A longitudinal slip estimation method for articulated vehicles

Abstract

The present invention discloses a longitudinal slip estimation method for an articulated vehicle, the purpose of which is to provide a more effective vehicle state estimation implementation scheme for a vehicle with an articulated steering structure, which includes the following steps: at the beginning of the estimation, the estimator data is initialized; then, according to the dynamic characteristics of the articulated wheeled vehicle, a two-degree-of-freedom vehicle model is built; the vehicle speed state information is estimated using an unscented Kalman filter algorithm; further, according to the kinematic characteristics of the articulated wheeled vehicle, the longitudinal slip of the articulated wheeled vehicle is estimated with the aid of the estimated speed state information and wheel speed information. The present invention overcomes the estimation algorithm distortion problem that may be caused by the strong nonlinearity of the articulated wheeled vehicle model by designing a two-degree-of-freedom dynamic model for articulated wheeled engineering vehicles, and designs an Unscented Kalman estimation method based on this model, thereby realizing accurate estimation of the four-wheel longitudinal slip state of the articulated wheeled vehicle.

Description

Technical Field

[0001] The present invention relates to the field of state estimation of autonomous driving vehicles, and in particular to a method for estimating the longitudinal slip rate of an articulated wheeled vehicle. Background Art

[0002] Articulated wheeled engineering vehicles are mainly divided into two parts, the front and rear body, which are connected by a rigid articulated body in the middle. The front and rear body are driven by hydraulic cylinders to fold and turn. They are widely used in non-structural scenes such as construction sites and mines. Accurate acquisition of vehicle status is essential for the intelligence, automation and vehicle safety of articulated wheeled vehicles. Many applications of articulated vehicles, such as path tracking control and safety status detection, all rely on state parameters such as yaw angular velocity, articulation angular velocity, front and rear body speed, and slip rate to varying degrees.

[0003] Among them, the longitudinal slip rate plays a vital role in the dynamic identification and safety control of articulated vehicles: on the one hand, the longitudinal slip rate is highly correlated with tire force modeling and is an indispensable parameter based on dynamic modeling; on the other hand, the longitudinal slip rate corresponds to the road adhesion coefficient. By controlling the slip rate, the tire can operate in the stable adhesion characteristic zone to ensure vehicle safety.

[0004] However, the longitudinal slip rate of a vehicle itself cannot be directly measured under the premise of practicality, so it needs to be estimated indirectly with the help of state estimation technology. At present, the research on vehicle state estimation methods based on model prediction is still mainly focused on ordinary vehicles, while the dynamic model of articulated vehicles is highly nonlinear, and there is also a strong coupling between the dynamic parameters involved. There is currently a lack of a dynamic model that can accurately describe the dynamic characteristics of articulated wheeled vehicles. Therefore, there are certain challenges in estimating the longitudinal slip rate of the object by building a vehicle dynamic model.

[0005] It can be seen that the current acquisition of the state of articulated wheeled vehicles is still insufficient and cannot accurately estimate the longitudinal slip rate of the vehicle. Summary of the invention

[0006] The main purpose of the present invention is to overcome the above-mentioned defects in the problem of estimating the longitudinal state of an articulated wheeled vehicle and to propose a method for estimating the longitudinal slip rate of an articulated wheeled vehicle.

[0007] The present invention adopts the following technical solution:

[0008] The present invention discloses a longitudinal slip rate estimation method for an articulated vehicle, which is characterized in that the dynamic characteristics of the articulated wheeled vehicle are fully considered from the aspects of vertical load, tire force, dynamic model, etc., and the Unscented Kalman method is used to overcome the estimation algorithm distortion problem that may be caused by the strong nonlinearity of the articulated wheeled vehicle model. The estimation method comprises the following steps:

[0009] 1) Estimator data initialization

[0010] 2) Build a two-degree-of-freedom vehicle model based on the dynamic characteristics of the articulated wheeled vehicle;

[0011] 3) Use the unscented Kalman filter algorithm to estimate the vehicle speed state information.

[0012] 4) According to the kinematic characteristics of the articulated wheeled vehicle, the longitudinal slip rate of the articulated wheeled vehicle is estimated with the aid of the estimated speed state information and wheel speed information.

