Method for further-processing a cyclically provided sensor signal

Abstract

The invention relates to a method for further-processing a cyclically provided sensor signal (S), wherein two different computing modes (FM, HRM) are performed in parallel to the compute of gradients (∆S / ∆t) of the cyclically provided sensor signal (S), wherein a first computing mode (FM) determines first gradient values (∆S / ∆t1) of the cyclically provided sensor signal (S) with a first time interval (∆t1), and a second computing mode (HRM) determines second gradient values (∆S / ∆t2) of the cyclically provided sensor signal (S) with a higher resolution and a higher accuracy with a second time interval (∆t2), which is longer than the first time interval (∆t1), wherein a signal difference (∆S) between two successively provided sensor signals (S) or between two items of signal information successively obtained from the provided sensor signals (S) is computed and compared with a threshold value (SW), wherein the comparison result is taken as a basis for selecting one of the two computing modes (FM, HRM), and either the first gradient values (∆S / ∆t1) determined by the first computing mode (FM) or the second gradient values (∆S / ∆t2) determined by the second computing mode (HRM) are output as gradient values (∆S / ∆t) of the cyclically provided sensor signal (S), and to a sensor arrangement (1) that is designed to carry out such a method.

Description

[0001] Description

[0002] title

[0003] Method for further processing a cyclically provided sensor signal

[0004] The invention relates to a method for further processing a cyclically provided sensor signal. The present invention also relates to a sensor arrangement for implementing such a method.

[0005] Vehicles are known from the state of the art in which advanced driver assistance systems (ADAS) and / or vehicle functions, such as ESP (Electronic Stability Program) or ESC (Electronic Stability Control), are implemented. Such driver assistance systems and / or vehicle functions use sensors to measure the vehicle's yaw movements around a yaw axis (vertical axis) and to measure vehicle stability around a roll axis (longitudinal axis) and a pitch axis (lateral axis). For example, an inertial sensor unit can be used to record the translation and rotation of the vehicle along or around the vertical axis, longitudinal axis, and lateral axis. In addition to acceleration and yaw rate signals from such inertial measurements, information about signal gradients of these signals can be used for specific vehicle functions.The gradient information of jerk and angular acceleration from inertial signals, which describe the translation and rotation of a vehicle, is the temporal change in acceleration a (da / dt = jerk) and angular velocity Q (d / dt = angular acceleration). According to Newton's law F = m*a, a change in acceleration a (da / dt) leads to a change in force dF / dt = m*(da / dt) and subsequently in torque.

[0006] The accuracy and resolution of the determined gradient signal are determined by the quantization (resolution) of the input signals (a, O) and the temporal separation of at least two observation points used to determine the gradient signal. Furthermore, stochastic signal variations (noise) in the input signals further limit the accuracy and thus increase the uncertainty in the determined gradient values. Noise suppression, achieving high gradient resolution, and correct gradient calculation for highly dynamic signals are contradictory requirements. Therefore, there is a trade-off between high dynamic range and high accuracy, which requires high resolution. For functional reasons and simplicity of implementation, it may be necessary to represent both states in a single signal.

[0007] Disclosure of the invention

[0008] The method for further processing a cyclically provided sensor signal with the features of independent patent claim 1 and the sensor arrangement with the features of independent patent claim 12 each have the advantage that a synchronous calculation of two independent gradient values ​​with parameter-based switching between the gradient values ​​enables a dynamic adaptation of the calculation mode for the gradients.

[0009] Embodiments of the present invention provide a method for further processing a cyclically provided sensor signal. Two different calculation modes are executed in parallel to calculate gradients of the cyclically provided sensor signal. A first calculation mode determines first gradient values ​​of the cyclically provided sensor signal with a first time interval. A second calculation mode determines second gradient values ​​of the cyclically provided sensor signal with a higher resolution and higher accuracy with a second time interval that is longer than the first time interval. A signal difference between two consecutively provided sensor signals or between two pieces of signal information successively obtained from the provided sensor signals is calculated and compared with a threshold value.Depending on the comparison result, one of the two calculation modes is selected, and either the first gradient values ​​determined by the first calculation mode or the second gradient values ​​determined by the second calculation mode are output as gradient values ​​of the cyclically provided sensor signal.

[0010] In addition, a sensor arrangement is proposed with at least one sensor which is designed to detect at least one physical quantity and to provide a corresponding cyclic sensor signal, and an evaluation and control unit which is designed to carry out such a method for further processing a cyclically provided sensor signal.

