DE4216301C2 - Method for determining the behavior of characterizing variables - Google Patents

Description

The invention relates to a method for determining the driving behavior of characterizing quantities.

There is already known a method according to which for determining the yaw angular velocity of the Sagnac effect is applied. This is monochromatic, coherent light divided and in the opposite direction by means of fiber optic cables guided on a circular path. Due to a rotation (yawing supply) of the optical fiber cable thus results for the split Light a rotating frame of reference. Because in this rotie reference frame corresponding to the electromagnetic waves behave differently in relativistic transformation equations As in a dormant frame of reference, so too do the Interference phenomena of divided light depending on from spin (yaw angle acceleration) and Rotational speed (yaw rate). By Auswer tion of these interference phenomena can thus be attributed to the ent closing down the quantities of the rotary motion.

In this process it might be considered disadvantageous that initially be a light source with interference-capable light be must be provided. In addition, the arrangement of the light be supported in terms of vibrations so that the interference can occur.  

Furthermore, a linear Einspurmodell a vehicle known, where the height of the center of gravity of the vehicle is neglected. Thus, in this approximation, the hard point of the vehicle in the plane of the contact points of the wheels laid. Since rolling and pitching movements are excluded, In this model, the wheels of an axle can become a wheel in the center of the axis are united. This model is included For example, in German book: Zomotor, Adam: chassis technology, Fahrver Hrsg. Jörnsen Reimpell, Würzburg: Vogel 1987, ISBN 3-8023-0774-7 on pages 99 to 116.

This representation is not apparent, as the greed angular velocity and the yaw angular acceleration of measurable Sizes can be derived.

From DE 36 08 420 C2 is a method for determining the Floating angle using the yaw rate as Measured variable known. There is using a vehicle model and the measured quantities of the vehicle longitudinal speed, Steering wheel angle, two in the vehicle longitudinal direction one behind the other horizontal vehicle lateral acceleration and the yaw angle speed a float angle of the vehicle is calculated.

From US 4,690,431 shows, due to vehicle parameters and the input variables steering angle, vehicle speed and two values of lateral acceleration, with the help of a Vehicle model the value of at least one dynamic variable, for example, the yaw acceleration or the yaw angle determine the speed, wherein the at least one dynami variable is measured by a sensor. Due to the Deviation between measured value and determined value of dynamic variables, a control intervention takes place.

In the determination of the driving behavior characterizing Size becomes thereby the forces acting between wheel and road not considered. The achieved by this approach Accuracy of the determined, characterizing the driving behavior  size is sufficient only if in a control loop Constant adjustment to a measured value of the driving behavior th characterizing size takes place. It is not possible, a value for the characterizing the driving behavior Size without feedback to a measured value available too put.

The object of the invention is to provide a method for determining the To create driving behavior of characterizing variables that the Provide one for use in the vehicle affect the control devices sufficiently accurate value of the size made possible light.

This task is used in a procedure for Determining the driving behavior characterizing sizes inventions according to the invention with the features of claim 1 solved, the features of the dependent claims advantageous and further developments.

This object is thus achieved in that the vehicle model considered the transverse force structure. This will be an accuracy achieved, which allows, at least the yaw rate Determine the size characterizing the driving behavior without that a feedback over a measured value of this size necessary is.

First, some equations are to be derived by means of which are the quantities to be determined depending on measurable Sizes are displayed. This is initially a table be created, the meaning of the following used Symbols clarified.

symbol meaning d # / dt first time derivative of a # size, which is one of the sizes included in this table d² # / dt second time derivative of a # size which is one of the sizes included in this table a Vehicle longitudinal acceleration a _q Vehicle lateral acceleration a _qh Vehicle lateral acceleration behind a _qv Vehicle lateral acceleration in front c _h Cornering Stiffness behind c _v Cornering Stiffness in front cs _h Rear spring stiffness cs _v Cross spring stiffness in front c _x Torsional stiffness during roll movement around vehicle longitudinal axis _h Installation height of the lateral acceleration sensor at the rear h _p Roll pole height (distance of the point during a rolling movement from the ground, which is stationary) h _s Gravity height h _v Installation height of the lateral acceleration sensor in front J _x Moment of inertia around the vehicle's longitudinal axis J _z Moment of inertia around the vehicle's vertical axis k _x Rotary damping during roll motion around vehicle longitudinal axis l wheelbase l _h Distance between lateral acceleration sensor rear and front axle l _v Distance between front lateral acceleration sensor and front axle l _s Distance between center of gravity and front axle m vehicle mass S _h Side force on the rear wheels S _v Lateral force on the front wheels   v Vehicle longitudinal speed v _q Vehicle lateral velocity α _h Slip angle on the rear axle a _v Slip angle on the front axle β float angle δ wheel steering angle .delta.t Integration step size Φ yaw t roll angle

The balance of forces in the transverse direction of the vehicle provides the Equation:

m _* a _q = S _{v *} cos (δ) + S _h (1).

