Patent Application
20220073041
The invention relates to a method for the traction control of a single-track motor vehicle.
Various methods for the traction control of both two-track motor vehicles, such as passenger vehicles, and also for single-track motor vehicles, such as motorcycles, are known in the prior art. The traction control is also called a traction control system (TCS) or also automatic slip regulation (ASR), wherein the various systems popular in the market for traction control have supplier-specific names.
In particular in the case of sudden acceleration processes or in the case of an underlying surface having low adhesive friction, for example as in the case of ice, snow, loose gravel, or wet conditions, the traction control prevents a drive wheel and thus in the case of motorcycles the rear wheel from spinning and the vehicle from breaking away laterally.
For this purpose, the drive torque on the rear wheel is deliberately influenced by the traction control by engaging in the engine management and thus the drive force of the engine is influenced.
The methods known in the prior art for traction control are usually based in motor vehicles solely on the wheel slip, i.e., the ratio of the speed of a driven wheel to the speed of a (hypothetical) nondriven wheel, which therefore rotates in a formfitting manner. In the case of single-track motor vehicles, a vehicle inclination is moreover often taken into consideration, due to which the tire properties or wheel properties of the motor vehicle can change and the determination of the wheel slip is influenced.
In addition, however, in single-track motor vehicles, the slip angle of the rear wheel is also relevant for the wheel slip and for a traction control on the drive wheel or rear wheel, but it is usually only acquirable by a complex and costly sensor system and is therefore not or is only rarely taken into consideration in the traction control. The slip angle of a wheel is understood as the angle between the velocity vector of the wheel in the wheel contact point on the roadway and the intersection line between wheel center plane and roadway plane.
The slip angle is not used for the traction control of a single-track motor vehicle in the majority of the known methods. In addition, in some methods in which the slip angle is taken into consideration, only the yaw angle of the motor vehicle is used for the indication of the location in space for a determination of the slip angle, so that a slip angle determined therefrom can be incorrect, since a motorcycle or a single-track motor vehicle has further degrees of freedom relevant for the traction control and for the slip angle, in particular pitch and roll angles. An incorrect slip angle can propagate in the traction control and cause incorrect or unpredictable behavior of the vehicle.
The invention is therefore based on the object of overcoming the above-mentioned disadvantages and providing a method for the traction control taking the slip angle of the rear wheel into consideration, wherein the slip angle is to be ascertainable easily and cost-effectively.
This object is achieved by the combination of features according to claim 1.
According to the invention, a method is proposed for determining a slip angle λr of a rear wheel of a single-track motor vehicle for the traction control of the rear wheel of the single-track motor vehicle. The traction control is carried out by a control loop, wherein the slip angle λr of the rear wheel is a feedback variable of the control loop. For the determination of the slip angle λr, it is provided that it is determined by at least one of the following steps:
In addition, the method according to the invention provides that the first slip angle λr1, the second slip angle λr2, or the third slip angle λr3 represents the slip angle λr. Alternatively thereto, the method provides that the slip angle λr is determined from at least two of the slip angles λr1, λr2, and λr3.
The basic concept is to determine the slip angle first in a model-based manner or by way of a state estimator in which a model of the vehicle is stored instead of directly measuring the slip angle.
However, if there is an excessive inaccuracy or tolerance of the slip angles λr1, λr2, λr3 determined in a model-based manner or if a high accuracy of the slip angle λr is required for the traction control to be implemented, it is thus provided that two of the slip angles λr1, λr2, λr3 or all slip angles λr1, λr2, λr3 are used to determine the slip angle λr.