[0011] The first calculation mode of the first gradient values ​​can be used for rapidly changing signals. Here, the focus can be placed on the measurement range and minimized time delay. Furthermore, the first time interval can be defined so that it can respond quickly to a highly dynamic stimulus and the determined gradient signal can meet the required accuracy even in transient conditions. Furthermore, the first time interval can be defined so that it can support an update rate of communication times during sensor data transmission. Furthermore, the gradient values ​​can remain within the gradient signal range even with significant changes in the signal differences. The second calculation mode of the second gradient values ​​can be used for slowly changing signals. Here, the focus can be placed on high accuracy and high resolution.In addition, the second time interval can be defined for a given signal resolution to achieve the desired resolution in the gradient value. By appropriately defining the second time interval, interfering noise in the second gradient values ​​can be eliminated to achieve the desired accuracy.

[0012] Here, switching criteria between the first calculation module and the second calculation mode can be defined based on the best possible accuracy for a given dynamic range with an acceptable noise error in the first calculation mode and based on high accuracy and higher resolution in the second calculation module.

[0013] To react to dynamic changes in the vehicle's condition, which can be caused, for example, by crosswinds at higher speeds, vibrations from uneven road surfaces, or bumps, the first gradient values ​​of the first calculation mode can be used. For example, the first gradient values ​​can be used for chassis control, braking intervention, and / or steering control to compensate for external events and keep the vehicle stable by quickly detecting the condition. In chassis control with wheel-based acceleration measurement or detected pitching or rolling motion, the first gradient values ​​of the first calculation mode can be used to control the damper speed or damper stiffness to stabilize the vehicle.Similarly, during braking, when the linear acceleration changes, this can lead to a change in momentum and thus to a pitching of the vehicle's front end. On the other hand, when negotiating a curve on a highway, the second gradient values ​​of the second calculation mode can be used to capture the actual movement with high accuracy, which exhibits a quasi-stationary rate of change of the signal difference between two consecutively provided sensor signals or between two consecutive signal information items obtained from the provided sensor signals, in order to establish a stable vehicle position.

[0014] In this case, the evaluation and control unit can be understood as an electrical assembly that receives and further processes or evaluates detected sensor signals. The evaluation and control unit can have at least one interface, which can be implemented in hardware and / or software. In a hardware implementation, the interfaces can, for example, be part of a so-called system ASIC, which contains a wide variety of functions of the evaluation and control unit. However, it is also possible for the interfaces to be separate integrated circuits or to consist at least partially of discrete components. In a software implementation, the interfaces can be software modules that are present, for example, on a microcontroller alongside other software modules.Also advantageous is a computer program product with program code stored on a machine-readable medium such as a semiconductor memory, a hard disk memory or an optical memory and used to carry out further processing and evaluation when the program is executed by the evaluation and control unit.

[0015] In this case, the at least one sensor can be understood as a structural unit comprising at least one sensor element that directly or indirectly detects a physical quantity or a change in a physical quantity and preferably converts it into an electrical sensor signal, which it transmits cyclically to the evaluation and control unit at a predetermined transmission rate and using a predetermined transmission protocol. For example, an inertial sensor unit can be used to detect the translation and rotation of the vehicle along or around the vertical axis, longitudinal axis, and transverse axis and output corresponding sensor signals.

[0016] The measures and further developments listed in the dependent claims enable advantageous improvements of the method specified in independent patent claim 1 for further processing a cyclically provided sensor signal.

[0017] It is particularly advantageous that the first gradient values ​​determined by the first calculation mode can be output as gradient values ​​of the cyclically provided sensor signal if the signal difference is greater than or equal to the threshold value. This allows for a sufficiently fast response to a highly dynamic stimulus or rapidly changing signals.

[0018] In an advantageous embodiment of the method, the second gradient values ​​determined by the second calculation mode can be output as gradient values ​​of the cyclically provided sensor signal if the signal difference is smaller than the threshold value. This allows for high resolution and high accuracy for slowly changing signals.

[0019] In a further advantageous embodiment of the method, a counter reading of a counter function can be incremented if the signal difference is greater than the threshold value or equal to the first threshold value. Alternatively, the counter reading of the counter function can be reduced if the signal difference is smaller than the threshold value. Depending on the counter reading and the currently selected one of the two calculation modes, this can be retained or switched to the other calculation mode. For example, in the second calculation mode and an output of the second gradient values, it is possible to switch to the first calculation mode and output the first gradient values ​​if the counter reading is incremented. In the first calculation mode and an output of the first gradient values, it is possible to switch to the second calculation mode and output the second gradient values ​​if the counter reading is reduced.To prevent constant switching between the two calculation modes and the output of the first gradient values ​​and the output of the second gradient values, a hysteresis can be provided. For example, a first number of upcounting operations can be specified before switching from the first calculation mode and the output of the first gradient values ​​to the second calculation mode and the output of the second gradient values. Similarly, a second number of downcounting operations can be specified before switching from the second calculation mode and the output of the second gradient values ​​to the first calculation mode and the output of the first gradient values.

