DE102005060219A1 - Estimating method for rating a friction coefficient between a road and a motor vehicle's tires uses the friction coefficient to enlarge a Kalman filter by the friction coefficient - Google Patents

Abstract

Front (lf) and rear (lr) wheel clearance equals the clearance between the front (22) and rear (21) wheels and a motor vehicle's mass center of gravity (CG). Front (alpha f) and rear (alpha r) slippage angle indicates the angle between front and rear wheel axles and their direction of movement. An independent claim is also included for a control device for a motor vehicle with an estimator.

Description

The The invention relates to a method for estimating the friction coefficient between the street and a tire of a motor vehicle and a control device for a Motor vehicle, which one set up for carrying out the method Estimator contains.

For the control of automobiles, electronic systems are increasingly being used, which complement or partially replace the driver's direct control. Examples include ESP (electronic stability program), EPAS (electric power assisted steering), E diff (electric differential) called. To fulfillment of their functions Such systems the current state of the motor vehicle. There not all required state variables of a simple measurement accessible are, and the number of sensors for cost reasons is limited, they must be estimated partly with the help of vehicle models, the estimates adjusted by comparison with measured values of observable quantities can be.

Farther is problematic that some state variables are not constant. For example the tire connects the vehicle with the road, and they form the forces that which are necessary to brake, accelerate and the direction to change. The contact between tires and the road is therefore not over the Time constant. Hung of various parameters, such as the type of tire used, tire pressure, tire tread and friction between tires and street from. The coefficient of friction varies from 0.1 on ice up to 1.0 dry asphalt. He is part of the vehicle models used for estimation algorithms be used. Such an estimate can be wrong appraisal the lateral forces on the wheels to lead. In the worst case, the vehicle model uses a constant one Friction coefficient, which is equal to 1, and the real vehicle is moving on ice. this leads to to overestimate the tire forces by a factor of 10.

each Control using such a false estimate is due the deviation between the reality and the estimate fail.

Of the Invention is therefore based on the object means for a reliable and at the same time fast estimation the condition of the motor vehicle, in particular the friction coefficient between the street and tires of a vehicle and the lateral speed.

These The object is achieved by the method according to claim 1 and the control device solved according to claim 10.

According to one In the first aspect, the invention relates to a method for estimating the Friction coefficient between road and tires of a motor vehicle, wherein the friction coefficient under Extension of a Kalman filter is estimated to the coefficient of friction on the assumption that the skew stiffness of the tire is proportional to the coefficient of friction.

The Assuming that the skew stiffness of the tire is proportional to the coefficient of friction is based on assuming that the skew stiffness of the tire due to friction between tires and road such changed is that the determined slip stiffness of the tire, which describes the forces caused by the contact between the road and the tire, from the product of the actual Skew stiffness of the Tire and the friction coefficient.

The appraisal of the friction coefficient ensures the correct estimation of the lateral velocity of the Vehicle at any road condition, as the lateral speed to a great extent depends on the friction coefficient. The estimated Lateral speed is usually used to the lateral forces to determine on each tire. A high accuracy of the lateral speed is therefore for Applications such as ESP, ARC (active roll control) or AFS (active front steering). The method according to the invention makes it possible both in the linear region and in the border region of the Tire a correct estimate to achieve the coefficient of friction.

In a preferred embodiment, the Kalman filter comprises a one-track model. The one-track model is used in the Kalman filter as a virtual model of the real motor vehicle. It provides the basic dynamics of the vehicle and contains all the necessary states and parameters to obtain an estimate of the friction coefficient. In the single-track model, the two front and two rear wheels are simplified to a front and a rear wheel. The center of gravity is set to zero. It will be on assuming that the longitudinal velocity is constant.

In an embodiment of the present invention will be or the lateral acceleration, the steering wheel angle, the yaw rate and / or the longitudinal speed measured and is an input variable or are input variables in the Single-track model. The steering wheel angle is an input that is determined by the driver.

The at least one measurement of the at least one above measured Sizes can be performed by at least one sensor of an ESP system.

Preferably serve the measurements of the lateral acceleration, the steering wheel angle, the yaw rate and / or the longitudinal speed as reference values. The reference values can be used to simulate different driving situations.

To the Estimate of the friction coefficient is advantageously a vehicle state, the in the context of the invention by the friction coefficient, the Lateralge speed and the yaw rate is defined at a given time based on a vehicle condition at a previous time determined.

