DE102020207825A1 - Method for monitoring the speed of a two-wheeled vehicle - Google Patents

Abstract

The invention relates to a method for monitoring the speed of a two-wheeler, in particular a motorcycle, in which, taking into account a curve radius (r) of a curve to be traveled and historical driving data (DH), an expected, driver-specific radius (rpred) and an expected, driver-specific inclined position (φpred) for the curve to be traveled at least at one point in the future, whereby a limit speed (vmax) at the at least one point in time for the to driving curve is determined, and if a current speed (v) of the limit speed (vmax) dependent speed threshold value (vs) exceeds, a countermeasure (G) is initiated.

Description

The present invention relates to a method for monitoring the speed of a two-wheeled vehicle, in particular a motorcycle, as well as a computing unit and a computer program for its implementation.

State of the art

For vehicles such as cars, routes to be driven and their conditions can be calculated in advance based on driving data and map data. This makes it possible, for example, to calculate certain maximum possible or permitted speeds in curves in advance. From the DE 10 2011 080 761 A1 For example, there is a known possibility of outputting a warning if a speed of the vehicle threatens to exceed a specific, predicted permissible speed in a curve. The idea of using driving data and / or map data, possibly also information from other road users or the like within a vehicle, is also known under the term of the electric horizon (or “eHorizon”) and is used, for example, in the DE 10 2012 210 454 A1 mentioned.

Disclosure of the invention

According to the invention, a method for monitoring the speed of a two-wheeled vehicle as well as a computing unit and a computer program for carrying it out are proposed with the features of the independent claims. Advantageous refinements are the subject matter of the subclaims and the description below.

The invention is concerned with speed monitoring in a two-wheeled vehicle, i.e. a single-track vehicle, in particular a motorcycle. Here, taking into account a curve radius of a curve to be traveled (e.g. a curve coming next on the currently traveled route section) and historical driving data, an expected, driver-specific cycle radius and an expected, driver-specific inclination for the curve to be traveled for at least one future point in time , ie one or in particular also several future points in time (then in each case for each of these points in time).

A limit speed for the at least one future point in time for the curve to be traveled is then determined on the basis of the (possibly respective) expected, driver-specific cycle radius and the expected, driver-specific inclination. Such a limit speed is to be understood, for example, as a speed which, if exceeded, would cause the two-wheeler to lose traction due to excessive centrifugal forces. Since this limit speed after entering the curve also depends on the associated real cycle radius and the inclined position, this limit speed can also behave differently for different points in time or change over time. In this respect, a corresponding value for the limit speed is calculated in each case for several, discrete points on the route or curve to be traveled.

If a current speed (of the two-wheeler) then exceeds a speed threshold value that is dependent on the limit speed, a countermeasure is initiated. The speed threshold value can be selected, for example, in such a way that at a current point in time there is still enough time to bring the two-wheeler to a speed below the limit speed by the future point in time (which is still in the future) (here, for example, the speed limit is also current speed to be taken into account again).

These variables can also be (re) determined continuously, for example, so that on the one hand the countermeasure only takes place when it is actually displayed, and on the other hand it is also possible to react to any, possibly also unexpected, changes in, for example, the speed.

The particular advantage of this procedure lies in the fact that the historical data takes into account, in particular, the driving ability of the individual driver. This means that, for example, experienced drivers who prefer a large lean angle do not have to take unnecessary countermeasures or warnings, since higher speeds are possible here than with low lean angles. In particular, the invention makes use of the knowledge that the (driving) radius with which a certain driver drives through a curve will as a rule deviate from the actual curve radius and from the cycle radius of other drivers. Likewise, a driver's lean angle will depend on his driving skills and will differ from the lean angles of other drivers. Within the scope of the invention, the cycle radius and lean angle are now predicted for each individual driver using historical driving data.

The countermeasure can include, for example, a warning to the driver of the two-wheeler, which can in particular take place acoustically and / or haptically and / or visually, i.e. also multi-modally, for example. For example, a vibration of the handlebar grip or a display in a possibly existing head-up display, the lighting up or flashing of a warning lamp and much more is conceivable.