If the accuracy or the correctness of the individual slip angles λr1, λr2, λr3 is inadequate or doubtful, the concept according to the invention for determining the slip angle λr is therefore preferably to first determine the slip angle λr in at least two model-based and in particular three different ways, which are each implementable cost-effectively, however, and in particular by way of state estimators, in each of which a model of the vehicle is stored, whereby the slip angles λr1, λr2, λr3 result, which each represent the slip angle λr as such, but can deviate from an actual slip angle at the rear wheel. To reduce the deviation, for error recognition, and to increase the accuracy, it is subsequently provided that a slip angle λr for the control of the drive torque on the rear wheel or for the traction control is to be calculated from the “estimated” or model-based slip angles λr1, λr2, λr3. The slip angle λr determined from at least two of the slip angles λr1, λr2, λr3 corresponds to the actual slip angle existing on the rear wheel more accurately than the individual slip angles λr1, λr2, λr3.
In particular in control technology, so-called state estimators, which are also referred to as state observers, are known, by way of which it is possible to approximately determine variables which cannot be directly observed or measured.
The steering angle δ is preferably acquired by a steering angle sensor on the front wheel of the single-track motor vehicle.
One advantageous refinement of the method provides that the first state estimator is a Kalman filter or alternatively an expanded Kalman filter.
In addition, it is advantageous if the second state estimator is a Kalman filter or an expanded Kalman filter, wherein variables of nonlinear systems can be observed or estimated by the expanded Kalman filter.
The orientation of the motor vehicle in space as an input variable of the first state estimator is determined in a likewise advantageous embodiment by a roll angle Φ, a yaw angle Ψ, and a pitch angle Θ of the motor vehicle.
In this case, the yaw angle Ψ describes the orientation of the vehicle around the z axis or the vertical axis of the vehicle. The roll angle Φ describes the orientation of the vehicle around the x axis or the longitudinal axis of the vehicle and the pitch angle Θ describes the orientation around the y axis or the transverse axis orthogonal to the longitudinal axis of the vehicle.
One advantageous variant of the method moreover provides that a curve radius R of a curve described by the motor vehicle is determined and the curve radius R is an input variable of the first state estimator.
It is moreover also advantageous if the curve radius R and the orientation of the motor vehicle are determined from a roll rate {dot over (ϕ)}, a yaw rate {dot over (ψ)}, and a pitch rate {dot over (θ)} as well as an acceleration of the motor vehicle in space and a vehicle velocity v. The vehicle velocity v is preferably a representative vehicle velocity, which is ascertained from the wheel velocities in consideration of a contact loss or a lift-off detection. The acceleration of the motor vehicle in space preferably corresponds to the acceleration in the direction of the spatial axes, so that the acceleration of the vehicle in space is composed of the accelerations ax, ay, and az.
An inertial measurement unit (IMU), which is present in the vehicle and is provided for further control systems in the vehicle in any case, is preferably used for the ascertainment or the measurement of the roll rate {dot over (ϕ)}, the yaw rate {dot over (ψ)}, and the pitch rate {dot over (θ)}, as well as the accelerations ax, ay, and az.
For this purpose, an IMU usually has acceleration sensors for acquiring the acceleration in three spatial directions and rotation rate sensors for acquiring the rotational velocity around the three spatial directions.
One refinement of the method, which is also advantageous, moreover provides that a vehicle mass m of the motor vehicle is determined, and a coefficient of friction μ between a roadway and a tire of the rear wheel is determined. In addition, the vehicle mass m and the coefficient of friction μ are input variables of the first state estimator.
It is moreover advantageous if the movement vector of the motor vehicle is determined as an input variable of the second state estimator from a change of a vehicle position ascertained using a GPS and/or from a change of an Earth's magnetic field measured using a magnetometer. The movement vector can alternatively be directly measured using the magnetometer.
One refinement moreover provides that a course angle, which refers to the angle between north direction and the movement vector or the movement direction, is determined from the movement vector or directly from the change of the position which was ascertained using the GPS and/or magnetometer. Alternatively to the movement vector or the movement direction, the course angle can also be used directly as an input variable of the second state estimator.