The torque balance around the vehicle's vertical axis provides the relationship hung:

J _{z *} d²Φ / dt = _s l S _{* v *} cos (δ) - (ll _{s) *} S _h (2).

In addition, the dynamics of a rolling motion are modeled by the Approach of a differential equation 2nd order:

J _{x *} d 2 τ / dt 2 + k _{x *} dτ / dt + c _{x *} τ = m _* (h _s -h _p ) _* a _q (3).

A modeling of the transverse force build-up on the tires is done after following equations:

These equations describing the lateral force structure have one strong dependence on the vehicle longitudinal speed v on.

The accelerations a _q and a depend on the temporal change of the amount of the respective speed as well as on the temporal change of the direction of the respective speed. The following equations result:

a _q = dv _q / dt + v _* dΦ / dt (8)

a = dv / dt -v _{q *} dΦ / dt (9).

As measured variables, the steering angle δ, the vehicle longitudinal speed v, and two transverse accelerations a _qv and a _qh should now be used. The two lateral acceleration sensors can be described in each case by the distance to the front axle and the jewei time level. Due to the geometry results for these two acceleration sensors:

a _qv = a + _q (l _s -l _{v) *} d²Φ / dt + (h _s -h _{v) *} d²τ / dt (10)

a _qh = a + _q (l _s -l _{h) *} d²Φ / dt + (h _s -h _{h) *} d²τ / dt (11).

Since equations (10) and (11) are thus a system of two equations which are linearly independent in that l _{h is} not equal to l _v , the yaw angle acceleration can thus be achieved given known or negligible acceleration due to acceleration d 2τ / dt 2 d²Φ / dt² and the vehicle lateral acceleration be true. Combining the two equations (10) and (11) yields:

Advantageously, when attaching the Querbeschleu nigungssensoren be provided that h _v h _h h h _p applies. Then the Wankbeschleunigungsterme account regardless of the size order of the roll acceleration, which greatly simplifies the evaluation.

Alternatively, the rolling dynamics by a known per se numerical evaluation of the differential equation ( 3 ) example, by means of the Runge-Kutta method or by means of the expli ziten Eulersfahren with the integration step size δt suc conditions:

τ _new : = τ + dτ / dt _* δt; dτ / dt _new : = dτ / dt + d²τ / dt²t _* Δt (15)

The yaw angle and lateral acceleration as well as the roll sizes Thus, using Equations (12), (13), (14) and (15) or by state estimates by means of equations (3), (10) and (11).

From the force balance according to equation (1) and the moment balance according to equation (2), the lateral forces can be obtained:

By differentiation of the two equations (16) and (17) results thus:

From the equations (4), (5), (6) and (7), two linear equations thus result, by means of which the quantities v _q and dΦ / dt can be determined from known quantities.

The equations (16), (17), (18) and (19) contain only variables which are measured directly (δ, v, a _qh, a _qv) or may be determined as described above on the basis of measured quantities. Time derivatives of known sizes can be derived by quotient education. Thus, the quantities v _q and dΦ / dt can be calculated at any time. The slip angle β is finally obtained:

β = arctane (v _q / v) (22).

So far it has been described, as from known parameters, that the Vehicle concern, the driving behavior characterizing sizes be determined. These parameters are subject to certain Fluctuations. The vehicle mass and the center of gravity vari It is characterized by different loadings. The tire dep gigen sizes vary with the tire temperature and with different road surfaces. The most significant fluctuation occurs at Cornering Stiffness. The following is a procedure presented with an adaptation of Cornering Stiffness is possible. Likewise, then an adaptation of the other parameters respectively. First of all, system equations have to be used for this adaptation are found, the only known or derivable sizes enthal and the linear are independent of the equations that are at used to determine the known or derived quantities  were. Suitable equations are obtained for example by the Differentiation of equations (20) and (21).

