolling resistance coefficient,
Figure BDA0003375794000000092
is the vehicle acceleration;
the longitudinal kinematic model of the road gradient is specifically as follows:
Figure BDA0003375794000000093
in the formula: a isxAcceleration values measured by a longitudinal acceleration sensor;
step three: a slope and quality estimator is built based on a longitudinal dynamics model, the slope and quality estimator estimates the road slope and the vehicle based on a double-extended Kalman filter of longitudinal dynamics, and the specific flow is as follows:
s3.1: parameter prediction:
Figure BDA0003375794000000094
Figure BDA0003375794000000095
s3.2: and (3) state prediction:
Figure BDA0003375794000000096
Figure BDA0003375794000000097
s3.3: and (3) state correction:
Figure BDA0003375794000000101
Figure BDA0003375794000000102
Figure BDA0003375794000000103
s3.4: parameter estimation:
Figure BDA0003375794000000104
Figure BDA0003375794000000105
Figure BDA0003375794000000106
in the formula, xp1＝[m]；xs1＝[v i]T；Ps1、Pp1Respectively a covariance matrix; qp1、Qs1Respectively, process noise covariance matrices; rs1、Rp1Respectively measuring a noise covariance matrix; jacobian matrix
Figure BDA0003375794000000107
Observation matrix Hs1＝h1＝[1 0](ii) a Observation matrix
Figure BDA0003375794000000108
Figure BDA0003375794000000109
Is a priori an estimate of the state variable,
Figure BDA00033757940000001010
is a priori estimate of the covariance matrix, Ks1、Kp1Respectively kalman gain.
A slope estimator is built based on a longitudinal kinematics model to estimate the road slope, and the slope estimator is based on an extended kalman filter of the longitudinal kinematics, as shown in fig. 3, specifically calculated as follows:
and (3) prediction:
Figure BDA00033757940000001011
Figure BDA00033757940000001012
and (3) correction:
Figure BDA00033757940000001013
Figure BDA00033757940000001014
Figure BDA00033757940000001015
in the formula, xs2＝[v i]T；Ps2Is a covariance matrix; qs2Is a process noise covariance matrix; rs2Is a measurement noise covariance matrix;
Figure BDA00033757940000001016
Hs2＝h2＝[1 0]。
step four: as shown in fig. 2, the gradient is fused with the results of the mass estimator and the gradient estimator, and a vehicle mass estimated value and a road gradient estimated value are output. The specific process is as follows:
s4.1: state interaction;
state estimation value x obtained according to k-1 times1(k-1) and xs2(k-1), covariance estimation value Ps1(k-1) and Ps2(k-1) and model mixture probability μ1(k) And mu2(k) To obtain a mixed state estimation value and a covariance estimation value;
predicted probability c of model jj：
Figure BDA0003375794000000111
Mixed probability mu of model i to model jij：
μ11(k-1)＝p11μ1(k-1)/c1
μ12(k-1)＝p12μ1(k-1)/c2
μ21(k-1)＝p21μ2(k-1)/c1
μ22(k-1)＝p22μ2(k-1)/c2
Hybrid estimation of states xs1i、xs2i：
Figure BDA0003375794000000112
Mixed estimation of covariance Ps1i、Ps2i：
Figure BDA0003375794000000113
S4.2: filtering; inputting the mixed state estimation value and covariance estimation value obtained in S4.1 into an estimator to enable the estimator to obtain a state estimation value and a covariance estimation value
Figure BDA0003375794000000114
Figure BDA0003375794000000115
After filtering, obtaining:
Figure BDA0003375794000000116
s4.3: updating the model probability, specifically:
respectively calculating innovation e of k timei：
Figure BDA0003375794000000121
Figure BDA0003375794000000122
Innovation covariance matrix calculation Si：
Figure BDA0003375794000000123
Figure BDA0003375794000000124
Updating the model probability by adopting a likelihood function, and normalizing:
Figure BDA0003375794000000125
Figure BDA0003375794000000126
Figure BDA0003375794000000127
Figure BDA0003375794000000128
in the formula, Λ1、Λ2Respectively, the likelihood of the model.
The beneficial effects of this embodiment: the method integrates the longitudinal dynamics model of the whole vehicle mass and the road gradient and the longitudinal kinematics model related to the road gradient, can simultaneously integrate the two conditions for estimation when predicting the whole vehicle mass estimation value and the road gradient estimation value, and has stronger adaptability, high estimation precision and good stability when facing the complexity and the variability of the vehicle running condition.
Example 3
An embodiment 3 of a method for estimating a vehicle mass and a road gradient is based on the above embodiment 1 or 2, and is different from the above embodiment in that the method further includes a fifth step, specifically:
parameters to be used for calculating the estimated mass of the entire vehicle and the estimated road gradient and other sensor parameters At1As characteristic input neural network, the vehicle mass estimated value and road gradient estimated value A in the fourth stept2Also used as characteristic input neural network, and the quality truth value and road gradient truth value Y of the experimental vehicletInput to the neural network as a tag. Training the neural network to learn the relationship between the features and the labels. For a trained network, At1And At2Inputting the network to obtain the vehicle mass and road gradient Y predicted by the networkt。
The beneficial effects of this embodiment: the method integrates the longitudinal dynamics model of the whole vehicle mass and the road gradient and the longitudinal kinematics model related to the road gradient, can simultaneously integrate the two conditions for estimation when predicting the whole vehicle mass estimation value and the road gradient estimation value, and has stronger adaptability, high estimation precision and good stability when facing the complexity and the variability of the vehicle running condition.
It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.

