Method for detecting resting state of vehicle

Abstract

To provide a method for detecting the resting state of a vehicle.SOLUTION: The method of the present invention includes: a step (a) of detecting the resting state of a vehicle on the basis of at least one sensor; a step (b) of executing an inspection routine to inspect the resting state of the vehicle when the resting state is identified in the step (a), the routine using at least one signal of at least one rotation rate sensor as an input value; and a step (c) of destroying the resting state of the vehicle detected in the step (a) when the absence of the resting state is identified on the basis of at least one rotation rate sensor in the step (b).SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] Prior Art The present invention relates to a method for detecting a stationary state of a vehicle. Furthermore, a computer program, a machine-readable storage medium, and a vehicle control device are also presented. The present invention is particularly applicable in a GNSS and INS-based positioning system for autonomous driving or semi-autonomous driving.

Background Art

[0002] For high-precision vehicle positioning by an inertial sensor system, high-quality sensor signals are required. The offset of an inertial sensor can be compensated during a stationary state. Here, it is important that no high dynamics of the signal occur during the stationary state, because such high dynamics have a negative impact on the compensation of the offset. In particular, movements induced by external infrastructure, for example, by a vehicle transporter, cause significant errors during the calibration of the offset if they are not identified by the detection of the stationary state. In particular, vehicle turntables are difficult to detect and are indistinguishable in current positioning systems.

Summary of the Invention

Means for Solving the Problems

[0003] Disclosure of the Invention Based on the above, means for detecting a stationary state of a vehicle will be described. First, an assumption of a stationary state is made using at least one sensor. Subsequently, the assumption of the stationary state is examined or confirmed using at least one other sensor. For this purpose, simply an existing vehicle sensor system is applied. According to the method described here, it is possible to detect a stationary state in a critical scenario, in particular, for example, a vehicle stationary on a rotating vehicle turntable.

[0004] Based on the above, a method for detecting a stationary state of a vehicle will be described. The method comprises at least the following steps, namely, a) A step of detecting the stationary state of the vehicle based on at least one sensor, b) If a stationary state is identified in step a), the step of performing an inspection routine for inspecting the stationary state of the vehicle, wherein the routine utilizes at least one signal from at least one rotational speed sensor as an input value, c) If it is determined in step b) that there is no stationary state based on the signal of at least one rotational speed sensor, the step of discarding the stationary state of the vehicle detected in step a), Includes.

[0005] In this context, a stationary state of a vehicle specifically means that the vehicle is not moving along or around its axes. These axes are typically the vehicle's longitudinal axis, transverse axis, and height axis. In a stationary state, the vehicle is at rest relative to its longitudinal axis, transverse axis, and height axis. Therefore, stationary state here also means stationary state relative to the ground. A vehicle stationary on a moving vehicle transport system should not be considered stationary in this context. The method described herein allows for the detection of a stationary vehicle, particularly on a rotating vehicle turntable.

[0006] In step a), the vehicle's stationary state is detected based on at least one sensor.

[0007] For this purpose, for example, at least one wheel sensor that calculates the wheel rotation speed can be used. If the calculated rotation speed is close to zero, it is suggested that the vehicle is stationary with a high probability.

[0008] Advantageously, the stationary state is detected using at least three wheel sensors that calculate the rotation speed of at least three wheels. Therefore, the calculation is robust against the failure of one wheel sensor or a wheel that locks up (slips) during braking.

[0009] However, simply detecting a stationary state using wheel sensor signals is insufficient. In particular, incorrect detection results may be sent if, for example, a stationary vehicle is on a rotating vehicle turning plate, or if, for example, the vehicle is moving at a very slow speed (e.g., less than 6-7 km / h).

[0010] In step a), the stationary state can also be detected by conventional methods, for example, from the ESP field. This is simply an assumption of a stationary state. Whether a stationary state truly exists is checked by a test routine in step b).

[0011] In step b), if a stationary state was identified in step a), an inspection routine is performed to check the stationary state of the vehicle. In this routine, at least one signal from at least one rotational speed sensor is used as an input value.

[0012] In order to perform the inspection routine, the signal from the rotational speed sensor is used as the input value in this case.

[0013] A rotational speed sensor is a specific type of inertial sensor used to detect rotational motion around an axis. For driving safety and comfort, typically at least three rotational speed sensors are installed to detect the rotational motion of the vehicle around its longitudinal axis, lateral axis, and height axis. Particularly advantageous is the observation of rotational motion around the height axis. This rotational motion is also called yawing and can be detected by a yaw rate sensor. Therefore, a rotational speed sensor that detects rotational motion around the vehicle's height axis is a yaw rate sensor. In the following, the concepts of "yaw rate sensor" and "rotational speed sensor" will be used partially synonymously. It is also conceivable that rotational motion around the longitudinal axis and / or rotational motion around the lateral axis can be detected at least partially in parallel by the corresponding rotational speed sensors.