[0020] In a particularly advantageous embodiment of the method, in the second calculation mode, it is possible to switch to the first calculation mode if the counter reading of the counter function exceeds a predetermined first counter value. At the same time, the counter reading can be increased to a predetermined second counter value. For example, the first counter value can be specified in the range from 2 to 8, which means that, depending on the specified first counter value, between 2 and 8 counting-up operations are required to switch from the first calculation mode to the second calculation mode. The second counter value can, for example, correspond to a maximum counter reading of the counter function. In addition, in the first calculation mode, it is possible to switch to the second calculation mode if the counter reading of the counter function falls below a predetermined third counter value.The second counter value and the third counter value can be specified in such a way that a switch from the first calculation mode to the second calculation mode can only occur after at least two second time intervals have elapsed. This can prevent the dynamic first gradient values ​​of the first calculation mode from negatively affecting the high accuracy of the second gradient values ​​of the second calculation mode.

[0021] In a further advantageous embodiment of the method, the threshold value can be specified as a function of a permissible error for the gradient values ​​of the cyclically provided sensor signal. If the first calculation mode is used for slowly changing signals, noise determines the first gradient values ​​without correct averaging. In the case of random, normally distributed noise, provided values ​​of the sensor signal can be statistically dependent on the actual bandwidth of the sensor data in the shorter first time interval. Therefore, averaging of the noise does not work, and large errors can occur in the determined first gradient values. The error can be calculated as a function of the boundary conditions of the sensor arrangement, such as the resolution and value range of the data word used to provide the sensor signal.To distinguish a true gradient from an error caused by noise, the determined gradient value or signal difference should have a minimum magnitude.

[0022] In a further advantageous embodiment of the method, the first time interval can correspond to a provision period of the sensor signals. This allows the first time interval to support an update rate of communication times during sensor data transmission. Furthermore, a ratio of the second time interval to the first time interval can be specified in the range of 5 to 20.

[0023] Embodiments of the invention are illustrated in the drawings and explained in more detail in the following description. In the drawings, like reference numerals designate components or elements that perform the same or similar functions.

[0024] Brief description of the drawings Fig. 1 shows a schematic flow diagram of an embodiment of a method according to the invention for further processing a cyclically provided sensor signal.

[0025] Fig. 2 shows a schematic block diagram of an embodiment of a sensor arrangement according to the invention.

[0026] Fig. 3 shows an error difference signal characteristic diagram with error characteristics of two calculation modes of the inventive method from Fig. 1.

[0027] Fig. 4 shows a counter reading-time characteristic diagram of a counter function of the inventive method from Fig. 1.

[0028] Embodiments of the invention

[0029] 1 and 2, the illustrated embodiment of a method 100 according to the invention for further processing cyclically provided sensor signal S comprises a step S100 in which the sensor signal S shown in Fig. 2 is provided. Here, two different calculation modes FM, HRM shown in Fig. 2 are executed in parallel to calculate gradients AS / At of the cyclically provided sensor signal S. For this purpose, in a step S110, a first calculation mode FM determines first gradient values ​​AS / At1 of the cyclically provided sensor signal S with a first time interval At1. In a step S120 executed in parallel, a second calculation mode HRM determines second gradient values ​​AS / At2 of the cyclically provided sensor signal S with a higher resolution and a higher accuracy with a second time interval At2 which is longer than the first time interval At1.In a step S130, a signal difference AS between two consecutively provided sensor signals S or between two consecutively obtained signal information items from the provided sensor signals S is calculated and compared with a threshold value SW. Depending on the comparison result, one of the two calculation modes FM, HRM is selected in step S140, and either the first gradient values ​​AS / At1 determined by the first calculation mode FM or the second gradient values ​​AS / At2 determined by the second calculation mode HRM are output as gradient values ​​AS / At of the cyclically provided sensor signal S.

