Method and device for determining driver fitness

Abstract

Procedure for determining the fitness to drive of the driver of a vehicle (1), comprising at least the following steps: Determination of a yaw rate (w(t)) of the vehicle (1) as a function of time (t) (St1), Determination of a yaw rate property function (G (t)) from the determined yaw rate (w(t)) (St2), Comparison of the yaw rate property function (G (t)) with at least one comparison value (tg1, tg2, tf1, tf2) (St2), and Determination of a driver fitness level (DL) of the driver depending on the comparison (St3), wherein the driver fitness level (DL) is an experience level (EL) or includes an experience level (EL), where a rate of change (fg) is determined as the yaw rate property function (G (t)) as the number of changes in the yaw rate (ω (t)) and / or the number of sign changes (n2) of the yaw rate (ω (t)) and is compared with at least one yaw rate rate change limit (tf1, tf2), where exceeding at least one yaw rate change limit (tf1, tf2) or a higher number of exceedances of at least one yaw rate change limit (tf1, tf2) results in a conclusion that the driver has a lower level of driving ability (DL), in particular a lower level of experience (EL), than if there were fewer exceedances.

Description

[0001] The invention relates to a method for determining the fitness to drive of the driver of a vehicle and to a corresponding device.

[0002] Various methods are known for determining the driver's level of attention: This allows, firstly, the driver himself to be monitored; for example, monitoring the driver's pulse rate or blink rate is known to determine if the driver is tired.

[0003] Furthermore, it is known to determine and evaluate the vehicle's driving behavior in order to infer the driver's level of attention. DE 44 00 207 A1 discloses a corresponding method in which, in addition to the driver's heart rate, steering angle data from the steering angle entered by the driver are used. DE 10 2006 043 676 A1 describes an evaluation of driver attention based on the longitudinal and / or lateral acceleration of the vehicle; higher longitudinal and / or lateral accelerations that are outside a comfort range are interpreted as indicating increased driver attention. If the determined lateral acceleration remains below a certain threshold, which can be empirically determined, a lower level of driver inattention is inferred. Depending on the determined level of attention, a warning function for dangerous driving situations can then be suppressed.

[0004] German patent DE 10 2009 009 975 A1 describes a method for determining a driver's attentiveness while driving, in which a statistical evaluation of a plurality of measured values ​​over a period of time is carried out. Here, a deviation of the driving direction from an ideal driving direction is determined, and the distribution of the deviation thus determined is compared with a Gaussian distribution function, which is intended to represent an attentive driver.

[0005] DE 10 2008 056 593 A1 describes a procedure for warning a driver of fatigue, in which sensory data are first used to assess inattention and, if a trigger threshold is exceeded, it is checked whether a critical event indicative of fatigue has occurred.

[0006] DE 10 2009 005 730 A1 describes a monitoring of a driver's attention, in which the driver's actual gaze direction is compared with a target gaze direction in which the driver should look when paying attention.

[0007] In DE 197 34 307 A1, the driver's direction of view is compared with the direction of travel, which is determined by a direction-determining device, e.g., via a yaw rate, a steering angle, or a lateral acceleration of the vehicle.

[0008] DE 10 2010 048 273 A1 describes a method and a driver assistance system for initiating a vehicle action based on attention, in which an actual driving path of the driver is compared with a target driving path, and the driver's attention is inferred from this.

[0009] In DE 10 2009 004 487 A1, data from a steering wheel angle sensor are used to detect steering quiescence phases and steering actions, whereby steering errors are determined and weighted.

[0010] DE 196 00 938 A1 discloses a driving condition monitoring device for vehicles that uses sensors and algorithms to analyze and evaluate a driver's driving behavior. Parameters such as speed are recorded to detect deviations from normal driving behavior.

[0011] DE 10 2009 009 975 A1 discloses a method for determining a driver's attention level while driving, based on sensor data, in particular on steering activity and following distance. The measured values ​​are statistically evaluated over a specific period of time, using parameters such as kurtosis and skewness of the distribution, as well as comparative measures from reference drives.

