Road friction estimation method and device based on combination of vehicle formation and extended Kalman filtering

Abstract

The invention discloses a road friction estimation method and device based on combination of vehicle formation and extended Kalman filtering. The method comprises the following steps: designing a state equation and a measurement equation of road friction estimation; all the vehicles except the tail vehicle in the vehicle formation send the estimation information of the road adhesion coefficient to the following vehicles of the vehicles through vehicle-to-vehicle communication; all the vehicles except the head vehicle in the vehicle formation decide whether to utilize the estimation information of the front vehicle according to the estimation information of the vehicle on the road adhesion coefficient; all vehicles in the vehicle formation utilize an extended Kalman filter to estimate the current road adhesion coefficient, and the estimation process is to solve a state equation and a measurement equation based on estimation information of the road adhesion coefficient at the previous moment. According to the method, the extended Kalman filter added with the judgment module is designed, the extended Kalman filters of formation members are combined by using vehicle-to-vehicle communication of the vehicle formation, and the estimation speed and precision of the road adhesion coefficient of the following vehicle are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle platooning, and in particular to a method and device for estimating road friction by combining vehicle platooning with extended Kalman filtering. Background Art

[0002] With the increase in the number of vehicles, road congestion is becoming more and more serious. Vehicle platooning technology is one of the solutions to this problem. The collision-free driving requirements of vehicle platoons not only refer to avoiding collisions between the platoon and other vehicles, but also require that members within the platoon maintain a safe distance to avoid collisions. Under the conditions of stable road environment and smooth communication, the distance between members of the vehicle platoon can be kept consistent because the road conditions experienced by the members are the same. However, once the road conditions change, especially when the platoon is long, vehicles at different positions may encounter different road conditions. This requires the platoon to respond quickly to road changes and adjust the distance between vehicles in a timely manner to ensure the safe driving of the entire platoon. In addition, the safe distance between members of the vehicle platoon is directly related to the braking distance of each vehicle, which is directly affected by the road adhesion coefficient. Therefore, a fast and accurate estimation of the road adhesion coefficient is crucial for adjusting the safe distance between platoon members. However, the road adhesion coefficient state estimator extended Kalman filter or unscented Kalman filter currently used in a single vehicle has a slow estimation speed, which makes the safety of the vehicle platoon where the road adhesion coefficient changes cannot be guaranteed. By improving the estimation method of the road adhesion coefficient, the driving safety and efficiency of vehicle platooning can be significantly improved, especially under variable road conditions. Summary of the invention

[0003] The purpose of the present invention is to solve the problems in the prior art.

[0004] The technical solution adopted by the present invention to solve the technical problem is: to provide a road friction estimation method combining vehicle platooning with extended Kalman filtering, comprising the following steps:

[0005] Design state equations and measurement equations for road friction estimation;

[0006] All vehicles except the last vehicle in the platoon send their estimated road adhesion coefficient information to their following vehicles through vehicle-to-vehicle communication; all vehicles except the leading vehicle in the platoon decide whether to use the estimated information of the preceding vehicle based on their own estimated road adhesion coefficient information;

[0007] All vehicles in the vehicle formation use the extended Kalman filter to estimate the current road adhesion coefficient. The estimation process is to solve the state equation and measurement equation based on the estimated information of the road adhesion coefficient at the previous moment.

