Method for determining rotational speed and / or movement variable derived from rotational speed, computer program product, data carrier, computer device

Abstract

The invention relates to a method for determining a rotational speed and / or a movement variable derived from the rotational speed by means of a rotational speed sensor (1), wherein a corrected third signal value ([omega] p) at the selected reference time point is extrapolated or interpolated on the basis of an estimated information age (TI) of a first signal value ([omega] m) of the rotational speed sensor (1) evaluated at a first evaluation time point (ta) and a second signal value ([omega] b) occurring temporally prior to the first signal value ([omega] m), and wherein a rotational speed and / or a movement variable is determined as a function of the third signal value ([omega] p).

Description

Technical Field

[0001] The invention relates to a method for determining a rotational speed and / or a movement variable derived from the rotational speed by means of a rotational speed sensor. The invention also relates to a computer program product which, when executed on a computer device, executes the above method. The invention also relates to a data carrier having such a computer program product and a computer device specially designed for executing the computer program product or the above method. Background Art

[0002] Methods of the type mentioned above are known from the prior art. Such methods are used, for example, in the context of vehicles. Therefore, many functions represented in the ABS / ESP control device use the wheel peripheral speed (ω·r) of the wheels of the motor vehicle and the speed of the wheel center point relative to the ground (v Rad ) or a variation normalized to this speed (the so-called wheel slip) as the central control variable. The peripheral wheel speed is usually determined by measuring the rotational speed using a rotational speed sensor. Also known are methods that are intended to improve the accuracy when determining the vehicle speed at low speeds by using such a rotational speed sensor. For example, the publications DE 10 2008 009 406 A1 and DE 10 2018 222 072 A1 each disclose a method in which the vehicle speed is determined during slow driving by extrapolation. In the last-mentioned publication, the data of multiple vehicle wheels are fused. Summary of the invention

[0003] The method according to the invention is characterized in that a corrected third signal value at a selected reference time is extrapolated or interpolated based on an estimated information age of a first signal value of the rotation speed sensor evaluated at a first evaluation time point and a second signal value occurring temporally before the first signal value, and the rotation speed and / or the displacement variable is determined based on the third signal value. The information age determines the actual valid time point of the signal value that is temporally before the evaluation time point, and the accuracy in determining the actual signal value, especially at low speeds, is advantageously improved. Here, with regard to the above-mentioned application case for determining the wheel peripheral speed of the wheels of a motor vehicle, the invention is based on the recognition that when the motor vehicle is decelerating, the actual wheel peripheral speed at the time point at which the evaluated signal value is available is less than the assumed value, because there is a certain delay between detection and evaluation. Systematic errors caused by non-coincidence signals when calculating wheel differential speeds or wheel slip are avoided or at least greatly reduced by the method according to the invention. In particular, in addition to the information age of the wheel peripheral speed, the information age of the wheel center speed required for determining the wheel slip is also taken into account. However, these considerations can be transferred to any type of rotation speed measurement, even outside the motor vehicle environment. In order to at least reduce the resulting inaccuracies, according to the invention, an information age is taken into account, which corrects the time associated with the signal value to an earlier time and extrapolates or interpolates the actual signal value to a reference time, for example by drawing a straight line between the previous signal value at which the valid time is known and the current time-corrected signal value and extending it to the evaluation time used as reference time. The method according to the invention is particularly advantageous in the case of measuring pulses of (magnetic) incremental encoders, but can be used in any type of rotational speed sensor (including inductive, etc.), since there is usually always a delay between detection and evaluation (for example due to the control device used).

[0004] According to a preferred development of the invention, it is proposed that the assumed valid time point of the first signal value is determined from the difference between the first evaluation time point and the information age. By determining the valid time point, it is advantageously ensured that the correct time is associated with the first signal value when the signal value is used as a grid point for extrapolation or interpolation.

[0005] It is particularly preferred that the first evaluation time point is selected as the reference time point. The advantage obtained by selecting the evaluation time point is that the extrapolation is performed at an already known time point, thereby associating the correct signal value with the evaluation time point.