If the movement vector, the movement direction, and/or the course angle is determined both from the change of the vehicle position ascertained using the GPS and also from the change of the Earth's magnetic field measured using the magnetometer, it is preferably provided that the respective ascertained changes are fused by means of a data fusion or a signal fusion, in particular by means of a Kalman filter, in order to obtain a measured value having higher accuracy and improved availability.
One advantageous variant provides that the slip angle λr is determined from precisely two of the slip angles λr1, λr2, and λr3 by a data fusion. Data fusion in general refers to the linkage of multiple values representing the same variable, wherein the data fusion is usually used to obtain items of information of better quality. A Kalman filter or an expanded Kalman filter can also be used, for example, for the data fusion.
For further error reduction, one refinement provides that a measurement deviation between the slip angle λr determined by the data fusion is determined as a measured value and the slip angle λr1, λr2, or λr3 not used in the determination of the slip angle λr is determined as a reference value. Measures for error correction can then be taken based on the measurement deviation.
It is furthermore advantageous if before the data fusion, a deviation between the slip angles λr1, λr2, and λr3 is determined and the slip angle λr is determined by the data fusion of the two slip angles of the slip angles λr1, λr2, and λr3 having the smallest deviation from one another.
For the determination of the third slip angle λr3, it is moreover advantageous if the vehicle status for determining the third slip angle λr3 is determined by a vehicle velocity v of the motor vehicle, a curve radius R of a curve described by the motor vehicle, and a roll angle Φ.
Moreover, it is provided in one variant that a tire lateral force Fs,f on a tire of the front wheel is determined from the curve radius R and the vehicle velocity v using a third state estimator and the slip angle λr of the front wheel is predetermined for the roll angle Φ and the tire lateral force Fs,f. The tire lateral force Fs,f can be determined, for example, by means of a state estimator from the curve radius and the vehicle velocity. The ratio between the slip angle λr and the roll angle Φ at the tire lateral force Fs,f can be mapped, for example, by a function, a characteristic curve, or a characteristic map and the slip angle λr can thus be determinable. The third state estimator is also implementable as a Kalman filter or expanded Kalman filter.
A further aspect of the invention relates to a method for the traction control of a rear wheel of a single-track motor vehicle by a control loop. A slip angle λr of the rear wheel as a feedback variable of the control loop is determined in this case by the method according to the invention.
Furthermore, one aspect of the invention relates to a system for the traction control of a rear wheel of a single-track motor vehicle. The system comprises a control unit, wherein the control unit is designed to determine a slip angle λr according to the method according to the invention.
The features disclosed above can be combined arbitrarily, if it is technically possible and they are not contradictory to one another.
Other advantageous refinements of the invention are characterized in the dependent claims or are described in greater detail in the following together with the description of the preferred embodiment of the invention on the basis of the figures.
The figures are schematic and by way of example. Identical reference signs in the figures indicate identical functional and/or structural features.
The roll rate {dot over (ϕ)}, yaw rate {dot over (ψ)}, and pitch rate {dot over (θ)}, as well as the accelerations ax, ay, and az in the three spatial directions x, y, and z are determined by the inertial measurement unit 10 (IMU) and provided to the conversion 20. In the conversion 20, the roll angle Φ, which is also referred to as the slip angle, the yaw angle Ψ, and the pitch angle Θ, as well as the curve radius R of the curve described by the motor vehicle are determined from roll rate {dot over (ϕ)}, yaw rate {dot over (ψ)}, and pitch rate {dot over (θ)}, and from the vehicle velocity v.
The three slip angles λr1, λr2, and λr3 are subsequently determined in three different ways, so that a possible error or a deviation of the respective slip angle λr1, λr2, and λr3 from the actual slip angle at the rear wheel may be compensated for.
The first state estimator 30 is a Kalman filter, in which a linear vehicle model is stored. The first slip angle λr1 is determined from a vehicle mass m, a coefficient of friction μ, the variables ascertained by the IMU 10, and the steering angle δ.
