DE102011080761A1 - Method for carrying out velocity warning for motor vehicle, involves adding maximum turning velocity and expected steering error in form of likely faulty curvature to track curve when calculation of maximum rate of turn is influenced - Google Patents

Abstract

The method involves computing predetermined road ahead curvature of the track by a control unit (5) on the basis of map data (2). A maximum cornering velocity is calculated as a recommendation output. The maximum turning velocity and expected steering error of the driver in the form of a likely faulty curvature are calculation and added to the track curve when calculation of the maximum rate of turn by the driver is influenced. An independent claim is included for device for carrying out velocity warning for motor vehicle.

Description

State of the art

The invention relates to a method and a device for carrying out a speed warning for a motor vehicle, in which a control unit calculates a curvature of the route on the basis of map data for a predeterminable leading section and from this a maximum speed is calculated and output as a recommendation, wherein the calculation the maximum speed can be influenced by the driver.

A generic method is for example from the DE 10 2006 028 277 A1 known.

Velocity alerters, in particular, curve detectors are well known in the art. The curve detectors calculate in advance, with which maximum speed a vehicle should pass the upcoming curve and output the result as a recommendation to the driver, in particular by a visual display. The actually driven in curves or mobile speed apparently depends on the physically possible speed, which is determined by the road geometry, such as curvature and road width, but also by environmental factors, such as road conditions, visibility of the road, as well as the driving behavior of the driver. The calculation of this curve limit speed is often based, compare for example DE 42 05 979 A1 , to the physically defined limit speed, which is calculated from the lateral acceleration a and the curve radius R. V = √ a · R , where a results inter alia from the coefficient of friction of the tires. Scientific investigations, such as KH Schimmelpfennig, N. Hebing, velocities in circular cornering Stability and safety limit, in: the traffic accident, May 1982, Issue 5, pages 97-99 , prove that in reality the curve is not traversed with this physically possible limit speed. Obviously, environment factors and personal driving style play into the actual driving behavior in a way that is difficult to explain. In the mentioned work many trips were examined and analyzed. It will result in a formula of the type V = c + b · lnR where R is the radius of curvature, c and b are each a regression constant.

However, this known model is of limited explanatory power. It does not explain why on motorways much higher cornering speeds are achieved than allowed by this formula. There is no explanation why the speed should increase only with the logarithm. A lack of the model described by the formula becomes particularly visible in the fact that for even distances the speed would grow by the logarithm to infinity. The known models can describe so far no capping the maximum cornering speed. The regression constants always apply only to a specific situation, that is to say depending on the respective lighting, road width, weather and many other parameters, and thus have to be learned for every situation. Since no justification is given for these parameters, a prediction of the parameters for unknown situations is not possible. A further disadvantage is that this formula has a very wide range of uncertainty. Finally, it is not clear which factors change the regression parameters. In summary it follows that with these very vague statements the programming of a curve alarm is only slightly supported.

Disclosure of the invention

Against this background, a method according to the invention and a device according to the independent patent claims are presented. Further embodiments of the invention will become apparent from the dependent claims and the description.

In the method according to the invention, it is provided that in the calculation of a first maximum speed the driver's expected steering errors are taken into account in the form of a probable wrong curvature, which is added to the line curvature.

The invention is based on the idea of a plausible, explanatory and easily expandable model of the cornering behavior, taking account of further aspects, which contains only a few free parameters and is suitable for easy implementation in a vehicle curve detector. The model assumes that the vehicle will not drive on the ideal line, but will fluctuate around this ideal line with the additional faulty curvatures. This allows learning and predicting the maximum speed or cornering speed even for unknown situations. The cornering way speed alert does not have to learn whole situations again and again. The method is also applicable to infinitely large curve radii, ie for straight line failure. By considering the Even in this case, real occurring Fehlkurümmungen results here a capping of the maximum speed, so that the corner warning even on a straight line provides a sensible speed recommendation.

According to a first development of the invention, the probable erroneous curvature is derived from the actual steering errors of the driver, the steering errors being determined by means of a steering angle sensor. The probable erroneous curvature can therefore be determined in a less complex manner or learned for future situations.

The training mentioned can be advantageously supplemented by the long-term planning of the driver, as it results from the visible curve area, is included in the curve warning. For this purpose, a development of the invention is proposed, according to which a second maximum speed is determined from the quotient of the visible curve area and a driver-typical time for the preparation of future steering maneuvers, and then the minimum of the first and second maximum speed is determined and output.

In a further development considered to be particularly advantageous, which complements the previously mentioned developments by also taking into account short-term steering corrections and which is also applicable to straight sections, a third maximum speed is calculated from the quotient of a roadway area still available for steering correction movements and a determined for the steering correction movements required driver-typical response time, and then the minimum of the determined maximum speeds is determined and output.