[0006] According to a preferred development of the invention, it is provided that the information age is estimated based on at least one characteristic variable of the signal processing of the rotational speed sensor, in particular the runtime of the integral signal filter and / or the signal transmission path. Taking such a characteristic variable into account advantageously ensures a further improvement in the accuracy when estimating the information age.

[0007] It is particularly preferred that the information age is estimated based on at least one speed-dependent characteristic variable of the sampling of the rotational speed sensor, in particular angularly equidistant sampling, in particular half of the time difference between the first signal value and the second signal value. Taking into account such a characteristic variable advantageously ensures a further improvement in the accuracy of the estimation of the information age. In particular, the rotational speed sensor sampling characteristic variable is taken into account in addition to the above-mentioned characteristic variables of the signal processing.

[0008] According to a preferred development of the invention, it is proposed that the information age is estimated based on at least one characteristic variable of the control device for evaluating the signal value, in particular a measured duration determined based on an expected value of a uniform distribution and / or determined by means of an interval nesting in an equidistant evaluation grid, which duration is used by the control device to convert angular equidistant sampling into temporal equidistant evaluation. Taking into account such a characteristic variable advantageously ensures a further improvement in the accuracy when estimating the information age. In particular, in addition to the above-mentioned characteristic variables of the signal processing and / or sampling, at least one characteristic variable of the control device for evaluating the signal value is also taken into account.

[0009] It is particularly preferred that the rotational speed sensor is designed as an incremental encoder, in particular a magnetic incremental encoder, and that the time intervals between temporally adjacent measuring events of the incremental encoder or values ​​derived therefrom are used as signal values. The advantages of the method according to the invention are particularly evident in this embodiment of the rotational speed sensor. In particular, the incremental encoder has a plurality of sensor elements and a magnetic incremental encoder wheel. Preferably, a predetermined signal value of the signal curve, in particular a zero crossing of the signal curve, is taken into account as a measuring event, and / or a measuring event is determined based on the signal curves of at least two sensor elements, in particular based on the difference signal curves of at least two of the sensor elements.

[0010] According to a preferred development of the invention, it is provided that the wheel peripheral speed is determined as the displacement variable. The method according to the invention makes it particularly advantageous to determine the wheel peripheral speed of a wheel, for example of a motor vehicle, simply and precisely.

[0011] It is particularly preferred that the extrapolation or interpolation is performed linearly or by means of an nth-order polynomial, where n is an integer and is greater than 1, in particular by means of a Savitzky-Golay filter. This advantageously further improves the accuracy when performing the method according to the invention.

[0012] The computer program product according to the invention for execution on a computer device is characterized in that it executes the method according to the invention during normal use. This results in the advantages already mentioned.

[0013] The data carrier according to the invention is characterized in that a computer program product according to the invention is stored thereon.

[0014] The computer device according to the invention is characterized in that it is specially designed to execute the method according to the invention or to execute the computer program product according to the invention. This also results in the above-mentioned advantages. Preferably, the computer device is associated with a motor vehicle, in particular a control device arranged in the motor vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Other preferred features and feature combinations can be derived from the above description. The present invention is explained in more detail below with reference to the accompanying drawings. Among them:

[0016] Figure 1 The speed sensor is shown.

[0017] Figure 2 shows a first angular acceleration-time diagram,

[0018] Figure 3 A method for determining a rotational speed and / or a displacement variable derived from the rotational speed is shown.

[0019] Figure 4 A second angular acceleration-time graph is shown. DETAILED DESCRIPTION

[0020] Figure 1 A known rotational speed sensor 1 is shown, which is used, for example, to determine the wheel peripheral speed ω·r, which can then, as mentioned above, be used as an input variable for ABS / ESP control. The rotational speed sensor 1 is designed to detect (measurement) events, such as magnetic flux changes, of an encoder or incremental encoder wheel 2 rotating with the wheel and transmit them to a computer device 3, here a control unit. The computer device 3 is in turn designed to determine the time interval Δt(i) between the i-th event and the preceding i-1-th event.