[0014] The detected rotational motion is used as a basis for examining the stationary state detected in step a) and is output in the form of an output signal from at least one rotational speed sensor.

[0015] Particularly advantageous, in order to obtain output signals of high quality for obtaining optimal inspection results, the offset of at least one rotational speed sensor is corrected before the output signal is used in stationary inspection. This prevents, for example, a vehicle moving very slowly, or a stationary vehicle rotating together on a rotating vehicle turning plate, from being identified as stationary.

[0016] Rotational motion can typically be described by rotational speed or rotation angle. In this case, rotational speed can be detected directly, for example, by a rotational speed sensor, while rotation angle can be detected indirectly by integrating the rotational speed.

[0017] When the vehicle is truly stationary, the detected rotational speed or rotational angle will remain at its minimum value. In other words, this means that the output of at least one rotational speed sensor is significantly greater in a non-stationary state than in a stationary state.

[0018] Based on the output of at least one rotational speed sensor, the stationary state detected in step a) can be inspected by the inspection routine in step b), and the assumed stationary state can be confirmed based on the inspection results.

[0019] Advantageously, the testing routine, in step b), - A subroutine that pre-corrects the signal of a rotational speed sensor having at least one offset of at least one rotational speed sensor by filtering. - A subroutine that integrates the signal within a pre-set integration time window (after the offset has been corrected in advance), and - A subroutine for inspecting the stationary state detected in step a) based on the integrated (pre-corrected) signal of at least one rotational speed sensor includes at least one of them.

[0020] Particularly preferably, in step b), the rotational speed signal / yaw rate signal of at least one rotational speed sensor / yaw rate sensor pre-corrected is used as an input value. The pre-corrected rotational speed signal / yaw rate signal is integrated in a predetermined integration window as a basis for inspecting the stationary state. Thus, in particular, it is possible to avoid that a vehicle located on a rotating vehicle turning plate (turntable) is detected as being in a stationary state.

[0021] If it is identified in step b) based on the signal of at least one rotational speed sensor that there is no stationary state, in step c), the stationary state of the vehicle detected in step a) is discarded.

[0022] According to the method described herein, the output signal of the pre-corrected rotational speed sensor is used for the detection of a vehicle transport device, for example, a vehicle turning plate. Such detection verifies the conventional detection of the stationary state based on the wheel sensor signal later and is used for the confirmation of the assumption of the stationary state.

[0023] In order to minimize false warnings, in the method proposed herein, a valid rotational speed measurement and a rotational speed measurement with insufficient calibration are distinguished. Even in the case of a valid rotational speed measurement, the measurement quality at the time of detecting the vehicle transport device is considered to achieve an optimal detection result.

[0024] The present invention is particularly applicable to a GNSS- and INS-based positioning system for autonomous driving or semi-autonomous driving.

[0025] Preferably, in step b), the variance of the signal of at least one rotational speed sensor is evaluated.

[0026] The variance of the signal typically corresponds to the output of the (averaged) AC component of the signal and describes the vibration intensity of the signal. When the signal vibrates strongly, the variance of the signal also becomes correspondingly large. The output signal with large vibrations from at least one rotational speed sensor can be an indicator of a non-static state.

[0027] Advantageously, the static state detected in step a) can be inspected in step b) as an inspection routine using the variance of the output signal of at least one rotational speed sensor.

[0028] Particularly advantageously, the variance of the signal is calculated after the offset of the corresponding rotational speed sensor has been corrected in advance. By doing so, the detected static state can be inspected in step b) without the negative influence of the offset.

[0029] Preferably, when the variance of the signal of the rotational speed sensor exceeds a first threshold value, it is identified that there is no static state.

[0030] When the vehicle is truly in a static state, the variance of the output signal of at least one rotational speed sensor is ideally close to zero or slightly larger than zero considering some possible uncertainties. If the variance exceeds (greatly) the minimum value, this indicates that there is no static state. Here, the minimum value is referred to as the first threshold value and can be preset according to the characteristics of the sensors involved and possible driving scenarios.

[0031] Preferably, step b) is performed particularly to identify the case where the vehicle is located on a rotary disk and moving with the rotary disk.

[0032] A rotary disc is typically a vehicle turning plate (turntable) that facilitates the turning of vehicles. Such turning plates are often installed in narrow, restricted parking spaces. Vehicles can be easily rotated to their desired entry or exit position using the turning plate. In conventional methods, it is difficult, or impossible, to distinguish a stationary vehicle from a stationary one due to its position on a rotating turning plate.

