DE102009050209B4 - Method and device for determining the standstill position of a motor vehicle - Google Patents

Abstract

Method for determining the position of the actual standstill point of a motor vehicle during a braking process of the motor vehicle, comprising the following steps: - determining the actual longitudinal acceleration of the motor vehicle, - determining a mean longitudinal acceleration taking into account the natural frequency of the oscillation of the motor vehicle at the end of the braking process, - determining a Longitudinal acceleration deviation by forming the difference between the actual longitudinal acceleration and the mean longitudinal acceleration, and determining the actual standstill point of the motor vehicle from the longitudinal acceleration deviation.

Description

The invention relates to a method for determining the standstill position of a motor vehicle according to claim 1 and a corresponding device according to the preamble of claim 7. Furthermore, the invention relates to the use of the device in a driver assistance system and in a motor vehicle.

Odometry is the science of determining the position of a mobile system using data from its propulsion system. In particular, wheel-driven systems use the number of wheel revolutions for this purpose, whereby the distance covered between two measuring points can be determined from the number of wheel revolutions together with the known wheel circumference. If the position of the starting point, for example the first measuring point, is known, the result is the position of the end point, i.e. the second measuring point, in conjunction with the steering angle and the distance covered.

However, like all dead reckoning methods, odometric methods are highly error-prone, with the position error increasing with longer-lasting measurement or the successive sequence of several measurements without intermediate support points.

• - Radgeometrie (Unrundheit, Verschleiß, fehlerhafte Messung des Raddurchmessers),
• - Bodenbeschaffenheit (Schlupf, Unebenheit der Straßenoberfläche),
• - Fahrgestellgeometrie (Spiel, fehlerhafte Messung der Radabstände und Lenkeinrichtung), und
• - Fahrzeuggewicht (ungleiche Gewichtsverteilung).

Variables that have an influence on the accuracy of the odometric position measurement are, for example:

• - wheel geometry (out-of-roundness, wear, incorrect measurement of the wheel diameter),
• - soil conditions (slippage, unevenness of the road surface),
• - Chassis geometry (play, incorrect measurement of the wheel spacing and steering device), and
• - Vehicle weight (uneven weight distribution).

To minimize errors in the odometric position determination, the document discloses DE 693 11 635 T2 a device with a first sensor for determining the rotational speed of a wheel and a second sensor for determining an instantaneous vertical speed of the wheel relative to the body, a processing device processing the measured variables of the two sensors. A correction filter is provided in the processing device, which corrects the displacement of the body due to the rotational speed of the wheel by means of the measured vertical speed.

Furthermore, from the EP 0 517 082 A2 a device for detecting the number of wheel revolutions is known, the number of revolutions of a part which is coupled to the drive and which rotates when the vehicle is in motion being detected by a sensor.

When braking a vehicle to a standstill, the longitudinal acceleration becomes negative after driving forward, in other words the vehicle is decelerated. At the end of this deceleration, the vehicle comes to a standstill and the deceleration should suddenly go to zero, which is not the case in reality the weight ratios in the vehicle can level off at a longitudinal acceleration other than zero. During this leveling out, for example, the distance pulse sensor of a rear wheel can still send distance pulses for the rear wheel, although the vehicle is obviously already at a standstill. The reason for the occurrence of such travel impulses despite the vehicle being at a standstill can, for example, be that the braking force is distributed during braking by means of the brake pedal, in which the front axle is subjected to more braking force than the rear axle. If the rear axle still has a small amount of play due to the suspension and suspension, further travel impulses can be generated despite the vehicle being at a standstill. It is now obvious that taking these path impulses into account in odometry would lead to incorrect position determination, which, for example, would lead to navigation errors and errors in the alignment of the vehicle in the parking space during an automatic parking process.

The invention is therefore based on the object of reducing errors in the odometric position determination, in particular when determining the standstill position after braking a vehicle.

This object is achieved by the method with the features of claims 1 and 5 and by the devices with the features of claims 7, 9 and 10. Preferred embodiments of the invention are the subject of the subclaims.