The invention will be described in more detail below with reference to exemplary embodiments. Show

1 a sketch to illustrate an embodiment for the geometric derivation of the visible curve area,

2 a sketch to illustrate an embodiment for the geometric derivation of the conditions to keep the vehicle despite steering correction movements within the lane,

3 1 is a schematic representation of a block diagram to illustrate an embodiment of the method according to the invention,

4 3 is a diagram illustrating a comparison between the maximum curve speed calculated according to the prior art and a calculated according to the invention,

5 a schematic representation of a device according to the invention. The curve radius is used for the following geometric derivations. The curvature χ is defined as the inverse of the radius of curvature R, that is: R = 1 / χ.

Looking at bends is easier for steering maneuvers, as curvature values can be added directly (turning on the steering wheel causes additional curvature in the trajectory of the vehicle). This can be described by simply adding the curvatures χ.

In the above-cited known approaches to calculate the maximum curve speed is always expected with a curvature resulting from the map data (possibly even with a curve cutting). In contrast, the erroneous curvature is still observed according to the invention, which results from the einkalkulierten steering maneuvers. Because the current speed of the vehicle must still be below the physically permissible maximum speed, if the curve radius from the map material nor the likely Fehlungskümmung is added, which corresponds to the expected steering errors of the driver. The expected faulty curvature can be determined, for example, with the aid of a steering angle sensor and the average value obtained can then be learned. Specifically, a number of measured values of the steering angle sensor can be recorded, from which subsequently - if appropriate after an evaluation of the individual steering maneuvers, whether these are to be considered relevant on the basis of specifiable criteria or not - an average value is formed, from which finally the expected misalignment is determined.

The maximum speed from the lateral stiffness is first determined, as known in the art, over the maximum transverse force a. This results in this 'maximum curve speed from physics' according to

This formula, which is known per se, is now modified by the addition of the wrong curvature from the expected steering errors to the actual (ie to be traveled in reality) section curvature, that is

According to this model according to the invention, higher speeds than an inexperienced one can be recommended to an experienced driver since the erroneous curvature is less in the case of an experienced driver. The mere addition of this wrong curvature in the formula known per se already ensures that the maximum cornering speed does not grow to infinity.

Further inventive considerations for the realistic modeling of the driving behavior in curves are based on the knowledge that the driver can only use the visible curve range for the planning of the future steering maneuvers. This visible area S _visible , compare 1 , can be measured by means of an environment detection system. This is particularly well suited to a video system. Alternatively, the visible range can also be estimated from the map data. With a curve radius of R and a roadway width of B results, cf. 1 , By means of simple geometric considerations for the illustrated case of a circular arc curve a visible area according to

For the planning or preparation of the future driving maneuvers following the visible area, the driver needs a typical time T. The present curve hinders, the more the smaller S is _visible , the driver's view of the future, so to speak. This results in the, maximum curve speed due to the long-term planning 'to v = S _visible / T. If the driver knows the distance traveled, he needs a much shorter time T 'to prepare the steering maneuvers. Thus, driving on a known route with a higher cornering speed. This approach also explains why it is slower to drive with very confusing curves than with clear terrain.

The visible distance S _visible can therefore be calculated by means of an environment detection system such as video, radar or from the map data. Likewise, the preparation time T can be learned for unknown and for known terrain, since this is a driver-typical measure. This allows a prediction of this maximum cornering speed for new routes.

Incidentally, the derivation of the visibility took place in accordance with 1 in the case of a circular arc curve. Similarly, the visibility can of course be deduced in the case of a clothoid curve or any other shape.

The exact waveform is in close range e.g. recorded by video, at medium distances by means of radar (by, for example, tracking vehicles ahead) and for the far-away area can be accessed on the stored digital map.

Visibility is also influenced by the current weather conditions. To reduce fog, darkness, rain, etc. the visibility. These environmental conditions can be determined according to very different methods and then be included in the determination of the maximum curve speed in question here. For example, the darkness - for the control unit - can be recognized, for example, by the lights switched on or simply by the time of day. Rain is detected, for example, by means of a rain sensor or an active windscreen wiper. Among other things, fog can be recognized by an active rear fog light or traffic reports.

In addition, the visibility is influenced, among other things, by the development or planting of the road. This can e.g. be detected by analysis of camera images, or removed from the map, if the card has corresponding data. Typical visibility is also derived from the currently used road class. On motorways, for example, there is a wider view than on country roads. The respective road class can be read off the digital map or also be derived by video on the basis of typical characteristics for the respective road class, for example by means of sign recognition. A recognized traffic sign, for example, "attention strong curve", is of course also a strong indication of a limited visibility.