Claims (10)

1. A vehicle mass and road grade estimation method, comprising the steps of:
the method comprises the following steps: acquiring vehicle parameters;
step two: establishing a longitudinal dynamics model related to the mass of the whole vehicle and the road gradient and a longitudinal kinematics model related to the road gradient according to the parameters in the step one;
step three: building a gradient and quality estimator based on a longitudinal dynamics model, and estimating the gradient of the road and the vehicle; building a gradient estimator based on a longitudinal kinematics model to estimate the road gradient;
step four: the gradient is fused with the results of the mass estimator and the gradient estimator, and a vehicle mass estimate and a road gradient estimate are output.
2. A vehicle mass and road gradient estimation method as claimed in claim 1, wherein in said step two, the mass and road gradient longitudinal dynamics model is specifically:
Figure FDA0003375793990000011
in the formula, TeIs the drive torque, igIs the transmission ratio of the variator, i0Is the main reducer transmission ratio, etaTIs the mechanical efficiency of the drive chain, r is the rolling radius of the wheel, CDIs the air resistance coefficient, A is the windward area, ρ is the air density, v is the vehicle speed, m is the vehicle mass, g is the gravitational acceleration, i is the slope, α is the slope angle, f is the rolling resistance coefficient,
Figure FDA0003375793990000012
is the vehicle acceleration;
3. a vehicle mass and road gradient estimation method according to claim 1, wherein in step two, the longitudinal kinematic model of the road gradient is specifically:
Figure FDA0003375793990000013
in the formula, axAcceleration values measured by the longitudinal acceleration sensor.
4. The vehicle mass and road gradient estimation method according to claim 1, characterized in that in the third step, a dual extended kalman filter based on longitudinal dynamics is constructed for the gradient and mass estimator, and the specific process is as follows:
s3.1: predicting parameters;
s3.2: predicting the state;
s3.3: correcting the state;
s3.4: and (6) parameter estimation.
5. The vehicle mass and road gradient estimation method according to claim 4, characterized in that a dual extended Kalman filter or a dual unscented Kalman filter of a gradient and mass estimator based on longitudinal dynamics is constructed, and the specific flow is as follows:
s3.1: parameter prediction, specifically:
Figure FDA0003375793990000021
Figure FDA0003375793990000022
s3.2: the state prediction specifically includes:
Figure FDA0003375793990000023
Figure FDA0003375793990000024
s3.3: and (3) state correction, specifically:
Figure FDA0003375793990000025
Figure FDA0003375793990000026
Figure FDA0003375793990000027
s3.4: parameter estimation, specifically:
Figure FDA0003375793990000028
Figure FDA0003375793990000029
Figure FDA00033757939900000210
in the formula, xp1＝[m]；xs1＝[v i]T；Ps1、Pp1Respectively a covariance matrix; qp1、Qs1Respectively, process noise covariance matrices; rs1、Rp1Respectively measuring a noise covariance matrix; jacobian matrix
Figure FDA00033757939900000211
Observation matrix Hs1＝h1＝[1 0](ii) a Observation matrix
Figure FDA00033757939900000212
Figure FDA00033757939900000213
Is a priori an estimate of the state variable,
Figure FDA00033757939900000214
is a priori estimate of the covariance matrix, Ks1、Kp1Respectively kalman gain.
6. The vehicle mass and road gradient estimation method according to claim 5, wherein in the third step, the gradient estimator is based on an extended Kalman filter of longitudinal kinematics, and the method comprises two procedures of prediction and correction.