[0012] Methods that rely on sensing the driver's body or behavior are inherently uncertain and depend on the driver's individual biological characteristics. For example, if the driver is wearing sunglasses or has an altered heart rate for reasons unrelated to their level of alertness, this can lead to a misinterpretation.

[0013] When evaluating a vehicle's handling, some of these methods require additional sensors, such as a lateral acceleration sensor, which are not otherwise present in the vehicle. This results in corresponding additional costs. Furthermore, these assessments incorporate certain assumptions about the ideal driving behavior of a driver, which are rather questionable in reality. For example, some methods assume that an attentive driver will typically make larger deviations from an ideal driving trajectory or even regular deviations from it; however, such assumptions generally cannot be further substantiated.

[0014] The invention is based on the objective of creating a method and a device for determining the driving behavior of a vehicle, which enable a relatively accurate determination with relatively little effort.

[0015] This problem is solved by the method according to claim 1 and the apparatus according to claim 12. The dependent claims describe preferred embodiments.

[0016] According to the invention, a driver's fitness level is generally determined. This is done using a yaw rate or yaw speed, i.e., the rotational movement of the vehicle around its vertical axis, which can be measured in particular by a yaw rate sensor.

[0017] The underlying principle is that yaw rate can be determined with relatively high accuracy compared to, for example, longitudinal and lateral accelerations. Thus, incorporating yaw rate allows for a better assessment than, for example, a lateral acceleration sensor, since lateral acceleration is also directly dependent on vehicle speed and therefore requires more complex compensation and can exhibit a higher error margin. According to the invention, yaw rate is also highly relevant for assessing the driver suitability level. Furthermore, yaw rate sensors are often already used by vehicle dynamics control systems, so existing sensors can potentially be integrated.

[0018] The driver's fitness level includes a driver attention level and / or a driver experience level, each of which can be determined based on specific criteria.

[0019] Low driver attention is caused by an inattentive driver. An inattentive driver is distracted or overtired and no longer fully participates in road traffic. Low driver experience is assumed in an inexperienced or nervous driver. While the inexperienced or nervous driver still participates attentively in road traffic, they nevertheless have difficulties controlling the vehicle and are therefore exposed to an increased risk of accidents, especially in critical situations. An attentive and experienced driver is assumed to make only small steering adjustments to maintain course. This will manifest as a low yaw rate when driving straight ahead, and as a constant yaw rate offset when cornering, depending on the radius of the curve and at a constant speed.

[0020] An inattentive or inexperienced driver can be identified by the following two behaviors: 1. Inexperienced drivers (nervous, unfamiliar) will exhibit increased swaying in their lane. This swaying differs from that of attentive and experienced drivers in that it involves more intense and frequent steering interventions. 2. An inattentive (overtired, distracted) driver will be recognizable by drifting out of their lane. A drift manifests itself through a minimal yaw rate, which, due to the driver's lack of course correction, leads to a change in the yaw angle, which in turn results in a continuously increasing deviation from the vehicle's intended course. At the end of the drift, the driver's corrective intervention causes a significant increase in the yaw rate. This corrective intervention can vary in intensity, depending on whether the driver is startled by the increased deviation from the course and makes a quick correction, or whether, due to inattentiveness, they generally make fewer corrections and therefore the corrective intervention does not result in an increased yaw rate. Regarding point 1: Recognizing the inexperienced driver's tendency to sway

[0021] The evaluation of the determined yaw rate is performed using a yaw rate property function. According to one embodiment, the temporal change of the yaw rate or its frequency within a measurement period can be used for this purpose. For example, the number of sign changes of the yaw rate can be used, thus providing an indication of counter-steering movements. Preferably, it is assumed that the yaw rate signal must exceed a minimum tolerance range around zero in order to detect a sign change and distinguish it from signal noise. Alternatively and additionally to the evaluation of the temporal behavior or frequency, an evaluation of the amplitude or magnitude of the yaw rate is also possible. Both the frequency or temporal change and the amplitude or magnitude of the yaw rate can be compared with one or more limit values.This allows us to determine whether one or both limit values ​​were exceeded once or several times during the measurement period.