[0033] In step b), the scenario described above is inspected in particular. For this purpose, at least one rotational speed sensor / yaw rate sensor is applied to detect the rotational speed / yaw rate around the vehicle's yaw axis (i.e., height axis). To obtain optimal inspection results, after the detection of the stationary state in step a) and before the use of the rotational speed signal / yaw rate signal, the offset of the rotational speed sensor / yaw rate sensor is corrected, for example, by a filter. The rotational speed signal / yaw rate signal is further integrated over a preset time window. When the integrated value reaches a threshold, it is identified that the vehicle is on the vehicle turning plate.

[0034] Preferably, in step b), the rotational speed sensor signal is detected in a state integrated within a preset integration window.

[0035] A rotational speed sensor typically detects the (instantaneous) rotational speed around a single axis. Particularly advantageous is that the rotational speed is integrated over an integration time window. The rotational speed integrated over the integration time window corresponds to the rotation angle around the axis. The integration time window can be preset depending on the application scenario. For example, to detect a stationary vehicle on a rotating vehicle turntable, the integration time window can be set to, for example, 5 seconds, preferably 3 seconds, according to the normally expected rotational speed of the vehicle turntable.

[0036] Preferably, in step b), the vehicle is detected to be stationary when the rotational speed sensor signal falls below a preset second threshold.

[0037] A second threshold can be set in advance, for example, by a conceivable driving scenario. A typical driving scenario, as described above, is when a stationary vehicle is on a vehicle turning plate that is rotating for turning. In this driving scenario, the vehicle can turn 360°. The second threshold can be determined, for example, by considering a pre-set integral time window and the normal rotation speed of the vehicle turning plate.

[0038] In the case of turning using a vehicle turning plate, for example, if the detected rotation angle does not reach 30° within 3 seconds, the vehicle can be detected as stationary. In contrast, a vehicle turning plate (i.e., a vehicle stationary on a rotating vehicle turning plate) is identified when the detected rotation angle reaches 30° within 3 seconds.

[0039] Preferably, the rotation angle sensor is a yaw rate sensor.

[0040] Lateral tilt and braking or acceleration pitching are usually harmless, but the rotation of the vehicle body around the height axis is typically an indicator of critical driving conditions. Therefore, yaw rate sensors are central components of any high-precision driving dynamics control, including rotational motion around the height axis (which is called yaw rate).

[0041] The yaw rate sensor also has the aforementioned (inertial) offset that needs to be corrected in a stationary state. If the vehicle is assumed to be in motion based on an incorrect detection of the vehicle's stationary state during the correction of the yaw rate sensor's offset, the accuracy of the subsequent measurements detected by the yaw rate sensor may be impaired.

[0042] According to the method described above, on the one hand, the vehicle's stationary state can be detected in an improved manner using an existing yaw rate sensor, and on the other hand, the vehicle's offset can be corrected in an improved manner by the improved detection of the stationary state.

[0043] The yaw rate sensor can, in particular, detect stationary vehicles on a rotating vehicle turning plate.

[0044] Preferably, the offset of the rotational speed sensor is corrected in advance by a filter.

[0045] The offset of the rotational speed sensor, in this context, specifically refers to the inertial offset, which arises from the characteristics of the rotational speed sensor and is usually also affected by degradation and / or temperature. This offset is output from the rotational speed sensor in the form of a value in a stationary state. This value is usually further corrected in a stationary state.

[0046] Prior correction, in this context, specifically means that the offset is corrected after the assumption of a stationary state is made, but before the assumption of a stationary state is confirmed. This prior correction is distinct from the correction of the offset after the assumption of a stationary state has been confirmed. This means that the vehicle may still move even after the offset has been corrected in advance.

[0047] Prior correction of the offset of the rotational speed sensor is advantageous in that it enables the rotational speed sensor to transmit a high-quality rotational speed sensor signal without the negative effects of the offset described above.

[0048] Prior correction of the offset is particularly advantageous for detecting slight motion on a moving vehicle transport system, such as a stationary vehicle on a vehicle turning plate. In this case, the offset is calculated by appropriate estimation and used to correct the rotational speed sensor signal. Preferably, the variance of the remaining rotational speed sensor signal (without the offset) is also calculated during this estimation.

[0049] When considering the quality of the rotational speed sensor used, Euclidean distance is not used for the analysis of the first threshold. Instead, the difference is weighted by variance. In this case, Mahalanobis distance is used instead of Euclidean distance, which is particularly advantageous.

[0050] Unlike the Euclidean distance approach, the Mahalanobis distance considers the covariance of two probe sets, i.e., two signals, for example, the covariance between the target output signal and the actual output signal of a rotational speed sensor. This allows for the efficient calculation of the similarity between these two unknown probe sets. In addition to the Euclidean distance, the variance of the two signals to be compared is also considered.

[0051] Preferably, if it is confirmed that a stationary state exists based on the signal from at least one rotational speed sensor, a method for calibrating the vehicle's inertial sensor system is initiated following step c).