[0008] Preferably, the state equation and measurement equation for designing road friction estimation include:

[0009] The 7-DOF vehicle model is used to construct the system state equation for each vehicle, expressed as:

[0010]

[0011] Among them, a x , a y are the longitudinal acceleration and lateral acceleration respectively, m is the vehicle mass; F x and F y are the longitudinal force and lateral force of the tire respectively. The subscripts fl, fr, rl and rr represent the left front wheel, right front wheel, left rear wheel and right rear wheel respectively. δ is the front wheel turning angle, r is the yaw velocity, is the yaw acceleration; I z is the moment of inertia of the vehicle around the z axis, l f and l r are the distances from the center of mass of the vehicle to the front axle and the rear axle, respectively, and b f and b r They are the front and rear track respectively;

[0012] The measurement equation of the system is constructed based on the system state equation, which is expressed as:

[0013] y(t)=h(x(t),u(t),v(t));

[0014] Among them, the state variable x(t) = [μ fl μ fr μ rl μ rr ] T , including the road adhesion coefficient μ fl , μ fr , μ rl , μ rr ; Input variables and are the normalized tire force; v(t) is the measurement noise; the measurement is [a x a y r] T .

[0015] Preferably, the process of estimating the road adhesion coefficient by using the extended Kalman filter comprises the following steps:

[0016] The correction module calculates a first state estimate and updates an error matrix;

[0017] The judgment module combines the first state estimate output by the vehicle correction module and the first state estimate output by the preceding vehicle correction module to calculate the second state estimate.

[0018] The prediction module calculates the current predicted state variable and the error matrix based on the first state estimate, the second state estimate and the updated error matrix;

[0019] The correction module calculates the first state estimate and updates the error matrix. The input data for the first calculation is the initial condition, which is the initial condition of the road adhesion coefficient set manually. The input for subsequent calculations is the current predicted state variable and error matrix output by the prediction module and the current measurement value.

[0020] Preferably, the calculation of the second state estimate is expressed as:

[0021]

[0022] in, represents the second state estimate output by the judgment module to the prediction module; x(t) represents the first state estimate output by the correction module; n+1 represents that the current vehicle is the n+1th vehicle in the vehicle formation, and n represents that the vehicle following the current vehicle is the n+1th vehicle in the vehicle formation; Y represents the logical value given by the judgment module, Y=1 indicates that the current vehicle enters the new road, and Y=0 indicates that the current vehicle has not yet entered the new road.

[0023] Preferably, the assignment of the logic value Y is specifically:

[0024] If the product of (x(t+3)-x(t+2)) / T and (x(t+2)-x(t+1)) / T is greater than zero, Y=1; otherwise Y=0.

[0025] Preferably, the assignment of the logic value Y is specifically:

[0026] When (x(t+3)-x(t+2)) / T>k and (x(t+2)-x(t+1)) / T>k hold, and (x(t+1)-x(t)) / T>k does not hold, Y=1; T represents a time interval;

[0027] In other cases, Y=0.

[0028] Preferably, the vehicle-to-vehicle communication of the vehicle formation is only between adjacent front and rear vehicles, the front vehicle only sends information to the rear vehicle, and the rear vehicle only receives information from the front vehicle.

[0029] The present invention also provides a road friction estimation method combining vehicle platooning with extended Kalman filtering, comprising:

[0030] Equation building module, designing the state equation and measurement equation of the road friction estimation system;

[0031] Information communication module: all vehicles except the last vehicle in the platoon send their estimated information of the road adhesion coefficient to their following vehicles through vehicle-to-vehicle communication; all vehicles except the leading vehicle in the platoon decide whether to receive the estimated information of the leading vehicle based on their estimated information of the road adhesion coefficient;

[0032] In the friction estimation module, the extended Kalman filters of all vehicles in the vehicle formation use the estimated information of the road adhesion coefficient at the previous moment to solve the state equation and the measurement equation to estimate the current road adhesion coefficient.

[0033] The present invention has the following beneficial effects: the present invention proposes a road adhesion coefficient estimation method based on a vehicle formation and an extended Kalman filter. In view of road conditions where the road adhesion coefficient changes, an extended Kalman filter with an added judgment module is used, and the vehicle-to-vehicle communication of the vehicle formation is coordinated to speed up the estimation speed of the new road adhesion coefficient by the following vehicle. Compared with the traditional single-vehicle estimation method, the estimation accuracy of the following vehicle is also improved, and more accurate information is provided for the longitudinal spacing control of the vehicle formation.