[0022] According to the invention, a high degree of driver inexperience is advantageously assumed when larger changes in yaw behavior are detectable, i.e., greater temporal changes or a higher frequency of changes, and / or a higher amplitude of changes or deviations of the yaw rate from a mean value. This distinguishes the method according to the invention from known systems of the type mentioned above, which only investigate inattention that is detected at smaller deviations, e.g., in longitudinal acceleration. In contrast, the invention assumes that an experienced driver will make only smaller corrections than an inexperienced driver who, due to, e.g., nervousness or a delayed reaction, makes sharper steering inputs and / or a greater number of steering movements, which will lead to a correspondingly higher number of yaw rate changes and / or larger changes in the amplitude of the yaw rate signal.

[0023] Advantageously, a corrected yaw rate can first be determined from the yaw rate signal. This corrected rate takes into account the curve of the driving lane, so that, for example, in a right or left curve, the respective yaw rate required to follow the lane is subtracted. This can be done by a simple subtraction, where a curve yaw rate or lane yaw rate is subtracted from the absolute yaw rate determined from the yaw rate signal. When determining this corrected yaw rate, other driving events can also be considered or subtracted, in particular lane changes and turning maneuvers.

[0024] For example, a turn signal and, if applicable, map data about turns or the number of lanes on one's own roadway can be used for evaluation.

[0025] To determine the fluctuation, a statistical variance can be calculated. If the variance is greater than a limit σ 2 AttThresYR This suggests an inattentive driver.

[0026] To calculate the variance, it is advantageous to use an expected value or mean value around which the variance is calculated. During straight-line driving or at the corrected yaw rate, the expected value is zero. If an expected value is not available, it can be determined by the mean value over the measurement period. If the measurement period is short enough, cornering maneuvers can also be included. To mask the transition into cornering, time windows in which the mean value has changed by more than a predefined threshold compared to the previous time window can be omitted from the analysis. This allows the transitions from cornering to straight-line driving to be hidden.

[0027] To avoid misinterpreting lane changes, where the driver's attention is likely to be high, as inattention using this method, an additional mechanism is preferably employed to filter out lane changes. A high change in yaw rate indicates a lane change. If the change in yaw rate exceeds a threshold, GR LaneChange , the measurement will not be considered for a certain period of time. Regarding point 2: Detecting drift

[0028] To detect drift, the yaw angle is preferably evaluated, i.e., the time integral of the yaw rate, which is used, for example, as an offset relative to traveling straight ahead. Such an evaluation can be performed, for example, as a statistical variance, i.e., an evaluation of the square of the deviation of the yaw angle from the expected value (target course).

[0029] Drift itself can be detected primarily by monitoring the road's trajectory. This signal can be acquired via a road sensor, such as a radar sensor. A radar sensor typically provides a signal indicating the road's trajectory (based on an analysis of existing vehicles and roadside structures). Map data from a navigation device can also be used to supplement this information. To detect drift, the deviation of the vehicle's heading from the road's trajectory (yaw angle deviation) is determined. This deviation is measured as the variance of the heading angle deviation. This variance is higher for an inattentive driver than for an attentive one. However, lane changes and turning maneuvers are preferentially filtered out.The yaw rate is measured; if it exceeds a threshold, the heading deviation is ignored until it falls below a threshold again. This means the measurement is only resumed when the vehicle is (almost) back in its lane. A high yaw rate combined with a small heading deviation suggests cornering. The influence of cornering can be further adjusted using an additional parameter. Turning maneuvers and lane changes can also be detected via the activated turn signal.