[0013] The vehicle state initialization formula is expressed as:

[0014] [x,y]'=[N,N]

[0015] Where x refers to the longitudinal velocity of the vehicle's total center of mass, y refers to the lateral velocity of the vehicle's total center of mass, and N refers to a very small constant that is not equal to 0.

[0016] The state coordination mode initialization formula is expressed as:

[0017]

[0018] Where P0 refers to the initial state covariance, E[] represents the expected value calculation, and x0 refers to the true value of the state vector matrix at the initial moment. Refers to the estimated value of the state vector matrix x0 at the initial moment, x0 is expressed as follows:

[0019] x0=[x,y]'

[0020] The two-degree-of-freedom dynamic model is as follows:

[0021]

[0022] where v x represents the longitudinal velocity of the vehicle's center of mass, v y represents the lateral velocity of the vehicle's center of mass, represents the longitudinal acceleration of the vehicle's center of mass, Indicates the lateral acceleration of the vehicle's center of mass. xii With F yii Refers to the longitudinal and lateral tire forces, F xfl 、F xfr 、F xrl and F xrr They represent the longitudinal tire forces on the front left wheel, front right wheel, rear left wheel and rear right wheel respectively, and F yfl 、F yfr 、F yrl and F yrrRepresents the lateral tire force on the front left wheel, front right wheel, rear left wheel and rear right wheel respectively. f and T r They represent the wheelbase of the front and rear wheels of the vehicle, γ represents the yaw rate, and δ represents the articulation angle. fb , l rb They refer to the distance from the center of mass of the front vehicle body and the center of mass of the rear vehicle body to the articulated axis, m is the mass of the vehicle, m fb 、m rb are the front vehicle body mass and the rear vehicle body mass respectively.

[0023] F xii With F yii It is expressed as:

[0024]

[0025] Where σ represents the longitudinal slip rate, α represents the sideslip angle, and λ t is a calculation process parameter, C σ , C α They represent the longitudinal and lateral stiffness of the tire, μ represents the ground friction coefficient, η is a calculation process parameter, and F z Indicates the vertical load on the tire.

[0026] The vertical load F in the two-degree-of-freedom dynamic model z The specific expressions are as follows:

[0027]

[0028] Where L fr , L fl , L r Respectively refer to the longitudinal distance from the front right wheel, front left wheel, and rear wheel to the center of mass of the vehicle, W fr , W fl It refers to the lateral distance from the two front wheels of the vehicle to the hinge point, g represents the acceleration due to gravity, and F f Indicates air resistance, F zd represents the vertical load at the hinge point, T r Rear wheel track, m f and m r Respectively represent the front and rear vehicle masses, a yf and a yr Represent the lateral acceleration of the front and rear vehicles respectively, and d refers to the distance from the center of mass of the vehicle to the ground.

[0029] According to the above complete dynamic model, the state equation and measurement equation required by the algorithm are determined as follows:

[0030] The state equation is expressed as:

[0031]

[0032] The measurement equation is expressed as:

[0033] y(t)=h(x(t))=[a x ,a y ]+v(t)

[0034] According to the state equation and measurement equation, the Unscented Kalman filter algorithm is used to estimate the speed state of the vehicle, as follows:

[0035] The steps of projecting the state quantity to the sigma point set are as follows:

[0036]

[0037] Among them, k refers to the kth moment, χ x (k-1) refers to the state quantity sigma sampling point set at time (k-1), Refers to the estimated value of the state quantity at time (k-1), L x represents the dimension of the state quantity, and P(k-1) represents the state covariance at time (k-1).

[0038] The weights required for sigma point weighting are:

[0039]

[0040] Where W (m) , W (c) is the weight required for subsequent weighting, n represents the dimension of the state vector, and α represents a [10 -4 ,1], λ is calculated as follows

[0041] λ=α 2 (L x +3-L x )-L x

[0042] The time update formula is:

[0043]

[0044] Among them, k refers to the k moment, refers to the state quantity sigma sampling point set, u refers to the input quantity, Refers to the prior prediction of the sigma point set, Refers to the prior prediction of the state quantity, W (c) Refers to the weight used for weighting the sigma point set, the subscript i refers to the i-th column of the matrix, P xx refers to the a priori estimated covariance matrix.