The yaw rate is given by equations (20) and (21) by eliminating v _q . The lateral forces are calculated by means of equations (16), (17), (18) and (19) as well as corresponding difference quotients. This results in current estimates for Cornering Stiffness in front and behind. An update of the values used in the other calculations Cornering Stiffness can be done by a L₂ approximation with sliding time averaging. The parameters to be used in the following calculations are denoted c _{v, akt} and c _{h, akt} . The previous parameters are denoted by c _v and c _h . The current estimates are denoted c _v (t) and c _h (t).

c _{v, act} = (l-Γ) _* c _v + Γ _* c _v (t) (27)

c _{h, act} = (l-Γ) _* c _h + Γ _* c _h (t) (28).

Here, Γ is a factor with a value between 0 and 1. The larger Γ, the more the current estimates are taken into account. With a small value of Γ a slow adaptation takes place. If the adaptation is to take place, for example, with a limiting frequency Ω _G , then Γ = Ω _{G *} δt.

An embodiment of the invention is shown cal diagrammatically in the drawing and will be described in more detail below. Figures 1 and 2 are to be taken from representations showing the sizes of the above table on a vehicle. Fig. 3 shows a possible course of the proceedings according to the invention. In a first step 301, the values are determined on the basis of which the parameters characterizing the driving behavior are determined. These values are the steering angle δ, the vehicle _{longitudinal speed} v and the transverse acceleration a _qv and a _qh . The steering angle is preferably measured directly, the vehicle longitudinal speed, for example, from the signals of speed sensors be true. The two transverse accelerations are preferably measured directly by means of suitably mounted transverse acceleration sensors. In a second step 302, the transverse acceleration a _q and the yaw angular acceleration d 2Φ / dt 2 are determined from these quantities, for example by means of equations (12) and (13). For example, by means of equations (14) and (15), the roll angle τ is then determined with its time derivatives in a step 303. Difference quotients then form the higher derivatives of the state variables, which are still needed in the following. By means of the equations (16), (17), (18) and (19), the lateral forces and their derivatives are then obtained in a step 304. By means of the equations (20), (21) and (22), the slip angle and the yaw angular velocity are then obtained in a step 305.

In addition, it can be seen from FIG. 4 that an adaptation of the parameters can take place. For this purpose, higher derivatives of the lateral forces are first determined in a step 401 by means of difference quotients. Equations (23), (24), (25) and (26) are then used to determine current estimated values of the parameters in a step 402. By means of equations (27) and (28), in a step 403, values of the parameters are then determined, which are then used in subsequent calculations to determine the driving behavior of characterizing quantities.

Fig. 5 shows a computing device 501 , which are fed as Eingangsig signals 502 , 503 , 504 , 505, the said variables corresponding to the step 301 of FIG . After the corre sponding variables in the computing means WUR by means of said equations according to the flowchart of Fig. 3 WUR, output signals 506 , 507 , 508 are output, which represent the particular sizes. These specific variables are the yaw rate and, if appropriate, the slip angle and / or a further characteristic characterizing the driving behavior, which was determined during the course of the method. Furthermore, according to the sequence diagram of FIG. 4, an adaptation of the parameters by means of which the vehicle model is described.

Claims (8)

1. A method for determining the characteristics characterizing the driving behavior, in which the steering angle (δ), the vehicle _{longitudinal speed} (v) and two in the vehicle _{longitudinal direction one} behind the other transverse accelerations (a _qv , a _qh ) measured and these signals representing signals ( 502 , 503 , 504 , 505 ) are supplied to a computing device ( 501 ) in which exclusively using all four measured variables, vehicle-specific variables and the lateral force structure on the tire considering vehicle model further, the driving behavior characterizing variables in several calculation steps (302, 303, 304) are derived and at least thereof the yaw angular velocity (dΦ / dt) is detected and output. 2. The method according to claim 1, characterized, that the determination of the yaw angular velocity (dΦ / dt) regardless of a particular driving condition. 3. The method according to claim 1, characterized, that further determines the slip angle (β) and is issued. 4. The method according to claim 1, characterized,  that in the vehicle model Wankbewegungen the vehicle be considered (303). 5. The method according to claim 1, characterized in that the lateral accelerations (a _qv , a _qh ) are measured by means of lateral acceleration sensors whose respective installation heights (h _v , h _h ) are equal and correspond to the center of gravity height (h _s ). 6. The method according to any one of claims 1 to 5, characterized, that an adaptation of the vehicle-specific variables takes place (401, 402, 403). 7. The method according to claim 6, characterized, that the adaptation by means of a sliding Mean value method (403). 8. The method according to any one of claims 1 to 5, characterized in that at least one further derived, the driving behavior characterizing size ( 506 , 507 , 508 ) is output.