[0030] In the illustrated embodiment of method 100, in step S140, the first gradient values ​​AS / At1 determined by the first calculation mode FM are output as gradient values ​​AS / At of the cyclically provided sensor signal S if the signal difference AS is greater than the threshold value SW or equal to the threshold value SW. If the signal difference AS is less than the threshold value SW, the second gradient values ​​AS / At2 determined by the second calculation mode HRM are output as gradient values ​​AS / At of the cyclically provided sensor signal S.

[0031] In the illustrated embodiment, the threshold value SW is specified as a function of a permissible error ER for the gradient values ​​AS / At of the cyclically provided sensor signal S. Here, the first time interval At1 corresponds to a provision period of the sensor signals S. This means that the first time interval At1 corresponds to a predetermined transmission rate or update rate with which the sensor signal S is cyclically provided. A ratio of the second time interval At2 to the first time interval At1 is specified in the range from 5 to 20.

[0032] As can be further seen from Fig. 2, the illustrated embodiment of a sensor arrangement 1 according to the invention comprises at least one sensor 3, which is designed to detect at least one physical variable and to provide a corresponding cyclic sensor signal S, and an evaluation and control unit 10, which is designed to carry out the inventive method 100 for further processing a cyclically provided sensor signal S. As can be further seen from Fig. 2, the evaluation and control unit 10 comprises a calculation block 12, which executes the first calculation mode FM and the second calculation mode HRM. In addition, the evaluation and control unit 10 comprises a comparison block 14, which in the illustrated embodiment calculates a signal difference AS between two successively provided sensor signals S and compares it with a threshold value SW.Depending on the comparison result, the comparison block 14 controls a selection block 15, which selects one of the two calculation modes FM, HRM, and outputs either the first gradient values ​​AS / At1 determined by the first calculation mode FM or the second gradient values ​​AS / At2 determined by the second calculation mode HRM as gradient values ​​AS / At of the cyclically provided sensor signal S to a buffer memory 16. Higher-level vehicle functions can read the gradient values ​​AS / At of the cyclically provided sensor signal S from the buffer memory 16. In this case, the buffer memory 16 can be considered optional. This means that the gradient values ​​AS / At of the cyclically provided sensor signal S can also be output directly to the higher-level vehicle functions.

[0033] In an alternative embodiment of the sensor arrangement 1 (not shown), the comparison block 14 calculates a signal difference AS between two successive signal information items obtained from the provided sensor signals S and compares this with the threshold value SW.

[0034] As can also be seen from Fig. 3, a first threshold value SW1 corresponds to a first error ER1 which, for example, corresponds to a 40% deviation, caused by noise, of the first gradient values ​​AS / At1 determined by the first calculation mode FM. A second threshold value SW2 corresponds to a second error ER2 which, for example, corresponds to a 5% deviation, caused by noise, of the first gradient values ​​AS / At1 determined by the first calculation mode FM. Due to the strong dependence on the hardware used, the errors ER1, ER2 are only given as examples. Depending on the desired permissible error ER, the threshold value SW in the illustrated embodiment can be determined in the threshold range SWB which is predetermined by the two threshold values ​​SW1, SW2.

[0035] As can also be seen from Fig. 2, the comparison block 14 for controlling the selection block 15 in the illustrated embodiment comprises a counter function ZF. The comparison block 14 with the counter function ZF is designed to control the selection block 15 for selecting one of the two operating modes FM, HRM. As can also be seen from Fig. 4, in the illustrated embodiment of the method 100, a counter reading ZS of the counter function ZF is increased if the signal difference AS is greater than the threshold value SW or equal to the threshold value SW. Alternatively, the counter reading ZS of the counter function ZF is reduced if the signal difference AS is smaller than the threshold value SW. Depending on the counter reading ZS and on the currently selected one of the two calculation modes FM, HRM, this is retained or the system switches to the other calculation mode HRM, FM.

[0036] As can be further seen from Fig. 4, at a time "0", the second calculation mode HRM is selected, so that the second gradient values ​​AS / At2 are output to the buffer memory 16 as gradient values ​​AS / At of the cyclically provided sensor signal S. After the first time interval At1 has elapsed, the comparison block 14 determines that the calculated signal difference AS exceeds the threshold value SW. This increases the counter reading ZS of the counter function ZF. As can be further seen from Fig. 4, the system switches from the second calculation mode HRM to the first calculation mode FM when the counter reading ZS of the counter function ZF exceeds a predetermined first counter value ZW1. This means that the comparison block 14 controls the selection block 15 so that it selects the first calculation mode FM and outputs the first gradient values ​​AS / At1 as gradient values ​​AS / At of the cyclically provided sensor signal S to the buffer memory 16.At the same time, the counter function ZF increases the counter reading ZS to a predefined second counter value ZW2. In the first calculation mode FM, the system switches to the second calculation mode HRM when the counter reading ZS of the counter function ZF falls below a predefined third counter value ZW3. This means that the comparison block 14 controls the selection block 15 so that it selects the second calculation mode HRM again and outputs the second gradient values ​​AS / At2 as gradient values ​​AS / At of the cyclically provided sensor signal S to the buffer memory 16.