After the driver has considered the long-term, future steering maneuvers, they only have to be implemented. To do this, the driver always has to readjust the steering wheel. The goal is to keep the vehicle within the allowed lane. The width of the available lane may e.g. be measured by means of video sensors.

den Fahrstreifen verlassen, wie aus 2 geometrisch ableitbar ist. Innerhalb der Reaktionszeit des Fahrers muss nun eine Korrektur erfolgen, wobei dem Fahrer genau dieser Fahrbahnbereich S_verlassen zur Verfügung steht, um noch zu reagieren. Somit ergibt sich eine ’maximale Kurvengeschwindigkeit aus der Kurzfristkorrektur’ zu v = S_verlassen/T. In a further exemplary embodiment of the invention, for the purpose of modeling this partial aspect of the cornering behavior, it is assumed that the driver makes a typical steering error or calculates this steering error in his driving behavior. This manifests itself in the way that a still curvature must be added to the actually required track curvature. Because of this wrong curvature that would Vehicle leave the lane. The wrong curvature corresponds to a false radius R _Fehl . For a remaining lane width B _whereabouts (this is the width of the lane minus the width of the car), the vehicle would after a remaining remaining length (lane area)

leave the lane as if out 2 is geometrically derivable. Within the reaction time of the driver is now a correction must be made, with the driver exactly this road section S _leave is available to respond yet. This results in a 'maximum curve speed from the short-term correction' to v = S _leave / T.

This formula shows, and empirically correct: the narrower the passable road curve, the slower the vehicle drives. Furthermore, with increasing experience of the driver results in a smaller Fehlkurümmung and thus a higher cornering speed is possible.

However, environmental conditions also have a significant influence on steering movements in this context, that is to say on the erroneous curvatures to be taken into account during the journey. Strong gusty crosswinds cause significant steering corrections to keep the vehicle on track. These enhanced, short-term steering maneuvers are learned and then automatically the recommended cornering speed is reduced accordingly. Similarly, a bad road with many potholes is driven at a lower cornering speed, since there the required readjustment is very large, so R _Fehl a rather small value, which leads to a low cornering speed.

The road width can be measured by environment detection (e.g., video). Similarly, the road condition can be estimated (potholes).

The steering error, so the steering safety of the driver can be determined by a steering angle sensor. This sensor measures the steering movements and thus gives a statement about the steering errors and steering safety of the driver. Thus, the values required in the above formula are directly measurable, or the steering error as a typical size for the driver also learnable.

The formula for the short-term correction was above, cf. 2 , Derived for the sake of simplicity from the width of the lane. In reality, not the entire lane is used. Rather, the driver always holds the vehicle within a pendulum area that is typical for the respective driver. This is the area the driver stops even if he drives over a wide area, for example. Of course, the width of this pendulum area, within which the vehicle moves like in a hose, depends on the respective environmental conditions, the driving safety of the driver and the respective speed. The width of the pendulum area is now measured and learned eg by means of video sensors.

Once the width of the pendulum area has been learned, it is used instead of the street width (or the minimum of the street width and the pendulum area). This makes it clear that the maximum cornering speed does not grow to infinity just because the vehicle is moving on a very wide road. It is then always the restriction from the pendulum area.

The block diagram of 3 shows a block 11 in which the determination of a first maximum speed or cornering speed according to the first embodiment above, ie from the line curvature plus steering error takes place. In doing so, according to block 14 Data of a digital map for determining the path curvature and data of a steering angle sensor according to block 15 used. A second maximum cornering speed is in block 10 determined on the basis of the driver's sight and preparation time T, wherein according to block 13 , as already described in more detail above, environment data and the learned preparation time for steering maneuvers T are used. Furthermore, in block 12 determines a third maximum cornering speed, which, as also explained above, based on the fact that the driver despite steering correction movements keeps the vehicle within the lane or its individual pendulum area. In turn, environment data and data of a steering angle sensor according to block 15 used. In the following step 16 the minimum of the three mentioned cornering speeds is determined, which is then to the corner warning to the driver according to block 17 is being used.

In the diagram of 4 is a comparison of the maximum curve speed as a function of the radius once after the known from the prior art logarithmic relationship and even after the just described embodiment of the invention with minimum formation of the three determined maximum cornering speeds, shown. It can be seen that, as a result, the method according to the invention is similar to the known logarithm. However, the method according to the invention provides an explanation of why the curve speed behaves and the Curve speed is plausible even for very large radii. It also explains why older people (longer reaction times) and novice drivers (larger control curve bends) drive slower than younger and more experienced vehicle drivers. By using the vehicle's surroundings detection systems, the recommended cornering speed is optimally adapted to the respective road and lighting conditions. Furthermore, the method according to the invention takes into account the fact that, when visibility is clear, the traffic is slower on narrow roads and why there is a capping effect on the cornering speed.