[0045] The observation update formula is:

[0046]

[0047] Where Q is the process noise covariance matrix; R is the measurement noise covariance matrix, refers to the prior prediction of the observation value sigma point set, h[] refers to the observation equation, is the prior prediction of the observed value, P yy refers to the covariance matrix of the observations, P xy Refers to the covariance matrix of state variables and observation variables.

[0048] The Kalman gain calculation and update formula is:

[0049] K(k)=P xy (k|k-1)P yy (k|k-1) -1

[0050]

[0051] P(k|k)=P(k|k-1)-K(k)P yy (k|k-1)K(k) T

[0052] Where K is the Kalman gain, is the posterior state estimator at time k, and P(k|k) is the posterior state covariance matrix.

[0053] The vehicle speed information estimated by the UKF algorithm and the wheel speed information obtained by measurement are used to estimate the sideslip angle and slip rate of the four wheels of the articulated wheeled vehicle.

[0054] The formula for longitudinal slip rate is:

[0055]

[0056] The formula for the sideslip angle is:

[0057]

[0058] Where r e represents the wheel radius, ω ij Indicates the wheel speed, v xij Indicates the longitudinal velocity of each wheel center, v yij Indicates the longitudinal velocity of each wheel center, v xfl 、v xfr 、v xrl and v xrr They represent the longitudinal speeds of the front left wheel, front right wheel, rear left wheel and rear right wheel, respectively, v yfl 、v yfr 、v yrl and vyrr Represents the lateral speed of the front left wheel, front right wheel, rear left wheel and rear right wheel respectively. xij 、v yij It is determined by the following formula.

[0059] v xij The formula is as follows:

[0060]

[0061] v yij The formula is as follows:

[0062]

[0063] v yrl =v y -γ(l ra -l rb )

[0064] v yrr =v y -γ(l ra -l rb )

[0065] Where L fl Indicates the distance from the front left wheel center to the hinge point, T f Indicates the front wheel track, L fr Indicates the distance from the front right wheel center to the hinge point, l ra Indicates the distance from the rear axle of the wheel to the hinge point, l rb Indicates the distance from the center of mass of the rear vehicle body to the hinge point.

[0066] It can be seen from the above description of the present invention that, compared with the prior art, the present invention has the following beneficial effects:

[0067] (1) The present invention proposes a method for estimating the longitudinal slip rate of an articulated wheeled vehicle. According to the dynamic characteristics of the articulated wheeled vehicle, a simple and applicable two-degree-of-freedom dynamic model is proposed, and an effective method for obtaining the vehicle speed state is provided, which provides a basic model basis for estimating the longitudinal slip rate.

[0068] (2) In the method provided by the present invention, the Unscented Kalman algorithm is introduced to estimate the front and rear vehicle acceleration information, which effectively overcomes the result distortion caused by the high nonlinearity of the articulated vehicle dynamics model.

[0069] (3) In the method proposed in the present invention, for any turning angle condition of an articulated wheeled vehicle, the vertical load and tire force are calculated in sequence through the vehicle torque balance relationship and the tire-road action relationship, and a two-degree-of-freedom dynamic model is introduced to calculate the vehicle speed information. Finally, the estimation of the four-wheel longitudinal slip rate under any turning angle condition is achieved, which overcomes the shortcomings of the existing articulated wheeled vehicle in state acquisition and can estimate the vehicle state more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a flow chart of the method provided by an embodiment of the present invention.

[0071] Figure 2 The two-degree-of-freedom vehicle model provided by the embodiment of the present invention is shown in FIG. Figure 1 .

[0072] Figure 3 It is a schematic diagram of solving the Unscented Kalman filtering algorithm provided by an embodiment of the present invention.

[0073] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. DETAILED DESCRIPTION

[0074] The present invention is further described below through specific implementation modes.

[0075] like Figure 1 , the present invention proposes a flow chart of a longitudinal slip rate estimation method for an articulated vehicle, specifically:

[0076] S101: Initializing data of articulated vehicle status;

[0077] In order to prevent non-positive definite matrices from appearing in the calculation, it is necessary to initialize some estimator parameters at the beginning of the estimation so that the entire estimation process can run normally. Therefore, the parameters are initialized in advance, such as Figure 3 As shown, the details are as follows:

[0078] x0=[x,y]'=[N,N]

[0079]

[0080] Where P0 refers to the initial state covariance, E[] represents the expected value calculation, and x0 refers to the true value of the state vector matrix at the initial moment. Refers to the estimated value of the state vector matrix x0 at the initial moment. x refers to the longitudinal velocity of the total center of mass of the vehicle, y refers to the lateral velocity of the total center of mass of the vehicle, and N refers to a very small constant that is not equal to 0.