[0034] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments, but the present invention is not limited to the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A method step diagram of an embodiment of the present invention;

[0036] Figure 2 This is a structural diagram of an extended Kalman filter according to an embodiment of the present invention;

[0037] Figure 3 A vehicle-to-vehicle communication topology diagram according to an embodiment of the present invention;

[0038] Figure 4 FIG. 4 is a diagram showing the structure of a device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] See also Figure 1 FIG. 1 is a diagram showing steps of a method according to an embodiment of the present invention, comprising the following steps:

[0040] S101, design state equations and measurement equations for road friction estimation;

[0041] S102, all vehicles except the rear vehicle in the vehicle formation send their own estimated information of the road adhesion coefficient to their following vehicles through vehicle-to-vehicle communication; all vehicles except the front vehicle in the vehicle formation decide whether to use the estimated information of the front vehicle based on their own estimated information of the road adhesion coefficient;

[0042] S103, all vehicles in the vehicle formation use the extended Kalman filter to estimate the current road adhesion coefficient, and the estimation process is to solve the state equation and the measurement equation based on the estimation information of the road adhesion coefficient at the previous moment.

[0043] Specifically, the process of estimating the road adhesion coefficient by the extended Kalman filter is as follows: Figure 2 As shown, the following steps are included:

[0044] S1031, the correction module calculates the first state estimate and updates the error matrix;

[0045] S1032, the judgment module calculates a second state estimate based on the first state estimate output by the vehicle correction module and the first state estimate output by the preceding vehicle correction module.

[0046] S1033, the prediction module calculates the current prediction state variable and the error matrix based on the first state estimation, the second state estimation and the updated error matrix;

[0047] Specifically, the correction module calculates the first state estimate and updates the error matrix. The input data for the first calculation is the initial condition, which is the initial condition of the road adhesion coefficient set manually. The input for subsequent calculations is the current predicted state variable and error matrix output by the prediction module and the current measurement value.

[0048] Specifically, see Figure 3 As shown, it is a communication topology diagram of an embodiment of the present invention. The vehicle-to-vehicle communication of the vehicle formation is only between the adjacent front and rear vehicles. The front vehicle only sends information to the rear vehicle, and the rear vehicle only receives information from the front vehicle; and the extended Kalman filter of the first vehicle (leader vehicle) of the vehicle formation does not need to be provided with a judgment module. The result of the correction module of the front vehicle is transmitted to the judgment module of the extended Kalman filter of the following vehicle through the vehicle-to-vehicle communication in the vehicle formation. The judgment module of the extended Kalman filter of the following vehicle determines whether to enter a new road surface according to its own estimation result of the road adhesion coefficient.

[0049] Specifically, the judgment result of the judgment module is to assign a logical value Y. In order to prevent the extended Kalman filter estimation fluctuation from causing the logical value to be 1, it is necessary to satisfy that the product of (x(t+3)-x(t+2)) / T and (x(t+2)-x(t+1)) / T is greater than zero, thereby avoiding the fluctuation of the extended Kalman filter when entering a new road surface.

[0050] The assignment logic of the embodiment of the present invention is shown in Table 1, which is a logic table of the judgment module.

[0051] Table 1

[0052]

[0053] It can be seen from Table 1 that when (x(t+3)-x(t+2)) / T>k and (x(t+2)-x(t+1)) / T>k are established, and (x(t+1)-x(t)) / T>k is not established, Y=1; T represents a time interval; in other cases, Y=0; k is the set threshold, k≠0.

[0054] When the vehicle's own estimation result satisfies Y=1, it is judged that it has entered a new road surface. At this time, it receives and uses the estimation result of the new road surface from the preceding vehicle to update the vehicle's own estimation result. The judgment conditions are as shown in the table, that is, the following vehicle only receives the information once. The details are as follows:

[0055]

[0056] The n in the upper left corner indicates the vehicle number. It represents the input of the extended Kalman filter prediction module, and Y represents the logical value of the extended Kalman filter judgment module. When the rate of change between two consecutive estimated values ​​exceeds the set threshold, it is assigned a value of 1, otherwise it is assigned a value of 0.