[0037] As can be further seen from Fig. 4, the second counter value ZW2 and the third counter value ZW3 are specified in such a way that a change from the first calculation mode FM to the second calculation mode HRM can only take place after at least two second time intervals At2 have elapsed.

Claims

Claims 1. A method (100) for further processing a cyclically provided sensor signal (S), wherein two different calculation modes (FM, HRM) are executed in parallel to calculate gradients (AS / At) of the cyclically provided sensor signal (S), wherein a first calculation mode (FM) determines first gradient values (AS / At1) of the cyclically provided sensor signal (S) with a first time interval (At1), and a second calculation mode (HRM) determines second gradient values (AS / At2) of the cyclically provided sensor signal (S) with a higher resolution and a higher accuracy with a second time interval (At2) which is longer than the first time interval (At1), wherein a signal difference (AS) between two successively provided sensor signals (S) or between two successively obtained signal information items from the provided sensor signals (S) is calculated and compared with a threshold value (SW),wherein, depending on the comparison result, one of the two calculation modes (FM, HRM) is selected, and either the first gradient values (AS / At1) determined by the first calculation mode (FM) or the second gradient values (AS / At2) determined by the second calculation mode (HRM) are output as gradient values (AS / At) of the cyclically provided sensor signal (S).

2. Method (100) according to claim 1, characterized in that the first gradient values (AS / At1) determined by the first calculation mode (FM) are output as gradient values (AS / At) of the cyclically provided sensor signal (S) if the signal difference (AS) is greater than the threshold value (SW) or equal to the threshold value (SW).

3. Method (100) according to claim 1 or 2, characterized in that the second gradient values (AS / At2) determined by the second calculation mode (HRM) are output as gradient values (AS / At) of the cyclically provided sensor signal (S) if the signal difference (AS) is smaller than the threshold value (SW).

4. Method (100) according to one of claims 1 to 3, characterized in that a counter reading (ZS) of a counter function (ZF) is increased if the signal difference (AS) is greater than the threshold value (SW) or equal to the first threshold value (SW), or the counter reading (ZS) of the counter function (ZF) is reduced if the signal difference (AS) is smaller than the threshold value (SW).

5. Method (100) according to claim 4, characterized in that depending on the counter reading (ZS) and on the currently selected one of the two calculation modes (FM, HRM), this is maintained or a change is made to the other calculation mode (HRM, FM).

6. Method (100) according to claim 5, characterized in that in the second calculation mode (HRM) a change is made to the first calculation mode (FM) when the counter reading (ZS) of the counter function (ZF) exceeds a predetermined first counter value (ZW1), wherein the counter reading (ZS) is simultaneously increased to a predetermined second counter value (ZW2).

7. Method (100) according to claim 5 or 6, characterized in that in the first calculation mode (FM) a change is made to the second calculation mode (HRM) when the counter reading (ZS) of the counter function (ZF) falls below a predetermined third counter value (ZW3).

8. Method (100) according to claim 6 and 7, characterized in that the second counter value (ZW2) and the third counter value (ZW3) are predetermined such that a change from the first calculation mode (FM) to the second calculation mode (HRM) can only take place after the expiration of at least two second time intervals (At2).

9. Method (100) according to one of claims 1 to 8, characterized in that the threshold value (SW) is predetermined as a function of a permissible error (ER) for the gradient values (AS / At) of the cyclically provided sensor signal (S).

10. Method (100) according to one of claims 1 to 9, characterized in that the first time interval (At1) corresponds to a provision period of the sensor signals (S).

11. Method (100) according to one of claims 1 to 10, characterized in that a ratio of the second time interval (At2) to the first time interval (At1) is specified in the range from 5 to 20.

12. Sensor arrangement (1) with at least one sensor (3) which is designed to detect at least one physical quantity and to provide a corresponding cyclic sensor signal (S), and an evaluation and control unit (10) which is designed to carry out the method (100) for further processing a cyclically provided sensor signal (S) according to one of claims 1 to 11.

Images