However, it is also conceivable that the countermeasure includes, for example, an active reduction in the speed of the two-wheeler, in particular by reducing an engine torque (here, depending on the two-wheeler, it can be the torque of an electrical machine and / or an internal combustion engine) and / or takes place by means of generating a braking torque, and / or that the countermeasure comprises, for example, generating a steering torque and / or setting a steering angle. This allows the radius of the curve and the inclination to be changed. For example, a larger radius with less inclination can be “set” or specified as a countermeasure. The driving dynamics can therefore generally be actively adapted. It is conceivable, for example, to use a brake control system in a targeted manner or to influence a characteristic curve that specifies the conversion of the position of the drive handle into the engine torque generated. If appropriate actuators are available, the steering angle specified by the driver or his steering behavior can also be influenced by intervening in the steering (steering torque and / or steering angle) in order, for example, to change the real cycle radius through the curve, which is a higher, possible one Speed causes.

These two types of countermeasures can also be used, for example, one after the other, ie first a warning is issued, and later, if the driver does not reduce the speed or does not reduce the speed enough or in good time, the active reduction of the speed and / or a steering intervention can be used as a countermeasure respectively. The use of two different speed threshold values for this is also conceivable.

The curve radius of the curve to be traveled is preferably determined using map data. These map data can, for example, already be stored on an executing computing unit, or they can also be obtained externally from another computing unit of the two-wheeler or via wireless communication links from a server, for example. Such map data, which has already been determined at an earlier point in time, can come from one or more of the following sources, for example: high-resolution maps from the field of automated driving, maps obtained from data from a fleet of two-wheelers, satellite images and geo-informational databases . Depending on availability, particularly precise information about the radius of the next curve can be determined.

Alternatively or additionally, however, it is advantageous if the curve radius of the curve to be traveled is determined using one or more sensors on the two-wheeler. For example, lane recognition cameras or radar, lidar and ultrasonic sensors can be used as sensors. These sensors, used per se for other purposes, nevertheless allow a curve radius to be determined (at least if this is within range of the sensors).

These two ways of determining the curve radius (map data and sensors) can also be combined to increase accuracy.

The historical driving data, which are also continuously recorded and stored in an executing computing unit, preferably include data on at least one of the following variables, in particular on a large number of curves driven: a speed of the two-wheeler, an incline, an (actual) curve radius and of a real cycle radius of the curve. The speed can be determined e.g. by means of a wheel speed sensor, GPS, radar and / or video sensors. The inclination indicates the inclination of the two-wheeler in relation to the roadway (as an angle) and can be determined, for example, using inertial sensors, GPS, radar, video sensors and / or lidar. The actually ridden cycle radius can, for example, be determined on the basis of measured driving dynamics data (e.g. supported by inertial sensors and / or GPS).

These historical driving data (or driving dynamics data) allow an assessment of the driving behavior or driving ability of the individual driver and thus also a prediction for future behavior in the next bend.

The expected, driver-specific cycle radius (i.e. within the scope of the prediction) is preferably determined using a model based on artificial intelligence (i.e. a machine learning model or a model based on machine learning), in particular a neural network. The same applies to the anticipated, driver-specific inclination, whereby separate models can be used for these two sizes, but a common model is also conceivable. In general, however, different implementations of the models come into consideration here. One type of model are parametric models, in which a distinction can be made between linear and non-linear regression models. For the latter, in turn, the (artificial) neural networks come into consideration, namely expediently recurrent neural networks such as long-term-short-term (LSTM) neural networks or so-called convolutional neural networks such as temporal convolutional neural networks. Another type of model are nonparametric models such as Gaussian processes or decision trees. Long-term-short-term (LSTM) neural networks in particular provide good results here.

wobei r_pred den zu erwartenden, fahrerindividuellen Fahrradius der Kurve und φ_pred die zu erwartende, fahrerindividuelle Schräglage angeben. Mit komplizierteren Modellen kann die Grenzgeschwindigkeit ggf. aber noch genauer bestimmt werden. So umfasst z.B. das vorgenannte Modell keine Reifeneinflüsse, die in detaillierteren Modellen z.B. umfasst sein können und damit auch nichtlineare Einflüsse berücksichtigen.The limit speed at the specific point in time for the curve to be traveled is preferably determined using a physical model of driving dynamics of the two-wheeler. In a simple case, this can be done using an ideal roll angle, for example. For this purpose, only the acceleration due to gravity g has to be taken into account and the limit speed v _max then results, for example, according to the following formula: $${\nu }_{\text{Max}}=\sqrt{\frac{G\cdot \text{tan}{\phi }_{pred}}{r_{pred}}},$$

where r _{pred indicates} the expected, driver-specific cycle radius of the curve and φ _pred the expected, driver-specific inclination. With more complicated models, the limit speed can, if necessary, be determined even more precisely. For example, the aforementioned model does not include any tire influences that can be included in more detailed models, for example, and thus also take non-linear influences into account.