[0021] From the number of events per revolution N, the predefined rolling radius r and the time interval Δt of the measurements, the peripheral wheel speed is estimated by the computer device 3 as ω m ·r=(2π·r) / N·1 / Δt. In principle, this is the average velocity over the arc length s=(2π·r) / N which is travelled between the events.

[0022] For example, assuming constant wheel acceleration, the determined wheel peripheral speed corresponds to the arithmetic mean of the actual wheel peripheral speeds occurring at the two events i and i-1 used. The valid time point of the speed information determined in this way is just between the two events, but its usable time point is at the detection time point of event i.

[0023] By the typical time-synchronous evaluation of the events in the control unit, for example in equidistant 5 ms time frames, a further delay is usually formed between the available time point and the evaluation time point. The delay can be represented (at least at low speeds assuming that at most one event is evaluated at a time) as a random variable uniformly distributed in the corresponding interval, for example in the interval [0 ... 5 ms]. At higher speeds, multiple events occur per evaluation grid, whereby the uniformly distributed intervals are preferably shortened to a small fraction for the respective last event of interest in the evaluation grid.

[0024] Here, the time difference between the evaluation time point and the valid time point is defined as the information age of the corresponding signal value of the event. Therefore, due to the angular equidistant or distance equidistant sampling by the speed sensor, the information age of the wheel peripheral speed has the above-mentioned two speed-dependent parts, namely a deterministic part and a random part.

[0025] This basic relationship is Figure 2 To this end, Figure 2 In the first angular acceleration-time diagram, the corresponding angular velocity ω is plotted over time t for events i and i-1. Thus, for event i-1, there is an angular velocity ω(i-1) at time point t(i-1), and for event i that follows at a time interval Δt(i), there is an angular velocity ω(i) at time point t(i).

[0026] Although only the angular velocity ω associated with event i is measured (and evaluated) m , but the angular velocity ω m In fact, at the effective time point t before the time point t(i) g is valid and is evaluated at time point t after time point t(i) a The difference between the validity time point and the evaluation time point Δ(t a -t g ) corresponds to the information age T I .

[0027] Refer to the following Figure 3 An advantageous method for determining a rotational speed and / or a displacement variable derived from the rotational speed by means of a rotational speed sensor, in particular by means of the above-described rotational speed sensor 1, is described. To this end, Figure 3The method is illustrated by means of a flow chart. In particular, the method ensures that systematic errors in determining the corresponding variables are at least significantly reduced by taking into account the age of the information as described above. The method is executed in particular by means of a computer device 3 .

[0028] In step S1, the method begins by detecting a first signal value of a rotational speed sensor and evaluating it at a first evaluation time. The rotational speed sensor is preferably designed as an incremental encoder, in particular a magnetic incremental encoder, wherein the time intervals between temporally adjacent measuring events of the incremental encoder or values ​​derived therefrom are used as signal values. In particular, as described above, the wheel peripheral speed, preferably also the wheel center speed and the wheel slip resulting therefrom are determined as displacement variables.

[0029] In any case, the corresponding signal value has a certain information age as described above. The information age is estimated in step S2. In particular, the corresponding information age of the signal value of another movement variable (eg wheel center speed) is additionally estimated.

[0030] Here, the wheel center speed, or in other words the speed of the wheel center relative to the ground, v Rad undergoes a different signal processing than the wheel circumferential velocity ω·r. Therefore, the wheel midpoint velocity is usually obtained by dividing the vehicle speed v Fahrzeug It is determined from the geometrical projection of the vehicle's center of gravity onto the corresponding wheel center and along the wheel's rolling direction. For the wheel center speed, validity and evaluation are usually assumed to be performed simultaneously, in particular any information age that may exist is not speed-dependent.