• - Bestimmen der tatsächlichen Längsbeschleunigung des Kraftfahrzeugs,
• - Bestimmung einer mittleren Längsbeschleunigung unter Berücksichtigung der Eigenfrequenz des Einpendelns des Kraftfahrzeugs am Ende des Bremsvorgangs,
• - Bestimmung einer Längsbeschleunigungsabweichung durch Differenzbildung der tatsächlichen Längsbeschleunigung und der mittleren Längsbeschleunigung, und
• - Bestimmung des tatsächlichen Stillstandspunkts des Kraftfahrzeugs aus der Längsbeschleunigungsabweichung.

The method according to the invention for determining the position of the actual standstill point of a motor vehicle with respect to a braking process of the motor vehicle has the following steps:

• - determining the actual longitudinal acceleration of the motor vehicle,
• - Determination of a mean longitudinal acceleration taking into account the natural frequency the leveling off of the vehicle at the end of the braking process,
• - Determination of a longitudinal acceleration deviation by calculating the difference between the actual longitudinal acceleration and the mean longitudinal acceleration, and
• - Determination of the actual standstill point of the motor vehicle from the longitudinal acceleration deviation.

Due to the at least partially elastic connection between the chassis and the body of the vehicle, in other words the suspension of the wheels, the steering and the axles in relation to the body, etc., the connecting elements of the corresponding elements being partially elastic and / or spring elastic, this oscillates At the end of the braking process, the vehicle comes to a standstill with a typical natural frequency around its zero position, until it comes to rest due to the system-related damping. This natural frequency of the vehicle oscillating around its zero position essentially depends on the vehicle type and can therefore be determined for each vehicle type in the sense of a calibration measurement. By considering the deviation of the longitudinal acceleration from the mean longitudinal acceleration as a function of the natural frequency, the actual standstill point of the vehicle can be recognized in an advantageous manner.

The mean longitudinal acceleration and the longitudinal acceleration deviation of the motor vehicle when braking to a standstill are advantageously determined by low-pass filters. The actual standstill point is detected by comparing the longitudinal acceleration deviation with a predetermined threshold value. The specified threshold value can be specified as a range, the threshold value and its range also being able to be determined as a function of the vehicle type through tests.

The above-described method for detecting the standstill point of a motor vehicle can advantageously be used to determine faulty travel impulses of at least one travel impulse sensor of a motor vehicle. The distance pulses are used to determine the position of the motor vehicle by means of odometry, but those distance pulses from the position pulse sensor that occur later than the actual standstill point of the motor vehicle are not taken into account for position determination by means of odometry or are compensated for afterwards.

To determine the “excess” travel impulses, an alleged standstill point of the motor vehicle is preferably determined, the alleged standstill point being defined in that no travel impulses are generated by the travel impulse sensor for a predetermined time. A time interval of 0.5 seconds, for example, can be used as the predetermined time. Distance impulses between the actual standstill point and the supposed standstill point are not taken into account in odometry.

A device according to the invention for a motor vehicle, in which "excess" distance pulses occurring after the actual standstill point are not taken into account, comprises at least one distance pulse sensor arranged on a wheel of the motor vehicle, an odometry device for determining the position of the motor vehicle by means of the distance pulses of the at least one distance pulse sensor and a Low-pass filter for determining the mean longitudinal acceleration as well as the longitudinal acceleration deviation of the motor vehicle when braking to a standstill. The filter function of the low-pass filter is a function of the natural frequency when the vehicle levels off.

The device preferably has a comparison device for comparing the longitudinal acceleration deviation with a predetermined threshold value, the actual standstill point of the motor vehicle being determined from the comparison.

The device according to the invention is used in particular in a driver assistance device for assisted or automatic parking of a motor vehicle.

A motor vehicle preferably has a driver assistance device explained above.

• 1 ein Beispiel eines zum Stillstand führenden Bremsvorgangs zeigt.

A preferred embodiment of the invention is explained below with reference to the single drawing, wherein

• 1 shows an example of a braking operation leading to a standstill.