Prefers the vehicle state is based on the previous time a measurement of at least the lateral acceleration and the yaw rate determined. That is, from a sensor system of the vehicle is only requires the measurement of two variables, Reliable by the coefficient of friction estimate.

The inventive method for estimation the coefficient of friction can be carried out in particular online, so that the current friction coefficient is always available.

The Invention further relates to a control device for a motor vehicle, that contains an estimator, which is adapted to a method of the above-described Kind of execute. That is, the estimator is capable of a coefficient of friction between Tire of a motor vehicle and road with the help of an extended Estimate Kalman filters.

The Control device is for a variety of different tasks in the context of control and monitoring a motor vehicle usable. In particular, the control device be used to electronic vehicle stabilization perform.

in the The invention will now be described with reference to the attached figure explained in more detail.

It shows:

1 a schematic drawing of an embodiment of a single-track model used in the invention.

1 shows a schematic drawing of a single track model, which can be used according to the present invention. The single-track model is the result of simplifying the two front wheels and the two rear wheels to a front wheel 22 and a rear wheel 21 , Slope, roll and lift movements are neglected. The CG focus is zero. The driver input is the wheel angle (δ). It is assumed that the longitudinal velocity is a constant or a little changing variable.

Like in the 1 is shown, the tire force F _r acts on the rear wheel 21 , And the tire force F _f acts on the front wheel 22 , The distance I _r is the distance of the rear wheel 21 to the center of mass CG of the vehicle, while the distance I _{f of} the distance of the front wheel 22 to the center of mass CG of the vehicle. The wheel slip angle α _f gives the angle between the wheel axle of the front wheel 22 and the direction of movement of the front wheel 22 while the wheel slip angle α _{r is} the angle between the wheel axle of the rear wheel 21 and the direction of movement of the rear wheel 21 indicates. Further influencing parameters in the single-track model are the skew stiffness of the tire c _{r of} the rear wheel 21 , and the skew stiffness of the tire c _{f of} the front wheel 22 , Further influencing variables in the single-track model are the yaw rate Ψ, the longitudinal velocity ν and the lateral velocity ν _Y. Furthermore, the coordinate system x, y, z is shown schematically.

definiert, wobei ν_y die Lateralgeschwindigkeit ist, Ψ . die Gierrate ist, c_f bzw. c_r die Schräglaufsteifigkeit des Reifens des Vorder- bzw. Hinterreifens ist, I_f bzw. I_r der Abstand der Vorder- bzw. Hinterachse und dem Massenzentrum des Fahrzeuges ist, I_Z das Trägheitsmoment des Fahrzeuges ist, m die Masse des Fahrzeuges ist, ν_x die Längsgeschwindigkeit ist, δ_w der Lenkradwinkel ist, und t die Zeit darstellt.The single-track model is in its continuous form, in particular by the matrix equation:

where ν _{y is} the lateral velocity, Ψ. the yaw rate is, c _f or c _{r is} the slip stiffness of the tire of the front and rear tires, I _f and I _{r is} the distance between the front and rear axle and the center of mass of the vehicle, I _{Z is} the moment of inertia of the vehicle , m is the mass of the vehicle, ν _{x is} the longitudinal speed, δ _{w is} the steering wheel angle, and t is the time.

The coefficients c _r , c _{f provide} the forces provided by the contact between the tire and the road and represent the tire slip resistance. The tire slip stiffnesses are parameters in the single track model describing the tire forces. The tire forces can be explained by the formulas F yf = c f · α f and Fyr = c r · α r . where F _{yf is} the front tire force, c _{f is} the tire front tire slip, α _{f is} the wheel slip angle of the front tire, F _{yr is} the tire rear tire force, c _{r is} the tire rear tire slip stiffness, and α _{r is} the wheel slip angle of the tire Rear tire is.

In the context of the invention it is assumed that the tire forces are changed due to friction, ie, increased or decreased. Therefore, the tire slip resistance changes with the current road condition. This leads to the following assumption: F yf = μ · c f · α r and F yr = μ · c r · α r . where μ is the friction coefficient and the other variables are as defined above, wherein the determined skew stiffness of the tire c ' _f or c' _r is composed of the product of the coefficient of friction μ and the actual skew stiffness of the tire c _f or c _r , so : c ' f = μ · c f and c ' r = μ · c r

That is, to include the friction coefficient in the tire gradient is set: c f + c r → μ · (c f + c r ) ie the friction coefficient is an additional state in the estimation algorithm. This includes the assumption that the tire slip stiffness includes the coefficient of friction. This, in turn, means that the tire's tire slip resistance, which describes the lateral forces in the linear range of tire forces, is affected by a change in road-tire contact.