[0031] Accordingly, it is preferred that, in particular, when calculating the wheel differential speed ω·rv Rad Or before wheel slip, estimate the measured wheel peripheral speed ω m The information age of r and the wheel midpoint velocity v Rad The information age of the speed signal is taken into account in such a way that the consistency of the speed signal used is established by subsequent extrapolation or interpolation. Thus, in particular in the case of low wheel speeds and current wheel accelerations, systematic errors in the speed difference / wheel slip signal derived therefrom are avoided.

[0032] Starting from the sensor detection of the respective physical variable, each of the two signals under consideration passes through a signal processing channel until the time of the joint algorithmic evaluation and thus achieves a different information age.

[0033] At the wheel peripheral speed ω mIn the case of r, the information age in the specific example consists of a speed-independent deterministic part caused by the signal filter, a speed-dependent deterministic part caused by the angularly equidistant sampling of the incremental encoder, and a speed-dependent random part, which is formed in the control unit when converting the sampling rate of the angularly equidistant or distance-equidistant incremental encoder into a temporally equidistant evaluation.

[0034] Speed ​​v at the wheel midpoint Rad In the case of the information age being essentially determined by the group runtime in the signal filter of the inertial sensor system, the calculation of the vehicle speed is based on the wheel center speed and is therefore assumed to be deterministic, possibly already corrected, and in particular speed-independent.

[0035] The information age is therefore preferably estimated from at least one characteristic variable of the signal processing of the rotational speed sensor or of the corresponding sensor system used, in particular from the runtime of an integrated signal filter and / or the signal transmission path. The speed-independent deterministic part of the information age is derived substantially directly from the specification of the upstream signal filter and is, for example, a few milliseconds, typically 10 milliseconds.

[0036] Alternatively or additionally, the information age is preferably estimated based on sampling, in particular angularly equidistant sampling, of the rotational rate sensor, in particular based on half the time difference between a first signal value and a second signal value temporally preceding the first signal value.

[0037] To estimate the speed-dependent delay between the validity and availability of the wheel peripheral speed, it is preferably assumed that the wheel speed between events is constant; then, for example, half of the measured time difference Δt / 2 between events is assumed as an estimate for this portion of the information age.

[0038] Alternatively or additionally, the information age is preferably estimated based on at least one characteristic variable of a control device for evaluating signal values, in particular a measured duration determined based on an expected value from a uniform distribution and / or determined by means of a nesting of intervals in an equidistant evaluation grid, which duration is used by the control device to convert angular equidistant sampling into temporal equidistant evaluation.

[0039] Preferably, a speed-dependent random delay between the available time and the evaluation time is measured. If the control unit determines the event time i by means of a continuous counter acting as a clock, this delay can also be read out during the evaluation, as is usually the case in ABS / ESP control units.

[0040] If it is not provided that the clock is read out by the counter at the evaluation time, the uniformly distributed expected value is preferably used as the delay. Then alternatively, under the assumption of an equidistant evaluation grid, the deviation of the evaluation is always further limited and estimated by means of randomly distributed events i and the interval nesting method.

[0041] If the information age is determined accordingly, then in step S3, one (or each) corrected third signal value at a selected reference time point (and taking into account the signal value of its other movement variable) is extrapolated or interpolated based on the estimated information age of the first signal value of the rotational speed sensor and / or the corresponding signal value of another movement variable and a second signal value that temporally precedes the first signal value.

[0042] In this respect, for each signal used, its last valid value is projected onto an estimated value at the evaluation time point by extrapolation / interpolation of its preceding signal curve, taking into account its information age, which is then used for the calculation.

[0043] When the method is first carried out, the temporally previous signal values ​​are temporally uncorrected values, since the information age has not yet been taken into account, so that the previous signal curve required for the extrapolation exists only from the second signal value onwards at the earliest, ie at the third event at the earliest.

[0044] In particular, the first evaluation time point is selected as the corresponding reference time point. Preferably, the extrapolation / interpolation is performed linearly or by means of a polynomial of order n, where n is an integer and is greater than 1, in particular by means of a Savitzky-Golay filter. In the first case, this is performed in terms of the computational complexity in the control device in the form of a linear extrapolation of the last valid speed, assuming a constant acceleration (e.g. the last estimated), or in the second case by means of a higher-order polynomial extrapolation.