[0081] S102: Building a two-degree-of-freedom vehicle model according to the dynamic characteristics of the articulated wheeled vehicle;

[0082] On this basis, in order to accurately describe the vehicle dynamics performance of the articulated wheeled vehicle, a two-degree-of-freedom dynamic model of the articulated wheeled vehicle is established, such as Figure 2 As shown, the details are as follows:

[0083]

[0084] where v x represents the longitudinal velocity of the vehicle's center of mass, v y represents the lateral velocity of the vehicle's center of mass, represents the longitudinal acceleration of the vehicle's center of mass, Indicates the lateral acceleration of the vehicle's center of mass. xii With F yii Refers to the longitudinal and lateral tire forces, v xfl 、v xfr 、v xrl and v xrr They represent the longitudinal speeds of the front left wheel, front right wheel, rear left wheel and rear right wheel, respectively, v yfl 、v yfr 、v yrl and v yrr Represents the lateral speed of the front left wheel, front right wheel, rear left wheel and rear right wheel respectively. f and T r They represent the wheelbase of the front and rear wheels of the vehicle, γ represents the yaw rate, and δ represents the articulation angle. fb , l rb They refer to the distance from the center of mass of the front vehicle body and the center of mass of the rear vehicle body to the articulated axis, m is the mass of the vehicle, m fb 、m rb are the front vehicle body mass and the rear vehicle body mass respectively.

[0085] In the dynamic model, the interaction between the vehicle and the ground is reflected by the tire force. The tires of the four wheels are calculated according to the DUGOFF tire model, and the longitudinal tire force F xii and the lateral tire force F yii Specifically expressed as:

[0086]

[0087] Where σ represents the longitudinal slip rate, α represents the sideslip angle, and λ t is a calculation process parameter, C σ , C α They represent the longitudinal and lateral stiffness of the tire, μ represents the ground friction coefficient, η is a calculation process parameter, and F z Indicates the vertical load on the tire.

[0088] The vertical load F described in the model zii The specific expressions are as follows:

[0089]

[0090] Where L fr , L fl , L r Respectively refer to the longitudinal distance W from the front right wheel, front left wheel, and rear wheel to the center of mass of the vehicle fr , W fl It refers to the lateral distance from the two front wheels of the vehicle to the hinge point, g represents the acceleration due to gravity, F f Indicates air resistance, F zd represents the vertical load at the hinge point, T r Rear wheel track, m f and m r Respectively represent the front and rear vehicle masses, a yf and a yr Represent the lateral acceleration of the front and rear vehicles respectively, and d refers to the distance from the center of mass of the vehicle to the ground.

[0091] According to the above complete dynamic model, the state equation and measurement equation required by the algorithm are determined as follows:

[0092] The state equation is expressed as:

[0093]

[0094] The measurement equation is expressed as:

[0095] y(t)=h(x(t))=[a x ,a y ]+v(t)

[0096] Among them, a x represents the longitudinal acceleration of the vehicle's total center of mass, a y It represents the lateral acceleration of the vehicle's total center of mass. Articulated vehicles have variable structural characteristics, so it is necessary to measure the acceleration information of the front and rear bodies separately and perform corresponding conversions.

[0097]

[0098] where a xf and a xr They represent the longitudinal acceleration of the front and rear vehicle mass centers, respectively, and a yf and a yr They represent the longitudinal acceleration of the front vehicle center of mass and the rear vehicle center of mass respectively.

[0099] S103: using an unscented Kalman filter algorithm to estimate vehicle speed state information;

[0100] On this basis, the Unscented Kalman filter algorithm is used to estimate the speed state of the articulated wheeled vehicle, such as Figure 3 As shown, the details are as follows:

[0101] The steps of projecting the state quantity to the sigma point set are as follows:

[0102]

[0103] Among them, k refers to the kth moment, χ x (k-1) refers to the state quantity sigma sampling point set at time (k-1), Refers to the estimated value of the state quantity at time (k-1), L x The size is equal to the dimension of the state quantity, and P(k-1) represents the state covariance at time (k-1).