[0057] Specifically, the S101 includes the following steps:

[0058] The 7-DOF vehicle model is used to construct the system state equation for each vehicle, expressed as:

[0059]

[0060]

[0061] Among them, a x , a y are longitudinal acceleration and lateral acceleration respectively, m is the vehicle mass, F x , F y are the longitudinal force and lateral force of the tire, fl, fr, rl, rr are the left front, right front, left rear, and right rear wheels, δ is the front wheel turning angle, r is the yaw rate, I z is the moment of inertia of the vehicle around the z axis, l f , l r are the distances from the center of mass of the vehicle to the front axle and the rear axle, respectively, and b f , b r They are the front and rear track respectively;

[0062] The measurement equation of the system is constructed as follows:

[0063] y(t)=h(x(t),u(t),v(t));

[0064] Among them, the quantity measurement y(t) = [a x a y r]T ; Input variables is the normalized tire force; state variable x(t) = [μ fl μ fr μ rl μ rr ] T ; v(t) is the measurement noise;

[0065] Estimation of road adhesion coefficient μ by extended Kalman filter algorithm fl , μ fr , μ rl , μ rr .

[0066] See also Figure 4 FIG. 1 is a system structure diagram of an embodiment of the present invention, including:

[0067] An equation building module 401 designs state equations and measurement equations of a road friction estimation system;

[0068] In the information communication module 402, all vehicles except the rear vehicle in the vehicle formation send their estimated information of the road adhesion coefficient to their following vehicles through vehicle-to-vehicle communication; all vehicles except the leading vehicle in the vehicle formation decide whether to receive the estimated information of the leading vehicle based on their estimated information of the road adhesion coefficient;

[0069] In the friction estimation module 403, the extended Kalman filters of all vehicles in the vehicle formation use the estimated information of the road adhesion coefficient at the previous moment to solve the state equation and the measurement equation to estimate the current road adhesion coefficient.

[0070] The present invention proposes a method for estimating the road adhesion coefficient based on a vehicle formation combined with an extended Kalman filter. In view of the estimation of the road adhesion coefficient by the vehicle formation, an extended Kalman filter with a judgment module is designed, and then the extended Kalman filters of the formation members are combined by using the vehicle-to-vehicle communication method of the vehicle formation. This method can not only speed up the estimation of the new road adhesion coefficient by the following vehicle of the vehicle formation, but also improve the estimation accuracy of the following vehicle. Compared with the traditional single vehicle estimation, this method does not need to install additional sensors, saves costs, and also provides more accurate information for the longitudinal spacing control of the vehicle formation.

[0071] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A road friction estimation method combining vehicle platooning with extended Kalman filtering, characterized in that: The following steps are involved: Design state equations and measurement equations for road friction estimation; All vehicles except the last vehicle in the platoon send their estimated road adhesion coefficient information to their following vehicles through vehicle-to-vehicle communication; all vehicles except the leading vehicle in the platoon decide whether to use the estimated information of the preceding vehicle based on their own estimated road adhesion coefficient information; All vehicles in the vehicle formation use the extended Kalman filter to estimate the current road adhesion coefficient. The estimation process is to solve the state equation and measurement equation based on the estimated information of the road adhesion coefficient at the previous moment.