With the proposed procedure, taking into account the individual ability or behavior of a driver of a two-wheeler, speed monitoring including a warning or, if necessary, intervention in the driving dynamics for a two-wheeler when cornering can be provided.

A computing unit according to the invention, for example a control unit of a motorcycle, is set up, in particular in terms of programming, to carry out a method according to the invention.

The implementation of a method according to the invention in the form of a computer program or computer program product with program code for performing all method steps is advantageous, since this causes particularly low costs, especially if an executing control device is used for other tasks and is therefore available anyway. Suitable data carriers for providing the computer program are, in particular, magnetic, optical and electrical memories, such as hard drives, flash memories, EEPROMs, DVDs, etc. A program can also be downloaded via computer networks (Internet, intranet, etc.).

Further advantages and embodiments of the invention emerge from the description and the accompanying drawing.

The invention is shown schematically in the drawing using an exemplary embodiment and is described below with reference to the drawing.

Figure list

• 1 shows schematically a two-wheeler in which a method according to the invention can be carried out.
• 2 shows schematically the two-wheeler 1 in an environment.
• 3 shows schematically a sequence of a method according to the invention in a preferred embodiment.

Embodiment (s) of the invention

In 1 is schematically a two-wheeler 100 , which is designed, for example, as a motorcycle or motorcycle in which a method according to the invention can be carried out, with a driver 101 shown. The two-wheeler 100 comprises a number of components or units which can be used in carrying out the proposed method.

A warning unit is an example of this 110 , an environmental sensor system 112 , a steering actuator 113 , a GPS unit 114 , an evaluation or arithmetic unit 116 , an ABS hydraulic system 118 , two wheel speed sensors 120 , an inertial sensor system 122 and a communication unit 124 intended.

The warning unit 110 can, for example, have a display for displaying a warning or warning message as a countermeasure when the speed threshold value is exceeded or can be communicatively linked to such a display. The environmental sensors 112 can for example include a camera, a lidar or an ultrasonic sensor and be used to determine the curve radius of a curve ahead, but also to determine or correct the speed of the two-wheeler if necessary. The steering actuator 113 can, for example, apply a steering torque to the steering or actively change the steering angle.

The GPS unit 114 can, for example, also be used to determine or, if necessary, correct the speed, but also to determine or correct, for example, the inclination or the angle of inclination of the two-wheeler. The evaluation or arithmetic unit 116 can - in the sense of a computing unit on which the proposed method is carried out - be used to evaluate the various data and calculate the relevant variables. For this purpose it is useful if the evaluation or arithmetic unit 116 can communicate with the other components or units accordingly.

The ABS hydraulics 118 basically serves to regulate the brakes on two-wheelers and can be used accordingly within the framework of the proposed method be controlled in order to actively reduce the speed of the two-wheeler, for example, as a countermeasure. The wheel speed sensors 120 can be used to first record the wheel speed (s), from which the speed of the two-wheeler can then be determined.

The inertial sensors 122 can be used to determine the lean angle or the angle of inclination of the two-wheeler, and the communication unit 124 can be used to receive map data from an external server, for example.

It should be noted that the components or units or sensors shown and described here are only exemplary, as is their use.

For further possibilities of obtaining corresponding data and measuring or determining variables, reference is also made to the above explanations.

In 2 is schematically the two-wheeler 100 out 1 in an environment, especially on a curve 200 that are on a currently traveled route or a route section 210 is shown. This is for the two-wheeler 100 generally an incline is shown as the angle of inclination φ, and for the curve 200 are an (actual or geometric) curve radius r (as it is typically determined based on the center of the lane), a real or current cycle radius r _real , as it is for a two-wheeler 100 or whose lane is to be assigned, as well as an expected, individual cycle radius r _pred .

In 3 a sequence of a method according to the invention is shown schematically in a preferred embodiment, as is the case with a two-wheeler, as shown in FIG 1 and 2 shown can be performed.

During the operation of the two-wheeler, in particular always in the area of curves that are being driven, various variables such as a speed v of the two-wheeler, actual curve radii r of the curves, bicycles actually ridden by the driver r _{real of} the curves as well as associated inclinations or angles of inclination φ ascertained or determined in the curves and _{stored as historical driving data D H} in, for example, a memory of the processing unit.

In addition, the actual curve radius r is determined for a curve to be driven next, for example using map data K and / or sensors S, as has already been explained in detail above.