[0030] Even without map data, deliberate changes in direction can be detected. For example, if the yaw rate exceeds a certain threshold, the course angle deviation can be ignored until it falls below the threshold again. Unlike conventional systems, large course deviations are therefore not interpreted as a sudden lapse in attention, but rather as a deliberate change of lane or roadway.

[0031] Another way to detect drift is by monitoring vehicles ahead. These are detected by a sensor, such as a radar sensor or camera, and thus made available to the system. It is assumed that these vehicles, like the vehicle in front, follow the lane. Using information about the lateral speed, distance, and offset of the vehicles ahead, as well as the current direction of travel of the vehicle in front, the yaw angle and yaw rate of the vehicle in front can be calculated relative to the road surface. This yaw angle and yaw rate can then be used as corrected yaw angle and corrected yaw rate. The accuracy of the calculation depends on the number of vehicles available.

[0032] For the quantitative determination of a driver's fitness level, i.e., a level of attention and / or experience, comparative values ​​are preferably used, in particular one or more threshold values. The exceeding of a single threshold value can be determined and evaluated, or the exceeding of multiple threshold values, e.g., a lower and an upper threshold value. In this way, the number of times one or more threshold values ​​are exceeded can be assessed.

[0033] The invention is explained in more detail below with reference to the accompanying drawings, which illustrate several embodiments. The drawings show: Fig. 1 a driving scene with a vehicle according to an embodiment of the invention with a first driving behavior with a high degree of attention; Fig. 2 a street scene with the vehicle from Fig. 1 in a second driving behavior with a low level of driver attention; Fig. 3 the vehicle out Fig. 1, Fig. 2 as a block diagram with relevant elements; Fig. 4 a flowchart of a process according to the invention; and Fig. 5 tables for evaluating attention.

[0034] Fig. Figure 1 shows a street scene 2, or driving situation or road situation of a vehicle 1, in particular a commercial vehicle, on a road 3 with three lanes 3a, 3b and 3c, which are separated by lane markings 4a. In the street scene 2 shown, the vehicle 1 is traveling together with other road users 5 in the same direction of travel, i.e. the X-direction; in principle, street scenes in which lanes with oncoming vehicles are adjacent are also relevant.

[0035] Vehicle 1 is driving according to Fig. 3 with a speed v and an acceleration a. The Fig. 1 and Fig. Figures 2 show the driving situation at time t0; Figure 6 shows the driving trajectory of the vehicle. Fig. 1 and the travel trajectory 7 of the Fig. Figures 2 show the position or driving path of vehicle 1 during a measurement period from t0 to a time tf.

[0036] According to the invention, the driving behavior of vehicle 1 is described in particular by its yaw rate ω (time-dependent change of the yaw angle), which is a function of time, i.e., ω = ω(t). For this purpose, vehicle 1 has, according to Fig. 3 Advantageously, a yaw rate sensor 8 is provided which measures the yaw rate ω as a function of time t and outputs a yaw rate signal S1 to a control unit 10. The control unit 10 can, in particular, be a control unit of an emergency braking system, e.g., a control unit of a vehicle dynamics control system with an emergency braking function.

[0037] The control unit 10 receives the yaw rate signal S1, determines a yaw rate property function G(t), and evaluates it by comparison. Various yaw rate property functions G(t) and evaluation criteria are provided as alternatives or supplementary options: According to a first evaluation criterion K1, the magnitude |ω (t)| of the yaw rate ω (t) is formed as the yaw rate property function G (t) and compared with one yaw rate limit tg1 or two yaw rate limit tg1, tg2. i.e., ω(t)>tg1 and ω(t)>tg2?

[0038] In this process, amplitude values ​​can first be determined as maximum values ​​or local maxima of the time function of the magnitude |ω (t)|, i.e., corresponding to the amplitude of a sine oscillation, and these amplitude values ​​can then be compared with the yaw rate limits tg1, tg2.