[0045] exist Figure 4 An example of a corresponding linear extrapolation is shown in FIG. 1 . In the second angular acceleration-time diagram, the corresponding angular velocity ω is plotted over time t for events i and i-1. As already mentioned with reference to Figure 2 As described, at the evaluation time point t a The angular velocity ω associated with event i m is evaluated as the first signal value. Now, the angular velocity is considered in its information age T I In the case of being projected to the corresponding reference time point, here is the effective time point t g .

[0046] Angular velocity ω b The second signal value before this point is now connected to the angular velocity ω by a straight line mThe time-corrected first signal value of is connected, and the straight line is extended to the evaluation time point so that the corresponding evaluation time point t a There is an angular velocity ω p The corrected third signal value.

[0047] In an alternative embodiment, the projection of the used signals can also be carried out at a time different from the evaluation time, preferably at an earlier time, for example a speed-dependent effective time of the wheel peripheral speed. This advantageously shortens the extrapolation duration or an interpolation can be carried out instead of an extrapolation, so that the accuracy of the derived variable can then be increased.

[0048] Finally, in step S4 , the rotational speed and / or the displacement variable is determined from the corresponding third signal value, for example, the wheel slip is determined from the wheel peripheral speed and the wheel center speed as described above. The method is preferably performed continuously.

Claims

1. A method for determining a rotational speed and / or a displacement variable derived from the rotational speed by means of a rotational speed sensor (1), According to the rotational speed sensor (1), at a first evaluation time point (t a ) evaluates the first signal value (ω m ) of the estimated information age (T I ) and the first signal value (ω m ) before the second signal value (ω b ) to extrapolate or interpolate the corrected third signal value (ω) at the selected reference time point p ),and According to the third signal value (ω p ) determine the rotational speed and / or the movement variable.

2. The method according to claim 1, characterized in that From the first evaluation time point (t a ) and the information age (T I ) to determine the first signal value (ω m ) is the assumed effective time point (t g ).

3. The method according to any one of the preceding claims, characterized in that The first evaluation time point (t a ) is selected as the reference time point.

4. The method according to any one of the preceding claims, characterized in that The information age (T) is estimated based on at least one characteristic variable of the signal processing of the rotational speed sensor (1), in particular the runtime of an integral signal filter and / or a signal transmission path. I ).

5. The method according to any one of the preceding claims, characterized in that The information age (T) is estimated based on at least one characteristic variable related to the rotation speed, in particular half of the time difference between the first signal value and the second signal value, sampled by the rotation speed sensor (1), in particular angularly equidistant sampling. I ).

6. The method according to any one of the preceding claims, characterized in that The information age (T) is estimated based on at least one characteristic variable of a control device (3) for evaluating signal values, in particular a measured duration determined based on an expected value of a uniform distribution and / or determined by means of a nesting of intervals in an equidistant evaluation grid. I ), the duration being used to convert angularly equidistant sampling into temporally equidistant evaluation by the control device (3).

7. The method according to any one of the preceding claims, characterized in that The rotational speed sensor (1) is designed as an incremental encoder, in particular a magnetic incremental encoder, and the time intervals between temporally adjacent measuring events of the incremental encoder or values ​​derived therefrom are used as signal values.

8. The method according to any one of the preceding claims, characterized in that The wheel peripheral speed is determined as the displacement variable.

9. The method according to any one of the preceding claims, characterized in that The extrapolation or interpolation is performed linearly or by means of a polynomial of order n, where n is an integer and is greater than 1, in particular by means of a Savitzky-Golay filter.

10. A computer program product for execution on a computer device, characterized in that The computer program product performs, in normal use, a method according to any one of the preceding claims.

11. A data carrier having a computer program product according to claim 10.

12. A computer device (3), in particular a control device for a motor vehicle, characterized in that The computer device (3) is specially designed to execute the computer program product according to claim 10.

Images