[0052] In other words, this means that the calibration of the vehicle's inertial sensor system is only performed after confirming the assumption of a stationary state. Therefore, the occurrence of high dynamics in the inertial sensor signals during calibration can be effectively avoided.

[0053] Preferably, a computer program for performing the method described herein is applied. In other words, this relates in particular to a computer program (product) that includes instructions for causing a computer to perform the method described herein when the program is executed by the computer.

[0054] More preferably, a machine-readable storage medium storing the computer program proposed herein is used. Typically, such machine-readable storage medium is a computer-readable data carrier.

[0055] Particularly preferably, a vehicle control device is provided that performs the method described herein.

[0056] The details, features, and advantageous configurations mentioned in relation to the method may also occur in the computer programs and / or storage media and / or location devices proposed herein, and vice versa. In this regard, the methodological descriptions of the detailed characterization of each feature are generally applicable.

[0057] The solutions and technical fields proposed herein will be described in detail below with reference to the drawings. It should be noted that the present invention is not limited to the illustrated embodiments. In particular, unless otherwise specified, partial aspects of the matters described in the drawings may be extracted and combined with other elements and / or other findings from other drawings and / or descriptions herein. The drawings schematically show the following: [Brief explanation of the drawing]

[0058] [Figure 1] This figure shows the flow diagram of the method for detecting the stationary state of a vehicle using a regular operation flow, as proposed herein. [Modes for carrying out the invention]

[0059] Figure 1 schematically illustrates the flow of the method proposed herein for detecting the stationary state of a vehicle by a regular operational flow. The order in which steps a), b), and c) of the method are illustrated in blocks 110, 120, and 130 is illustrative only. In block 110, the stationary state of the vehicle is detected based on at least one sensor. In block 120, if the stationary state is identified in step a), an inspection routine is executed to check the stationary state of the vehicle, which utilizes at least one signal from at least one rotational speed sensor as an input value. In block 130, if it is identified in step b) that the stationary state of the vehicle does not exist based on the signal from at least one rotational speed sensor, the stationary state of the vehicle detected in step a) is discarded.

[0060] In particular, step b) of the method is - The offset of at least one rotational speed sensor is pre-corrected by a filter, and the rotational speed sensor uses the pre-corrected, high-quality signal as the input value in the inspection routine. - A test routine that integrates the signal within a pre-set integration time window (after the offset has been corrected in advance), and - An inspection routine that checks the stationary state detected in step a) based on the integrated (pre-corrected) signal from at least one rotational speed sensor. It includes at least one of the following.

[0061] In particular, in step b) of the method, a pre-corrected yaw rate signal from at least one yaw rate sensor is used as the input value. The pre-corrected yaw rate signal is integrated over a predetermined integration window as a basis for inspecting the stationary state.

[0062] The method described above makes it possible to avoid detecting a vehicle located on a rotating vehicle turning plate (turntable) as being stationary.

Claims

1. A method for detecting a stationary state of a vehicle, comprising at least the following steps: a) detecting a stationary state of the vehicle based on at least one sensor; b) if a stationary state is identified in step a), executing a check routine for checking the stationary state of the vehicle, said routine utilizing at least one signal of at least one rotational speed sensor as input value; c) discarding the stationary state of the vehicle detected in step a) if it is determined in step b) that a stationary state does not exist based on the signal of the at least one rotational speed sensor; The method includes:

2. 2. The method according to claim 1, wherein in step b) the variance of the signals of the at least one rotational speed sensor is evaluated.

3. 3. The method according to claim 2, wherein the absence of a stationary condition is identified if the variance of the signal of the rotational speed sensor exceeds above a first threshold value.

4. 2. The method according to claim 1, wherein step b) is performed in particular to identify when the vehicle is located on a rotary disc and moving together with said rotary disc.

5. 2. The method according to claim 1, wherein in step b) the signal of the at least one rotational speed sensor is detected in an integrated state in a predefinable integration window.

6. 6. The method according to claim 5, wherein in step b) a stationary state of the vehicle is detected if the integrated signal of the at least one rotational speed sensor falls below a predefinable second threshold value.

7. The method of claim 1 , wherein the rotational speed sensor is a yaw rate sensor.

8. 2. The method of claim 1, wherein the offset of the rotational speed sensor is pre-corrected by a filter.

9. 2. The method of claim 1, wherein the offset of the rotational speed sensor is pre-corrected based on a weighted difference and a weighted variance of the output signal.

10. 10. The method according to claim 9, further comprising the step of: if a stationary condition is determined to exist based on the signal of the at least one rotational speed sensor, following step c) a method of calibrating an inertial sensor system of the vehicle is initiated.

11. A computer program for carrying out the method according to any one of claims 1 to 10.

12. A machine-readable storage medium storing the computer program of claim 11.

13. A control device for a vehicle configured to carry out a method according to any one of claims 1 to 10.

Images