The solution to the problem according to the invention is essentially about recognizing the actual standstill point of the vehicle and recognizing all travel impulses that are generated after this point in time and excluding them from consideration in order to achieve an improved odometry in this way.

For example, if a distance pulse corresponds to a distance of 2 to 3 cm and if the number of distance pulses for the left and right rear wheel is different, the odometry would calculate an inclined position of the vehicle in the parking space during an automatic parking process or the vehicle would not park in the parking space parallel to the parking space axis.

The natural frequency of the vehicle's oscillation around its zero position is used to determine the actual standstill point of a vehicle. This oscillation frequency is a characteristic variable of a vehicle and can be determined by calibration measurement. In particular, the oscillation frequency is a characteristic variable of a vehicle type, so that it does not have to be determined individually for each vehicle, but rather it is sufficient to determine the oscillation frequency once for a vehicle type and to use the value determined in this way for this type. In terms of magnitude, the oscillation frequency for a mid-range vehicle is around 3 Hz.

In 1 in the upper part is the longitudinal acceleration in m / s ² versus time t displayed in seconds, the display starting at a point in time of 7.5 seconds. The curve A. shows a braking process until the vehicle comes to a standstill, ie the course of the longitudinal acceleration a _{x over time is} shown. As can be seen, the longitudinal acceleration oscillates around the zero value with a typical natural frequency.

In the lower part of the 1 are the measured distance impulses Wl, for example a rear wheel, as a function of time t as a graph WZ shown. It can be seen that between the times 7.5s and 7.75s, three travel impulses are generated, which increases the number of travel impulses from 461 to 464. After overshooting the longitudinal acceleration curve A. Two more travel impulses are generated over the zero line in the time interval 7.75s to 8s, although the vehicle is obviously stationary.

eine Art „Mittelwert“ 〈a_x〉 der Längsbeschleunigung a_x unter Verwendung der Einpendelfrequenz ω₀ gebildet, der der Längsbeschleunigung gut folgt und in 1 oben durch den Graphen B dargestellt ist. Im Prinzip ist (1) eine Laplace-Darstellung des Tiefpasses.To determine the actual standstill point, the shape $$〈a_x〉=\frac{1}{1+s\text{/}{\omega }_0}{a}_x$$

a kind of "average"<a_x> of the longitudinal acceleration a _x using the Einpendelfrequenz ω ₀ formed that follows the longitudinal acceleration and good in 1 above through the graph B. is shown. In principle (1) is a Laplace representation of the low pass.

und ist in 1 durch die Kurve C dargestellt. Dabei kann vorteilhafterweise der gleiche Tiefpass wie in (1) verwendet werden.From this central longitudinal acceleration and longitudinal acceleration, the longitudinal acceleration deviation <a _x is - <a _x >> constituted as follows: $$\begin{array}{l}〈a_x-〈a_x〉〉=\left(1-\frac{1}{1+s\text{/}{\omega }_0}\right)\frac{1}{1+s\text{/}{\omega }_0}{a}_x\\ =\frac{s\text{/}{\mathrm{<figure-callout\; id="0"\; label="\omega "\; filenames="DE102009050209B4\_0003.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_0}{1+2s\text{/}{\mathrm{<figure-callout\; id="0"\; label="\omega "\; filenames="DE102009050209B4\_0003.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_0+{\left(s\text{/}{\omega }_0\right)}^2}{a}_x\\ =\frac{s{\omega }_0}{{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="DE102009050209B4\_0003.png"\; state="\{\{state\}\}">s</figure-callout>}}^2+2s{\mathrm{<figure-callout\; id="0"\; label="\omega "\; filenames="DE102009050209B4\_0003.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_0+{\mathrm{<figure-callout\; id="0"\; label="\omega "\; filenames="DE102009050209B4\_0003.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_0^2}{a}_x\end{array}$$

and is in 1 through the curve C. shown. The same low-pass filter as in (1) can advantageously be used here.

The above type of signal processing is only to be understood as an example and the low-pass filter in (1) and (2) can differ from the form given above.