The state equation resulting from this change is nonlinear, that is, the Kalman gains are not constant over time, and there is no transmission factor solution because the solution depends on the current state of the vehicle. The solution is therefore advantageously calculated online by means of a transfer function (or transfer function) which transfers the state to a state t _{k + 1} at a time t _k .

wobei ν_x die Längsgeschwindigkeit darstellt, ν_y die Lateralgeschwindigkeit darstellt, Ψ . die Gierrate darstellt, μ den Reibkoeffizienten darstellt, T eine Transferfunktion darstellt, I_Z das Trägheitsmoment des Fahrzeugs darstellt, c die Schräglaufsteifigkeit des Reifens darstellt, I den Abstand zwischen Massenzentrum und der Achse des Fahrzeugs darstellt, m die Masse des Fahrzeugs darstellt, δ den Radwinkel darstellt, t die Zeit darstellt, der Index I für Vorderreifen steht, der Index r für Hinterreifen steht, und der Index k für den momentanen Zeitpunkt steht.The state space description is then performed using the transfer function according to the matrix equation:

where ν _{x represents} the longitudinal velocity, ν _{y represents} the lateral velocity, Ψ. represents the yaw rate, μ represents the friction coefficient, T represents a transfer function, I _{Z represents} the moment of inertia of the vehicle, c represents the slip stiffness of the tire, I represents the distance between the center of mass and the axis of the vehicle, m represents the mass of the vehicle, δ the Wheel angle, t represents the time, the index I stands for front tire, the index r stands for rear tire, and the index k stands for the current time.

The determination of the friction coefficient at time t _{k + 1} takes place on the basis of a vehicle condition at time t _k , which in turn is obtained from the lateral acceleration measured for time t _k and the yaw rate measured for time t _k .

wobei a_Y die Lateralbeschleunigung darstellt, wobei ν die Längsgeschwindigkeit darstellt, Ψ . die Gierrate darstellt, μ den Reibkoeffizienten darstellt, c die Schräglaufsteifigkeit des Reifens darstellt, I den Abstand zwischen Massenzentrum und der Achse des Fahrzeugs darstellt, m die Masse des Fahrzeugs darstellt, δ_W den Radwinkel darstellt, t die Zeit darstellt, der Index I für Vorderreifen steht, der Index r für Hinterreifen steht, und der Index k für den momentanen Zeitpunkt steht.The lateral acceleration and the yaw rate are measured, for example by means of a sensor arrangement from the ESP system. As an intermediate result, which is used for determining the vehicle state at time t _k , is then obtained:

where a _{Y represents} the lateral acceleration, where ν represents the longitudinal velocity, Ψ. represents the yaw rate, μ represents the friction coefficient, c represents the slip stiffness of the tire, I represents the distance between the center of mass and the axis of the vehicle, m represents the mass of the vehicle, δ _{W represents} the wheel angle, t represents the time, the index I for Front tire stands, the index r stands for rear tire, and the index k stands for the current time.

Claims (11)

Method for estimating the coefficient of friction between the street and a tire of a motor vehicle, wherein the coefficient of friction under extension of a Kalman filter by the friction coefficient estimated on the assumption will that skew stiffness of the tire is proportional to the coefficient of friction. Method according to claim 1, characterized in that that the Kalman filter comprises a single-track model. Method according to claim 1 or 2, characterized that the lateral acceleration and / or the wheel angle and / or the Yaw rate and / or the longitudinal speed is measured and be and an input variable or input variables in the Single track model. Method according to one of claims 1 to 3, characterized that the at least one measurement of at least one sensor of a ESP system performed becomes. Method according to one of claims 1 to 4, characterized that measurements of the lateral acceleration, the wheel angle, the yaw rate and / or the longitudinal speed serve as reference values. Method according to claim 5, characterized in that that the reference values for the simulation of different driving situations serve. Method according to one of claims 1 to 6, characterized that to treasure of the friction coefficient on a vehicle state at a time Basis of a vehicle condition at a previous time is determined. Method according to claim 3 and claim 7, characterized characterized in that the vehicle state at the previous time on the basis of a measurement of at least the lateral acceleration and the yaw rate is determined. Method according to one of claims 1 to 8, characterized that the estimate of the coefficient of friction is carried out online. Control device for a motor vehicle, comprising an estimator, which is adapted to a method according to at least one of the claims 1 to 9 Control device according to claim 10, characterized characterized in that it is designed to be an electronic Perform vehicle stabilization.