[0039] For example, it can be determined whether a yaw rate limit tg1, tg2 is exceeded one or more times during the measurement period t0 to tf, whereby different numbers n1 of exceedances can be determined for the various yaw rate limits tg1, tg2. From the number n1 of exceedances of the individual yaw rate limits tg1, tg2 during the measurement period t0 to tf, a driver attention level AL is then determined, e.g., by means of a tabular assignment.

[0040] The evaluation of the magnitude |ω (t)| of the yaw rate ω (t) can also be done by statistical evaluation, i.e. by determining an average value and e.g. a statistical variance σ, which thus represents the yaw rate property function G(t), and a comparison with target variance values ​​or classification of the variance as a degree of attention.

[0041] According to a second embodiment, a second evaluation criterion K2 is used, in which a temporal behavior, in particular the temporal derivative dω (t) / dt or the frequency or a number of changes of the yaw rate ω (t), is evaluated as the yaw rate property function G (t).

[0042] When evaluating the time derivative dw(t) / dt, the first derivative can be set equal to zero, according to standard calculation methods in analytical mathematics, in order to determine local minima and maxima, i.e., dw(t) / dt = 0, and the sign of the second derivative d 2 ω(t) / dt 2 .

[0043] In the simplified case of a sinusoidal driving trajectory 6, a temporal sinusoidal function of the yaw rate ω (t) also results - at least at constant driving speed v - so that an oscillation with a fixed frequency can be assumed here; in reality, however, the vehicle 1 will perform these changes irregularly.

[0044] Advantageously, the number of changes in direction or the number n2 of sign changes of the yaw rate ω (t) in the measurement period t0 to tf can be used and compared with a threshold tf1 or a lower threshold tf1 and an upper threshold tf2.

[0045] The number n2 of sign changes of the yaw rate ω (t) in the measurement period t0 to tf is referred to as the "change frequency" fg of the yaw rate w(t); this change frequency fg can be compared with yaw rate change thresholds tf1, tf2 and the driver's level of attention AL or experience EL can be determined from this.

[0046] It is recognized here that the evaluation criteria K1 and K2 are initially meaningful when the vehicle is traveling straight ahead, whereas the results of the evaluation criteria may need to be corrected for driver-initiated lane changes or cornering. Instead of the measured absolute yaw rate, a corrected yaw rate can be used as the yaw rate ω(t). For example, when driving in a curve, a curve yaw rate ωk corresponding to the curve trajectory is advantageously calculated; this can be determined, for example, from map data or by detecting the lane markings 4a and thus serves as the zero line or reference value for the determinations of the preceding evaluation criteria K1 and K2. In the simplest case, the current absolute yaw rate w(t) can therefore be reduced by this value, i.e., a corrected yaw rate wc(t) = w(t) - ωk is used, where ωk can also be time-dependent.

[0047] Furthermore, turning maneuvers and deliberate lane changes in lane 3b can also be detected, for which either a curve yaw rate ωk is determined for these maneuvers or no evaluation takes place during the maneuver. For the detection of turning maneuvers and lane changes, a turn signal and, if necessary, map data can be used, for example.

[0048] Advantageously, the vehicle 1 additionally has a road sensor 20, e.g. radar sensor and / or an optical sensor unit, for determining the lane 3b or the lane boundary 4a, which outputs road signals S4 to the control unit 10.

[0049] As an additional variant, instead of the corrected yaw rate wc(t), a yaw rate wc(t) = ωv(t) derived from the direction of movement of the vehicles ahead can be used. According to a further embodiment, a combination of the yaw rates can be applied using different weighting factors ωc(t) = a * wc(t) + (1 - a) ωv(t), where the factor a varies between 0 and 1 depending on the number of vehicles ahead. For example, if there are many vehicles ahead, a = 1. If there are no vehicles ahead, a = 0.