In the middle of the 1 is the logic L. the signal processing in the form of various logic levels and a correction value as a function of time t shown.

This is the logic level D. , also called bit D. called, set to "True" if the longitudinal acceleration deviation exceeds <a _x -<a _x >> a predetermined value. In the present example, the threshold is 0.2 m / s ² .

Will this bit D. is set to "True" and the vehicle is in motion for longer than a specified time, the bit E. set to the value "True", which is in the graph E. is shown. When the level of this bit rises E. the current values of the displacement pulses at this point in time are temporarily stored, which in this example is 464, as shown in the graph WZ can be seen.

The vehicle sets another bit F. when it thinks it has stalled. The bit F. is set, for example, when no more distance pulses are generated for a specified period of time, the time interval in this example being approximately 0.5 seconds.

By the bit F. becomes the logic level E. , ie the bit E. , set back to the value "False" and a trigger pulse G which causes the difference between the current value and the temporarily stored value of the displacement pulses to be formed. In the present case, the difference has the value 2, which is represented by the level of the correction value of "-2" shown in H. These are the "faulty" path impulses that have to be taken into account to correct the odometry.

List of reference symbols

LB

Longitudinal acceleration [m / s ² ]

L.

logic

WI

Travel impulses

t

time in seconds

A.

Longitudinal acceleration a _x

B.

central longitudinal acceleration <a _x>

C.

Longitudinal acceleration deviation <a _{x x} -<a >>

D.

Longitudinal acceleration deviation> threshold

E.

Threshold exceeded AND vehicle has not yet recognized a standstill

F.

no travel impulses for a specified time

G

Trigger

H

Distance pulse correction

WZ

Distance pulse number of the distance pulse sensor

Claims (10)

Method for determining the position of the actual standstill point of a motor vehicle during a braking process of the motor vehicle, comprising the following steps: - determining the actual longitudinal acceleration of the motor vehicle, - Determination of a mean longitudinal acceleration taking into account the natural frequency of the vehicle's leveling-out at the end of the braking process, - Determination of a longitudinal acceleration deviation by calculating the difference between the actual longitudinal acceleration and the mean longitudinal acceleration, and - Determination of the actual standstill point of the motor vehicle from the longitudinal acceleration deviation. Procedure according to Claim 1 , characterized in that the mean longitudinal acceleration and the longitudinal acceleration deviation are determined by low-pass filters. Procedure according to Claim 2 , wherein a filter function of the low-pass filter is a function of the natural frequency. Method according to one of the preceding claims, characterized in that the actual standstill point takes place by comparing the longitudinal acceleration deviation with a predetermined threshold value. Method for determining faulty travel impulses of at least one travel impulse sensor of a motor vehicle using the method according to one of the preceding claims, the travel impulses being used to determine the position of the vehicle by means of odometry, characterized in that travel impulses of the travel impulse sensor which occur later than the actual standstill point of the vehicle , for position determination by means of odometry are not taken into account. Procedure according to Claim 5 , characterized in that a supposed standstill point of the motor vehicle is determined if no distance pulses are generated by the distance pulse sensor for a specified time and distance pulses between the actual standstill point and the supposed standstill point are not taken into account in the odometry. Device for performing the method according to one of the Claims 5 or 6th for a motor vehicle, with at least one displacement pulse sensor arranged on a wheel of the motor vehicle and with an odometry device for determining the position of the motor vehicle by means of the displacement pulses of the at least one displacement pulse sensor, characterized in that the device has a low-pass filter for determining the mean longitudinal acceleration and for determining the longitudinal acceleration deviation of the motor vehicle when braking to a standstill. Device according to Claim 7 , characterized in that the device has a comparison device for comparing the longitudinal acceleration deviation with a predetermined threshold value, the actual standstill point of the motor vehicle being determined from the comparison. Driver assistance device for assisted or automatic parking of a motor vehicle with a device according to one of the Claims 7 or 8th . Motor vehicle with a driver assistance device according to Claim 9 .

Images