Both the vehicle mass m and also the coefficient of friction μ can be “estimated” based on sensor data. For example, the vehicle mass can be composed of individual values added to one another. For this purpose, an empty weight of the vehicle can be known, a fuel weight can be determined by a tank fill level, and, for example, a weight of the persons on the vehicle can be ascertained from a spring behavior acquired by sensors.
The following applies for the determination of the first slip angle λr1:
The functions f1 and f2 are each stored here in the first state estimator 30.
The second slip angle λr2 is determined subsequently or in parallel to the first slip angle λr1. For this purpose, a second state estimator 40 implemented as an expanded Kalman filter is used, which uses the yaw angle Ψ, the course angle ν, and the steering angle δ as input variables, wherein the observation or the use of a nonlinear model of the vehicle is possible due to the expanded Kalman filter.
The course angle ν can be determined from a change of the vehicle position, which can be ascertained by a GPS. Alternatively thereto, it is possible to determine the course angle ν by way of a change of the Earth's magnetic field measured using a magnetometer. To obtain an exact course angle ν, the embodiment shown provides that a course angle ν is used which originates from a data fusion. For this purpose, a first course angle ν1 is ascertained with the aid of the GPS and a second course angle ν2 is ascertained with the aid of the magnetometer and these two course angles ν1, ν2 are offset to form a course angle ν. One very simple option is, for example, to determine the mean value of the course angles ν1, ν2 and use it as the course angle ν. Alternatively thereto, however, a data fusion can also be carried out by means of a further Kalman filter.
Subsequently or in parallel, a third slip angle λr3 is ascertained by the determination 50. For this purpose, the steering angle δ and the roll angle Φ determined by the IMU 10 and the radius R are used as essential input variables. The determination 50 of the third slip angle λr3 is explained in greater detail by way of example with respect to
For the determination of all three slip angles λr1, λr2, and λr3, a database acquired at the same time is used in each case, so that, for example, the steering angle δ is identical in each case.
After the three slip angles λr1, λr2, and λr3 have been determined they are offset with one another by the data fusion 60. The data fusion 60 determines in the embodiment shown the slip angle λr, which corresponds to the actual slip angle at the rear wheel with a higher probability than each of the three slip angles λr1, λr2, and λr3 as such, by means of a further Kalman filter. The slip angle λr thus determined is subsequently provided to the traction control 70.
Alternatively to the method shown in
The models or the state estimators and the calculations and the constants required for this purpose can be stored, for example, in a control unit or in the control unit of the traction control, so that the determination of the slip angle λr can be carried out using the control unit, wherein the further required sensor values or variables are provided to the control unit, for example, by the IMU 10 and a steering angle sensor.
In the determination 51 of the kinematic steering angle Δ, i.e., the theoretical steering angle resulting in the actual cornering, the actual steering angle δ and also the roll angle Φ and the castor angle ε are taken into consideration, which jointly result in the kinematic steering angle Δ.
Moreover, the Ackermann angle ΔA is determined by the calculation 52, which results according to the single-track model under the assumption of small angles for ΔA=p/R.
Furthermore, the following applies for the single-track model at small angles:
To determine the front slip angle λr, with the aid of a third state estimator 53, the tire lateral force Fs,f on the front tire is determined from a traveled curve radius R and the vehicle velocity v. The ratio of the front tire lateral force Fs,f at an inclination described by the roll angle Φ in relation to the front slip angle λr is known and is stored, for example, by a characteristic map or a function, so that the front slip angle λr is determinable therefrom and the rear slip angle λr is determinable as the third slip angle λr3 therefrom.
The invention is not restricted in its embodiment to the above-described preferred exemplary embodiments. Rather, a number of variants is conceivable, which also makes use of the described solution in fundamentally differently designed embodiments.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2019 101 392.5 | Jan 2019 | DE | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2019/083275 | 12/2/2019 | WO | 00 |