[0050] According to a third embodiment, a time integration of the corrected yaw rate (ωc(t)) is performed, yielding a corrected yaw angle ψc(t) with respect to a reference direction. The corrected yaw angle ψc(t) can then be statistically evaluated, for example, by calculating a statistical variance from the square of the deviations from an expected value. The expected value of the corrected yaw rate is zero. If the corrected yaw rate wc(t) is not available, the uncorrected yaw rate ω(t) can also be used, with the expected value being determined from the average value of the yaw rate ω(t). Thus, the statistical function of the yaw angle ψ(t) or the corrected yaw angle ψc(t) represents the yaw rate property function (G(t)), which can be compared with reference values.In this context, a high value of the yaw rate property function (G (t)) and thus a high variance suggests an inattentive driver.

[0051] In both embodiments, the evaluation assumes that an attentive driver drives with greater concentration and makes fewer or smaller corrections to the current driving state of vehicle 1, resulting in very small values ​​in the variance of the course angle deviation. An inattentive driver will make very few, but usually drastic, corrections and exhibit a very high variance in the course angle deviation. A nervous or inexperienced driver, on the other hand, will make many interventions with medium amplitudes and medium variance in the course angle deviation. (See Table 1 of the...) Fig. Section 5 provides an example of the relationship between the expected characteristics and the investigated criteria K1, K2, and K3.

[0052] For quantitative evaluation, a table can be used, for example, which assigns the respective threshold violations to a specific level of attention AL (e.g., AL with values ​​between 0 and 10, with AL = 10 representing the highest level of attention) and a level of experience EL (e.g., EL with values ​​between 0 and 10, with EL = 10 representing the highest level of experience) according to the evaluation criteria K1, K2, and K3. This is illustrated in Table 2. Fig. 5 shown.

[0053] For each evaluation criterion, the level of experience and the level of attention are determined separately. Table 2 shows the points awarded for each criterion, depending on whether the thresholds are exceeded. Finally, the points for the three levels of attention are preferably added together. An inattentive driver is identified if the sum of the attention levels is less than a certain threshold. An inexperienced or nervous driver is identified if the sum of the experience levels is less than a certain threshold.

[0054] According to the first evaluation criterion K1, lower amounts |ω(t)| or |ωc(t)| are to be expected for an attentive and experienced driver.

[0055] Thus, the driver is very attentive, i.e., there is a high level of attention AL, e.g. AL= 10, if both limit values ​​tg1, tg2 are not exceeded at all during the measurement period t0 to tf. Exceeding only the lower threshold tg1 indicates a slightly lower level of attention AL, e.g. AL= 8. If the lower threshold tg1 and / or the upper threshold tg2 are exceeded multiple times, the level of attention AL is again lower, e.g. AL = from 2 to 6. The same applies to the experience level EL.

[0056] According to the second evaluation criterion K2, an experienced driver is recognized, i.e. EL= 10, if the driver performs a very small number of sign changes of the yaw rate (ω (t)) during the measurement period t0 to tf, i.e. the change frequency fg is low and the yaw rate change limits tf1, tf2 are not exceeded at all or only rarely.

[0057] If only the lower threshold tf1 is exceeded, a slightly lower experience level EL is present, e.g. EL= 5.

[0058] If the lower threshold tf1 and / or the upper threshold tf2 are exceeded multiple times, a lower experience level EL is indicated, e.g., EL = 0 to 5. For attention detection, a high level of attention is preferably assumed when the upper threshold tf2 is exceeded, e.g., Al = 10. The same points are awarded for exceeding and falling below the threshold tf1, as a precise differentiation between attentive and inattentive behavior is not possible.

[0059] The third evaluation criterion, K3, provides a clearer statement regarding attention.

[0060] Depending on the transmitted level of attention AL, different procedures can be carried out. In particular, a display signal S2, e.g., a warning signal, can be output to a display device 12 for visual and / or audible indication to the driver, thus informing them of their level of attention. Furthermore, the display signal S2 can also be recorded by a tachograph.