[0104] The weights required for sigma point weighting are:

[0105]

[0106] Where W (m) , W (c) is the weight required for subsequent weighting, n represents the dimension of the state vector, and α represents a [10 -4 ,1], λ is calculated as follows

[0107] λ=α 2 (L x +3-L x )-L x

[0108] The time update formula is:

[0109]

[0110] Among them, k refers to the k moment, refers to the state quantity sigma sampling point set, u refers to the input quantity, Refers to the prior prediction of the sigma point set, Refers to the prior prediction of the state quantity, W (c) Refers to the weight used for weighting the sigma point set, the subscript i refers to the i-th column of the matrix, P xx refers to the a priori estimated covariance matrix.

[0111] The observation update formula is:

[0112]

[0113] Where Q is the process noise covariance matrix; R is the measurement noise covariance matrix, refers to the prior prediction of the observation value sigma point set, h[] refers to the observation equation, is the prior prediction of the observed value, L x is a calculation parameter, which in the context of the present invention is equal to the dimension of the state variable, P yy (k|k-1) refers to the covariance matrix of the observations, P xy (k|k-1) refers to the covariance matrix of the state and observation quantities.

[0114] The Kalman gain calculation and update formula is:

[0115] K(k)=P xy (k|k-1)P yy (k|k-1) -1

[0116]

[0117] P(k|k)=P(k|k-1)-K(k)P yy (k|k-1)K(k) T

[0118] Where K is the Kalman gain, is the posterior state estimator at time k, and P(k|k) is the posterior state covariance matrix.

[0119] S104: estimating the longitudinal slip rate of the articulated wheeled vehicle based on the kinematic characteristics of the articulated wheeled vehicle and using the estimated speed state information and wheel speed information;

[0120] Through the above process, the velocity information of the center of mass of the vehicle can be obtained. On this basis, the longitudinal slip rate and sideslip angle of the wheels are calculated by introducing the four-wheel speed information as the final result output and the input parameter for the next moment estimation.

[0121] The formula for longitudinal slip rate is:

[0122]

[0123] The formula for the sideslip angle is:

[0124]

[0125] Where r e represents the wheel radius, ω ij Indicates the wheel speed, v xij Indicates the longitudinal velocity of each wheel center, v yij Indicates the longitudinal velocity of each wheel center, v xfl 、v xfr 、v xrl and v xrrThey represent the longitudinal speeds of the front left wheel, front right wheel, rear left wheel and rear right wheel, respectively, v yfl 、v yfr 、v yrl and v yrr Represents the lateral speed of the front left wheel, front right wheel, rear left wheel and rear right wheel respectively. xij 、v yij It is determined by the following formula. For the convenience of calculation, the longitudinal velocity and lateral velocity of the rear vehicle's center of mass are converted accordingly, as follows: Figure 2 As shown, it means as follows:

[0126]

[0127] where v xr is the longitudinal velocity of the rear vehicle’s center of mass, and v yr is the lateral velocity of the rear vehicle center of mass, and δ is the articulation angle.

[0128] By introducing the vehicle kinematic parameters including the rear body yaw rate, articulated steering angular velocity, vehicle center of mass velocity information, and vehicle geometric parameters as well as the measured four-wheel speed information, the longitudinal velocity and lateral velocity of the four-wheel center are calculated according to the kinematic relationship, as follows: Figure 2 As shown, it is determined by the following formula:

[0129] Longitudinal wheel speed v xij The formula is as follows:

[0130]

[0131] Lateral wheel speed v yij The formula is as follows:

[0132]

[0133] v yrl =v y -γ(l ra -l rb )

[0134] v yrr =v y -γ(l ra -l rb )

[0135] Where l ra Indicates the distance from the rear axle of the wheel to the hinge point, l rb Indicates the distance from the center of mass of the rear vehicle to the hinge point, T f Indicates the front wheel track, L fta , L rta Indicates the distance from the front and rear wheel axles to the articulation axis.