7. A vehicle mass and road grade estimation method according to claim 6, wherein, in the grade estimator,
the prediction is specifically as follows:
Figure FDA00033757939900000215
Figure FDA00033757939900000216
the correction is specifically as follows:
Figure FDA0003375793990000031
Figure FDA0003375793990000032
Figure FDA0003375793990000033
in the formula, xs2＝[v i]T；Ps2Is a covariance matrix; qs2Is a process noise covariance matrix; rs2Is a measurement noise covariance matrix;
Figure FDA0003375793990000034
Hs2＝h2＝[1 0]。
8. the vehicle mass and road grade estimation method according to claim 7, wherein the step four is specifically performed by:
s4.1: state interaction;
state estimation value x obtained according to k-1 times1(k-1) and xs2(k-1), covariance estimation value Ps1(k-1) and Ps2(k-1) and model mixture probability μ1(k) And η2(k) To obtain a mixed state estimation value and a covariance estimation value;
predicted probability c of model jj：
Figure FDA0003375793990000035
Mixed probability mu of model i to model jij：
μ11(k-1)＝p11μ1(k-1)/c1
μ12(k-1)＝p12μ1(k-1)/c2
μ21(k-1)＝p21μ2(k-1)/c1
μ22(k-1)＝p22μ2(k-1)/c2
Hybrid estimation of states xs1i、xs2i：
Figure FDA0003375793990000036
Mixed estimation of covariance Ps1i、Ps2i：
Figure FDA0003375793990000037
S4.2: filtering; inputting the mixed state estimation value and covariance estimation value obtained in S4.1 into an estimator to enable the estimator to obtain a state estimation value and a covariance estimation value
Figure FDA0003375793990000041
Figure FDA0003375793990000042
After filtering, obtaining:
Figure FDA0003375793990000043
s4.3: updating the model probability, specifically:
respectively calculating the innovation of k timeei：
Figure FDA0003375793990000044
Figure FDA0003375793990000045
Innovation covariance matrix calculation Si：
Figure FDA0003375793990000046
Figure FDA0003375793990000047
Updating the model probability by adopting a likelihood function, and normalizing:
Figure FDA0003375793990000048
Figure FDA0003375793990000049
Figure FDA00033757939900000410
Figure FDA00033757939900000411
in the formula, Λ1、Λ2Respectively, the likelihood of the model.
9. A vehicle mass and road grade estimation method according to any one of claims 1-8, further comprising step five, specifically:
parameters to be used for calculating the estimated mass of the entire vehicle and the estimated road gradient and other sensor parameters At1As characteristic input neural network, the vehicle mass estimated value and road gradient estimated value A in the fourth stept2Also used as characteristic input neural network, and the quality truth value and road gradient truth value Y of the experimental vehicletInput to the neural network as a tag. Training the neural network to learn the relationship between the features and the labels. For a trained network, At1And At2Inputting the network to obtain the vehicle mass and road gradient Y predicted by the networkt。
10. A vehicle mass and road gradient estimation method according to any one of claims 1-8, characterized in that in step one, structural parameters of the whole vehicle are obtained and parameters of the vehicle during driving are obtained by using sensors.
CN202111417841.9A 2021-11-25 2021-11-25 Vehicle mass and road gradient estimation method Active CN113978473B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111417841.9A CN113978473B (en) 2021-11-25 2021-11-25 Vehicle mass and road gradient estimation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111417841.9A CN113978473B (en) 2021-11-25 2021-11-25 Vehicle mass and road gradient estimation method

Publications (2)

Publication Number Publication Date
CN113978473A true CN113978473A (en) 2022-01-28
CN113978473B CN113978473B (en) 2023-05-23