[0061] Furthermore, depending on the attention level AL, settings of other systems can also be made; for example, the massage function of a vehicle seat can be changed. Additionally, brake setting signals S3 at wheel brakes 11 can be used to adjust the brake setting or brake preload depending on the determined attention level AL, in order to subsequently enable faster braking.

[0062] Thus, according to the invention, a driver fitness level DL is determined which includes the attention level AL and / or the experience level EL. The relevant, i.e., the more problematic, of these two values ​​can therefore be determined, so that, for example, a low driver fitness level DL is formed by the low attention level AL, or by the low experience level EL.

[0063] From the attention level AL and the experience level EL, a scalar value of the driver suitability level DL can additionally be formed, e.g. as an arithmetic mean; however, this is not necessary, since it is already recognized according to the invention that, for example, very high inattention, i.e., very low attention, cannot always be compensated for by high experience and can therefore already lead to dangerous driving situations.

[0064] The inventive method according to Fig. 4 therefore shows the following steps: After starting in step St0, in step St1 the yaw rate w(t) is continuously recorded as a function of time t by a yaw rate sensor 8 and output as a yaw rate signal S1 to the control unit 10, which can store the yaw rate signal S1, e.g., in an internal or external storage device 14, at least for a past time t0 to tf, in order to subsequently determine a yaw rate property function G(t) in step St2 and to perform an evaluation according to the first evaluation criterion K1 and / or the second evaluation criterion K2 and / or the third evaluation criterion K3.

[0065] In step St3, the attention level AL and experience level EL are determined, which are then used, if necessary, to output a display signal S2 or for subsequent evaluation. The procedure is carried out continuously during the journey, so that after step St3 the procedure is reset to the point before step St1. Reference numeral list (part of the description) 1 vehicle 2 Street scene 3rd Street 3a, 3b, 3c, lanes 4a Lane markings 5 other road users 6. Travel trajectory of the Fig. 1 in measurement period t0, tf. 7. Travel trajectory of the Fig. 2 in measurement period t0, tf. 8 Yaw rate sensor 10 Control unit 11 wheel brakes 12 Display device 14 internal or external storage devices 20 road sensor a acceleration DL driver fitness level AL Driver's level of attention EL experience level fg yaw rate change frequency ω(t) G (t) Yaw rate property function K1 first evaluation criterion K2 second evaluation criterion K3 third evaluation criterion n1 Number of yaw rate limit exceedances n2 Number of sign changes of the yaw rate ω (t) S1 Yaw Rate Signal S2 indicator signal S3 Brake Adjustment Signal S4 track signals t time t0, tf Measurement period tf1 lower yaw rate change limit tf2 upper yaw rate change limit tg1 lower yaw rate limit tg2 upper yaw rate limit v Driving speed X direction of travel ψ (t) yaw angle σ statistical variance ω = ω (t) Yaw rate ωk Curve yaw rate ωc (t) = w(t) - ωk corrected yaw rate dω (t) / dt time derivative of the yaw rate ω (t) Step St0 Start Step St1: Recording the yaw rate w(t) Step St2 Determining the yaw rate property function G(t) and evaluation Step 3: Determining the level of attention AL