2. The method for estimating road friction by combining vehicle platooning with extended Kalman filtering according to claim 1, characterized in that: The state equation and measurement equation for designing road friction estimation include: The 7-DOF vehicle model is used to construct the system state equation for each vehicle, expressed as: Among them, a x , a y are the longitudinal acceleration and lateral acceleration respectively, m is the vehicle mass; F x and F y are the longitudinal force and lateral force of the tire respectively. The subscripts fl, fr, rl and rr represent the left front wheel, right front wheel, left rear wheel and right rear wheel respectively. δ is the front wheel turning angle, r is the yaw velocity, is the yaw acceleration; I z is the moment of inertia of the vehicle around the z axis, l f and l r are the distances from the center of mass of the vehicle to the front axle and the rear axle, respectively, and b f and b r They are front and rear track respectively; The measurement equation of the system is constructed based on the system state equation, which is expressed as: y(t)=h(x(t),u(t),v(t)); Among them, the state variable x(t)=[μ fl μ fr μ rl μ rr ] T , including the road adhesion coefficient μ fl , μ fr , μ rl , μ rr ; Input variables and are the normalized tire force; v(t) is the measurement noise; the measurement is [a x a y r] T .

3. The method for estimating road friction by combining vehicle platooning with extended Kalman filtering according to claim 1, characterized in that: The process of estimating the road adhesion coefficient by using the extended Kalman filter comprises the following steps: The correction module calculates a first state estimate and updates an error matrix; The judgment module combines the first state estimate output by the vehicle correction module and the first state estimate output by the preceding vehicle correction module to calculate the second state estimate. The prediction module calculates the current predicted state variable and the error matrix based on the first state estimate, the second state estimate and the updated error matrix; The correction module calculates the first state estimate and updates the error matrix. The input data for the first calculation is the initial condition, which is the initial condition of the road adhesion coefficient set manually. The input for subsequent calculations is the current predicted state variable and error matrix output by the prediction module and the current measurement value.

4. The method for estimating road friction by combining vehicle platooning with extended Kalman filtering according to claim 1, characterized in that: The second state estimation is calculated as: in, represents the second state estimate output by the judgment module to the prediction module; x(t) represents the first state estimate output by the correction module; n+1 represents that the current vehicle is the n+1th vehicle in the vehicle formation, and n represents that the vehicle following the current vehicle is the n+1th vehicle in the vehicle formation; Y represents the logical value given by the judgment module, Y=1 indicates that the current vehicle enters the new road, and Y=0 indicates that the current vehicle has not yet entered the new road.

5. The method for estimating road friction by combining vehicle platooning with extended Kalman filtering according to claim 1, characterized in that: The assignment of the logical value Y is specifically: If the product of (x(t+3)-x(t+2)) / T and (x(t+2)-x(t+1)) / T is greater than zero, Y=1; otherwise Y=0.

6. The method for estimating road friction by combining vehicle platooning with extended Kalman filtering according to claim 1, characterized in that: The assignment of the logical value Y is specifically: When (x(t+3)-x(t+2)) / T>k and (x(t+2)-x(t+1)) / T>k hold, and (x(t+1)-x(t)) / T>k does not hold, Y=1; T represents a time interval; In other cases, Y=0.

7. The method for estimating road friction by combining vehicle platooning with extended Kalman filtering according to claim 2, characterized in that: The vehicle-to-vehicle communication in a vehicle platoon is only between the adjacent front and rear vehicles. The front vehicle only sends information to the rear vehicle, and the rear vehicle only receives information from the front vehicle.

8. A road friction estimation method combining vehicle platooning with extended Kalman filtering, characterized in that: include: Equation building module, designing the state equation and measurement equation of the road friction estimation system; Information communication module: all vehicles in the platoon, except the rear vehicle, send their estimated road adhesion coefficient information to their following vehicles through vehicle-to-vehicle communication; All vehicles in the platoon, except the lead vehicle, decide whether to receive the estimated information of the preceding vehicle based on their own estimated information of the road adhesion coefficient; In the friction estimation module, the extended Kalman filters of all vehicles in the vehicle formation use the estimated information of the road adhesion coefficient at the previous moment to solve the state equation and the measurement equation to estimate the current road adhesion coefficient.

Images