Family

ID=79732011

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111417841.9A Active CN113978473B (en) 2021-11-25 2021-11-25 Vehicle mass and road gradient estimation method

Country Status (1)

Country Link
CN (1) CN113978473B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114655224A (en) * 2022-03-21 2022-06-24 潍柴动力股份有限公司 Road gradient estimation method, electronic device and storage medium
CN114919585A (en) * 2022-07-22 2022-08-19 杭州宏景智驾科技有限公司 Vehicle weight and road gradient estimation method based on vehicle dynamics model

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112613253A (en) * 2021-01-06 2021-04-06 东南大学 Vehicle mass and road gradient combined self-adaptive estimation method considering environmental factors

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112613253A (en) * 2021-01-06 2021-04-06 东南大学 Vehicle mass and road gradient combined self-adaptive estimation method considering environmental factors

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
付翔;刘会康;黄斌;: "基于CRLS的路面坡度及轮毂电机车辆质量估计", 数字制造科学 *
任志英;沈亮量;黄伟;刘星星;: "基于AEKF的车辆质量与道路坡度实时估计", 振动.测试与诊断 *
李尊;孔盼;史艳龙;: "重型商用车质量和道路坡度估算方法", 重型汽车 *
赵健;李至轩;朱冰;李雅欣;孙玉泽;: "基于交互多模型的车辆质量与道路坡度估计", 中国公路学报 *
郝胜强;罗培培;席军强;: "基于稳态卡尔曼滤波的车辆质量与道路坡度估计", 汽车工程 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114655224A (en) * 2022-03-21 2022-06-24 潍柴动力股份有限公司 Road gradient estimation method, electronic device and storage medium
CN114919585A (en) * 2022-07-22 2022-08-19 杭州宏景智驾科技有限公司 Vehicle weight and road gradient estimation method based on vehicle dynamics model
CN114919585B (en) * 2022-07-22 2022-11-04 杭州宏景智驾科技有限公司 Vehicle weight and road gradient estimation method based on vehicle dynamics model

Also Published As

Publication number Publication date
CN113978473B (en) 2023-05-23

Similar Documents

Publication Publication Date Title
CN112613253B (en) Vehicle mass and road gradient combined self-adaptive estimation method considering environmental factors
CN113978473A (en) Vehicle mass and road gradient estimation method
CN103661393B (en) Kinematic road gradient is estimated
CN110588657B (en) Joint estimation method for vehicle motion state and road gradient
CN109583571B (en) Mobile robot soft ground trafficability prediction method based on LSTM network
CN113806861B (en) Method for estimating vehicle mass and road gradient of two-step structure
CN103661352A (en) Static road gradient estimation
CN107458380A (en) A kind of road grade real-time estimation method being applied under comprehensive driving cycles
CN113830088B (en) Intelligent semi-trailer tractor trajectory tracking prediction control method and vehicle
CN111198032A (en) Real-time estimation method for automobile mass
CN112180361A (en) Vehicle-mounted radar target tracking method based on dynamic federal filtering
JP5337090B2 (en) Vehicle characteristic information estimation device and warning device using the same
CN115271040A (en) Vehicle quality estimation method based on adaptive extended Kalman filtering
CN114802223A (en) Intelligent vehicle controllable capacity grade prediction method
CN117719519A (en) Vehicle running state estimation method
CN116674571A (en) Real-time estimation method for automobile quality and gradient based on data confidence factor
CN116538286A (en) Commercial vehicle gear shifting system and method considering NVH characteristics
CN113158569B (en) Tank car side-tipping state high-reliability estimation method based on long-short term memory network
CN112417365A (en) Automatic driving truck quality estimation method based on extended Kalman filtering
CN111703429A (en) Method for estimating longitudinal speed of wheel hub motor driven vehicle
CN111695196A (en) Tank car side-tipping state pre-judging method based on vehicle body kinematics information
CN117077281A (en) UKF-CNN-LSTM-based heavy-load transport vehicle centroid slip angle estimation method
CN113721509B (en) Parameter estimation method based on container truck signal acquisition device
CN114357624B (en) Vehicle weight estimation algorithm based on second-order linear differential tracker and parameter bilinear model
CN116588119B (en) Vehicle state estimation method based on tire model parameter self-adaption

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