Claims

[1] Method for determining the fitness to drive of the driver of a vehicle (1), comprising at least the following steps: Determination of a yaw rate (w(t)) of the vehicle (1) as a function of time (t) (St1), Determination of a yaw rate property function (G (t)) from the determined yaw rate (w(t)) (St2), Comparison of the yaw rate property function (G (t)) with at least one comparison value (tg1, tg2, tf1, tf2) (St2), and Determination of a driver fitness level (DL) of the driver depending on the comparison (St3), wherein the driver fitness level (DL) is an experience level (EL) or includes an experience level (EL), where a rate of change (fg) is determined as the yaw rate property function (G (t)) as the number of changes in the yaw rate (ω (t)) and / or the number of sign changes (n2) of the yaw rate (ω (t)) and is compared with at least one yaw rate rate change limit (tf1, tf2), where exceeding at least one yaw rate change limit (tf1, tf2) or a higher number of exceedances of at least one yaw rate change limit (tf1, tf2) results in a conclusion that the driver has a lower level of driving ability (DL), in particular a lower level of experience (EL), than if there were fewer exceedances. [2] Method according to claim 1, characterized by , that the yaw rate property function (G (t)) has or represents an absolute value (|ω (t)|) of the yaw rate (w(t)) and is compared with at least one yaw rate limit (tg1, tg2), where it is determined whether or how often the at least one yaw rate limit value (tg1, tg2) is exceeded in a measurement period (t0; tf) and, depending on this determination, a driver fitness grade (DL) of the driver is determined. [3] Method according to claim 2, characterized by , that The driving ability level (DL) is an attention level (AL) or includes an attention level (AL), wherein a number (n1) of exceedances (tf1, tf2) of the at least one yaw rate limit value (tg1, tg2) is determined in the measurement period (t0; tf) and a higher number (n2) of exceedances is used to conclude that the driver has a lower level of fitness (DL), in particular a lower level of attention (AL) than a lower number (n1) of exceedances. [4] Method according to claim 2 or 3, characterized by , that at least a lower yaw rate limit (tg1) and an upper yaw rate limit (tg2) are provided and that if the lower yaw rate limit (tg1) and the upper yaw rate limit (tg2) are exceeded, it is concluded that the driver has a lower level of driving ability (DL; AL, EL) than if only the lower yaw rate limit (tg1) is exceeded. [5] Method according to any one of claims 2 to 4, characterized by , that a statistical function, e.g. a statistical variance (σ) of the absolute value (|ω (t)|) of the yaw rate (w(t)) is formed as the yaw rate property function (G (t)). [6] Method according to any one of the preceding claims, characterized by , that by integrating the yaw rate (ω (t)) over time a yaw angle (ψ(t)) is determined. [7] Method according to claim 6, characterized by , that a statistical function of the yaw angle (ψ(t)) is determined as the yaw rate property function (G (t)) in a measurement period (tf1, tf2), in particular a statistical variance (σ) which is subsequently evaluated to determine the driving ability (DL) of the driver. [8] Method according to claim 7, characterized by, that exceeding the lower limit (ts1) of the statistical variance (σ) of the yaw angle (ψ(t)) and the upper limit (ts2) of the statistical variance (σ) of the yaw angle (ψ(t)) indicates a lower level of attention (AL) and / or experience (EL) of the driver than merely exceeding the lower limit (ts1) of the statistical variance (σ) of the yaw angle (ψ(t)). [9] Method according to any one of the preceding claims, characterized by , that from a yaw rate signal (S1), which represents an absolute yaw rate (ω (t)) of the vehicle (1), a corrected yaw rate (wc(t)) is first determined by subtracting or calculating from the absolute yaw rate (ω (t)) a curve yaw rate (wk) corresponding to the lane (3b) of the vehicle (1), and subsequently the corrected yaw rate (ωc (t)) is used to determine the yaw rate property function (G (t)). [10] Method according to any of the preceding claims, characterized by that the corrected yaw rate (wc(t)) is determined based on measured signals from vehicles ahead. [11] Device for determining the fitness to drive of the driver of a vehicle (1) for carrying out a method according to one of the preceding claims, wherein the device comprises at least: a yaw rate sensor (8) for determining a yaw rate (w(t)) of the vehicle (1) and outputting a yaw rate signal (S1), and a control device (10) for receiving the yaw rate signal (S1) and to determine a yaw rate (w(t)) of the vehicle (1) as a function of time (t), to determine a yaw rate property function (G (t)) from the determined yaw rate (w(t)) and to compare the yaw rate property function (G (t)) with at least one comparison value (tg1, tg2, tf1, tf2) (St3), and to determine a driver aptitude level (DL) of the driver as a function of the comparison. [12] Vehicle (1) with a device according to claim 11 suitable for carrying out a method according to any one of claims 1 to 10.

Images