The 5 shows an advantageous embodiment of a device according to the invention 1 , The device 1 has a memory 3 in which a digital map 2 is stored. Also, the memory can 3 only with the device 1 be connected so that the device the digital card 2 use another device, such as can read. In this case, the device itself or the other device may be a navigation system. The device 1 is by means of a position determining means 4 For example, by means of GPS, able to determine the current position and lying ahead of the device driving distance of a vehicle. Furthermore, the device 1 a control unit 5 on, which by means of the position data of the position determining means 4 and the map data of the digital map 2 can determine the road ahead. In addition, the device has 1 via a display and / or warning means 7 by which the device can inform and / or warn the operator of the device or vehicle about a certain maximum cornering speed. The control unit 5 can further with data of a steering angle sensor 8th as well as data from environment sensors 9 be supplied. Learned input values, in particular a driver-typical time for the preparation of future steering maneuvers, for example, in the input means 6 be stored.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 202006028277 A1 [0002]
• DE 4205979 A1 [0003]

Cited non-patent literature

• KH Schimmelpfennig, N. Hebing, velocities in circular cornering Stability and safety limit, in: the traffic accident, May 1982, Issue 5, pages 97-99 [0003]

Claims (12)

Method for carrying out a speed warning for a motor vehicle, in which a control unit ( 5 ) based on map data ( 2 ) calculates a curvature (χ _curve ) of the route for a predeterminable preceding section and from which a maximum speed is calculated and output as recommendation, wherein the calculation of the maximum speed can be influenced by the driver, characterized in that when calculating a first maximum speed (V _max) is the expected steering error of the driver in the form of a probable incorrect curvature of the (χ _curve) is added in addition to the track curvature (χ _defects) are considered. A method according to claim 1, characterized in that the likely failure of curvature (χ _failure) is derived from the driving behavior of the driver. A method according to claim 2, characterized in that the likely failure of curvature (χ _failure) is derived from the actual steering errors, the steering error (using a steering angle sensor 8th ) be determined. Method according to one of claims 1 to 3, characterized in that a second maximum speed (V _max ) from the quotient of the visible curve area (S _visible ) and a driver-typical time for preparing future steering maneuvers is determined, and that the minimum of the first and second maximum speed (V _max ) is determined and output. A method according to claim 4, characterized in that the visible curve area (S _visible ) by means of environment sensors ( 9 ), in particular a video system, and / or from the map data ( 2 ) is determined. A method according to claim 4 or 5, characterized in that the driver-typical time for the preparation of future steering maneuvers, in particular a first time for a known and a second time for unknown terrain, is derived from the driving behavior of the driver. Method according to one of claims 4 to 6, characterized in that for determining the visible curve area (S _visible ) also environment information ( 9 ), in particular on current weather conditions, are used. _(Leave S) A method according to any one of claims 1 to 7, characterized in that a third maximum speed (V _max) from the quotient of a still available for the steering correction movements roadway area and a time required for the steering correction movements driver typical reaction time is determined, and that determines the minimum of the determined maximum speeds (V _max ) ( 16 ) and issued ( 17 ) becomes. A method according to claim 8, characterized in that for determining the still available roadway area (S _leave ) the roadway width (B) by means of environment sensors ( 9 ), in particular a video system, and / or from the map data ( 2 ), and the steering correction movements ( 8th ) are determined by means of a steering angle sensor. A method according to claim 8 or 9, characterized in that the steering correction movements in their dependence on environment information ( 9 ), in particular on the occurrence of gusty crosswinds or the presence of potholes, recorded, learned and used in the occurrence of the corresponding environmental conditions. Method according to one of claims 8 to 10, characterized in that instead of the actual roadway width (B) detects a driver-typical, in particular dependent on the respective environmental conditions, pendulum range, learned and used to calculate the third maximum speed (V _max ). Device which is designed to carry out a speed warning, in particular a curve warning for a motor vehicle, according to a method according to one of claims 1 to 11, and the first determination means ( 2 . 4 ) for the determination of a forward path curvature (χ _curve ), second determination means ( 6 . 8th ) for determining a likely faulty curvature (χ _fault ) attributable to the driver's expected steering errors (χ faulty), and a control unit ( 5 ) having means calculating a first maximum velocity (V _max ).

Images