[0136] S105: After the calculation is completed, the estimation result of the longitudinal slip rate is output as output. At the same time, the longitudinal slip rate and sideslip angle of the four wheels will return to step S102 to calculate the tire force of the next estimation cycle, and then estimate the longitudinal slip rate information of the vehicle in the next cycle.

[0137] The present invention proposes a method for estimating the longitudinal slip rate of an articulated wheeled vehicle, proposes a simple and usable two-degree-of-freedom dynamic model based on the dynamic characteristics of the articulated wheeled vehicle, and provides an effective method for obtaining the vehicle speed state, providing a basic model basis for estimating the longitudinal slip rate.

[0138] In the method provided by the present invention, the Unscented Kalman algorithm is introduced to estimate the front and rear vehicle acceleration information, thereby effectively overcoming the result distortion caused by the high nonlinearity of the articulated vehicle dynamics model.

[0139] In the method proposed in the present invention, for any turning angle condition of an articulated wheeled vehicle, the vertical load and tire force are calculated in sequence through the vehicle moment balance relationship and the tire-road action relationship, and a two-degree-of-freedom dynamic model is introduced to calculate the vehicle speed information. Finally, the estimation of the four-wheel longitudinal slip rate under any turning angle condition is achieved, which overcomes the shortcomings of the existing articulated wheeled vehicle in state acquisition and can estimate the vehicle state more accurately.

[0140] The above is only a specific implementation of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial changes to the present invention using this concept shall be deemed as an infringement of the protection scope of the present invention.

Claims

1. A method for estimating the longitudinal slip rate of an articulated vehicle, characterized in that: The steps include: Articulated vehicle status data initialization; According to the dynamic characteristics of articulated wheeled vehicles, a two-degree-of-freedom vehicle model is built; The unscented Kalman filter algorithm is used to estimate the vehicle speed state information; estimating the longitudinal slip rate of the articulated wheeled vehicle based on the kinematic characteristics of the articulated wheeled vehicle and using the estimated speed state information and wheel speed information; At the beginning of the estimation, the data is initialized as follows: The vehicle state initialization formula is expressed as: [x,y]'=[N,N] Where x refers to the longitudinal velocity of the vehicle's total center of mass, y refers to the lateral velocity of the vehicle's total center of mass, and N refers to a very small constant that is not equal to 0; The state coordination initialization formula is expressed as: Where P0 refers to the initial state covariance, E[] represents the expected value calculation, and x0 refers to the true value of the state vector matrix at the initial moment. Refers to the estimated value of the state vector matrix x0 at the initial moment, x0 is expressed as follows: x0=[x,y]' The two-degree-of-freedom vehicle model is constructed as follows: where v x represents the longitudinal velocity of the vehicle's center of mass, v y represents the lateral velocity of the vehicle's center of mass, represents the longitudinal acceleration of the vehicle's center of mass, Indicates the lateral acceleration of the vehicle's center of mass; F xii With F yii Refers to the longitudinal and lateral tire forces, F xfl 、F xfr 、F xrl and F xrr They represent the longitudinal tire forces on the front left wheel, front right wheel, rear left wheel and rear right wheel respectively, and F yfl 、F yfr 、F yrl and F yrr Represent the lateral tire forces on the front left wheel, front right wheel, rear left wheel and rear right wheel respectively; T f and T r They represent the wheelbase of the front and rear wheels of the vehicle, γ represents the yaw rate, δ represents the articulation angle, is the angular velocity of the hinge angle, l fb , l rb They refer to the distance from the center of mass of the front vehicle body and the center of mass of the rear vehicle body to the articulated axis, m is the mass of the vehicle, m fb 、m rb are the front vehicle body mass and the rear vehicle body mass respectively; F xii With F yii It is expressed as: Where σ represents the longitudinal slip rate, α represents the sideslip angle, and λ t is a calculation process parameter, C σ , C α They represent the longitudinal and lateral stiffness of the tire, μ represents the ground friction coefficient, η is a calculation process parameter, and F z Indicates the vertical load of the tire; The unscented Kalman filter algorithm is used to estimate the speed state information of the vehicle, as follows: The steps of projecting the state quantity to the sigma point set are as follows: Among them, k refers to the kth moment, χ x (k-1) refers to the state quantity sigma sampling point set at time (k-1), Refers to the estimated value of the state quantity at time (k-1), L x represents the dimension of the state quantity, P(k-1) represents the state covariance at time (k-1); The weights required for sigma point weighting are: Where W (m) , W (c) is the weight required for subsequent weighting, n represents the dimension of the state vector, and α represents a [10 -4 ,1], λ is calculated as follows λ=a 2 (L x +3-L x )-L x The time update formula is: (x i x (k|k-1)) i =f((x x (k-1)) i ,u(k-1)) Among them, k refers to the k moment, refers to the state quantity sigma sampling point set, u refers to the input quantity, Refers to the prior prediction of the sigma point set, Refers to the prior prediction of the state quantity, W (c) Refers to the weight used for weighting the sigma point set, the subscript i refers to the i-th column of the matrix, P xx refers to the a priori estimated covariance matrix; The observation update formula is: Where Q is the process noise covariance matrix; R is the measurement noise covariance matrix, refers to the prior prediction of the observation value sigma point set, h[] refers to the observation equation, is the prior prediction of the observed value, P yy refers to the covariance matrix of the observations, P xy Refers to the covariance matrix of state and observation; The Kalman gain calculation and update formula is: K(k)=P xy (k|k-1)P yy (k|k-1) -1 P(k|k)=P(k|k-1)-K(k)P yy (k|k-1)K(k) T Where K is the Kalman gain, is the posterior state estimator at time k, and P(k|k) is the posterior state covariance matrix.