A (first) neural network M ₁ (or possibly another model based on artificial intelligence) is then used for this curve to be driven next, taking into account the curve radius r of this curve and the historical data D _{H as input variables} , individual cycle radius r _{pred determined} in the curve for a future point in time. In addition, for this curve to be driven next, taking into account the curve radius r of this curve and the historical data D _H as input variables, a (second) neural network M ₂ (or possibly also another model based on artificial intelligence) is expected to be , individual inclination φ _{pred determined} in the curve.

On the basis of the expected, driver-specific bicycle radius r _pred and the expected, driver-specific inclination φ _{pred , a limit speed v max} for the future point in time is then determined using a physical model M _P , which, for example, depicts the driving dynamics of the two-wheeler and which was explained above by way of example determined or determined for the curve to be traveled. The prediction horizon, ie the future point in time or possibly also the future points in time for which the limit speed is determined, can preferably also depend on the speed. The higher the current speed, the further ahead into the future the limit speed should be determined, ie more and / or points in time further in the future will be expedient.

The limit speed v _max or the underlying variables are preferably determined continuously or again and again for the current point in time, so that a countermeasure can always be initiated in good time. In particular, the limit speed v _{max is} constantly updated with knowledge of the specific circumstances, such as the current driving speed, current inclination and current cycle radius r _real. However, it is also conceivable that the limit speed v _max or the underlying variables are only determined for one or, if necessary, several specific locations in the curve.

In addition, a current speed v of the two-wheeler is then continuously compared with the limit speed v _max by determining a speed threshold value vs that is dependent on the limit speed v _max. If the current speed v exceeds this speed threshold value vs, the countermeasure G is initiated, for example a warning is output and / or the two-wheeler is braked. The speed threshold value vs can, for example, be selected (and also continuously adapted) in such a way that the countermeasure is sufficient that the Limit speed v _{max is} not exceeded. The speed threshold value vs can, for example, _{correspond to the limit speed v max} or also be below it, in order to take reaction times or road surface conditions into account, for example. For example, braking or steering interventions in wet or slippery conditions cannot be as strong as in dry conditions.

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was 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.

Patent literature cited

• DE 102011080761 A1 [0002]
• DE 102012210454 A1 [0002]

Claims (12)

A method for monitoring the speed of a two-wheeler (100), in particular a motorcycle, in which, taking into account a curve radius (r) of a curve (200) to be driven and historical driving data (D _H ), an expected, driver-specific cycle radius (r _pred ) and a to expected, driver-specific inclination (φ _pred ) for the curve to be negotiated for at least one future point in time, with a limit speed (v _max ) _{based on the expected, driver-specific cycle radius (r pred} ) and the expected, driver-specific inclination (φ _pred) is determined for the at least one future point in time for the curve (200) to be traveled, and if a current speed (v) exceeds a speed threshold (vs) dependent on _{the limit speed (v max), a countermeasure (G) is initiated.} Procedure according to Claim 1 wherein the countermeasure (G) comprises a warning to a driver (101) of the two-wheeler (100), which in particular takes place acoustically and / or haptically and / or visually. Procedure according to Claim 1 or 2 , wherein the countermeasure (G) comprises an active reduction of the speed (v) of the two-wheeler, which takes place in particular by reducing an engine torque and / or by generating a braking torque, and / or generating a steering torque and / or setting a steering angle. Method according to one of the preceding claims, wherein the curve radius (r) of the curve to be traveled is determined using map data (K). Method according to one of the preceding claims, wherein the curve radius (r) of the curve to be traveled is determined using one or more sensors (S, 112, 114) on the two-wheeler. Method according to one of the preceding claims, wherein the historical driving data (D _H ) include data on at least one of the following variables, in particular in each case on a plurality of curves driven: a speed (v) of the two-wheeler, an incline (φ), a curve radius (r) of the curve and a real cycle radius (r _real ) of the curve. Method according to one of the preceding claims, wherein the expected, driver-specific cycle radius (r _pred ) is determined using a model (M ₁ ) based on artificial intelligence, in particular a neural network. Method according to one of the preceding claims, wherein the driver-specific inclination (φ _pred ) to be expected is determined using an artificial intelligence-based model (M ₂ ), in particular a neural network. Method according to one of the preceding claims, wherein the limit speed (v _max ) for the at least one future point in time for the curve (200) to be traveled is determined using a physical model (M _P ) of driving dynamics of the two-wheeler. Computing unit (116) which is set up to carry out all method steps of a method according to one of the preceding claims. Computer program that causes a processing unit (116) to perform all method steps of a method according to one of the Claims 1 until 9 to be carried out when it is executed on the computing unit (116). Machine-readable storage medium with a computer program stored thereon Claim 11 .

Images