2. The longitudinal slip rate estimation method for an articulated vehicle according to claim 1, characterized in that: According to the kinematic characteristics of the articulated wheeled vehicle, the longitudinal slip rate and sideslip angle of the vehicle are estimated with the help of the estimated speed state information and wheel speed information, as follows: The formula for longitudinal slip rate is: The formula for the sideslip angle is: Where r e represents the wheel radius, ω ij Indicates the wheel speed, v xij Indicates the longitudinal velocity of each wheel center, v yij Indicates the lateral velocity of each wheel center, v xfl 、v xfr 、v xrl and v xrr They represent the longitudinal speeds of the front left wheel, front right wheel, rear left wheel and rear right wheel, respectively, v yfl 、v yfr 、v yrl and v yrr Respectively represent the lateral speeds of the front left wheel, front right wheel, rear left wheel and rear right wheel; v xij 、v yij Determined by the following formula; v xij The formula is as follows: v yij The formula is as follows: v yrl =v y -γ(l ra -l rb ) v yrr =v y -γ(l ra -l rb ) Where L fl Indicates the distance from the front left wheel center to the hinge point, L fr Indicates the distance from the front right wheel center to the hinge point, l ra Indicates the distance from the rear axle of the wheel to the hinge point, T f Indicates the front wheel track, l rb Indicates the distance from the center of mass of the rear vehicle body to the hinge point.

3. The longitudinal slip rate estimation method for an articulated vehicle according to claim 1, characterized in that: In the two-degree-of-freedom vehicle model, the vertical loads of the four wheels are calculated respectively by the vehicle moment balance relationship based on the ground, with the help of vehicle dynamics parameters and the vertical loads at the measured hinge points; the vertical load model is specifically as follows: Where L fr , L fl , L r Respectively refer to the longitudinal distance from the front right wheel, front left wheel, and rear wheel to the center of mass of the vehicle, W fr , W fl It refers to the lateral distance from the two front wheels of the vehicle to the hinge point, g represents the acceleration of gravity, and F f Indicates air resistance, F zd represents the vertical load at the hinge point, T r Rear wheel track, m f and m r Respectively represent the front and rear vehicle masses, a yf and a yr They represent the lateral acceleration of the front and rear vehicles respectively, and d refers to the distance from the center of mass of the vehicle to the ground.

4. The longitudinal slip rate estimation method for an articulated vehicle according to claim 1, characterized in that: The unscented Kalman filter algorithm is used to estimate the vehicle's speed state information. The state equation and measurement equation used are as follows: Equation of state It is expressed as: The measurement equation y(t) is expressed as: y(t)=h(x(t))=[a x ,a y ]+v(t) Among them, a x represents the longitudinal acceleration of the vehicle's total center of mass, a y It represents the lateral acceleration of the vehicle's center of mass.

Images