DE102015215852A1 - Method and device for detecting a non-automated vehicle - Google Patents

Abstract

In a method of detecting a non-automated vehicle, movement data (BD) of a mobile radio device is provided. Depending on the movement data (BD), it is determined whether the mobile radio device is in a non-automated vehicle.

Description

The invention relates to a method and a device for detecting a non-automated operated vehicle.

In modern vehicles, many assistance systems are often used, such as an adaptive cruise control and / or a lane departure warning system.

The object underlying the invention is to contribute to improving such assistance systems.

The object is solved by the features of the independent claims. Advantageous embodiments are characterized in the subclaims.

The invention is characterized by a method for the detection of a non-automated operated vehicle. The invention is further characterized by a device which is designed to carry out the method for the detection of a non-automated vehicle.

In the method of detecting a non-automated vehicle, motion data of a mobile radio device is provided. Depending on the movement data, it is determined whether the mobile radio device is in a non-automated vehicle.

Vehicle movement data can be used, for example, for self-learning vehicle assistance systems such as adaptive cruise control (ACC) and / or lane departure warning systems. However, the movement data must not originate from vehicles in which the respective vehicle assistant is active. Therefore, it is necessary to recognize whether a vehicle is not being operated automatically. Since many mobile radio devices can transmit motion data, a great deal of motion data can be obtained for learning such vehicle assistance systems in this way.

After it has been determined whether the mobile radio device is located in a non-automated vehicle, it is thus possible in particular to associate with the movement data whether the mobile radio device is in a non-automated vehicle. Then, for example, the movement data that has been assigned to the mobile radio in a non-automated vehicle may then be used for self-learning vehicle assistance systems such as proximity control (ACC) and / or lane departure warning assistants.

The movement data comprise in particular position data and / or speed data and / or acceleration data and / or angle data of the mobile radio device. The speed data and / or acceleration data may be representative of lateral speeds or lateral accelerations and / or longitudinal speeds or longitudinal accelerations.

In particular, the mobile radio device comprises a portable device with a wireless network interface (WLAN interface) and / or a mobile radio interface. The mobile radio device comprises in particular a smartphone and / or a tablet PC and / or a laptop and / or a mobile navigation device.

In particular, a non-automated vehicle is a vehicle in which no distance control (ACC) or lane departure warning system is active for the period of time for which the movement data is representative.

For example, in order to determine whether the mobile radio is in a non-automated vehicle, a probability is first determined as to whether the mobile radio is in a non-automated vehicle and then comparing the determined probability to a predetermined threshold.

According to an optional embodiment, the movement data are representative of a course of speed data and / or position data and / or angle data of the mobile radio device. Depending on the movement data, frequency data are determined which are representative of a frequency spectrum of the course of speed data and / or position data and / or angle data. Depending on the frequency data, it is determined whether the mobile radio device is in a non-automated vehicle.

For example, frequency data of a non-automated vehicle often has more high-frequency components than frequency data of an automatically-operated vehicle. Thus, by analyzing the frequency data, it can be easily and reliably determined whether the mobile radio device is in a non-automated vehicle.

The determination of the frequency data takes place for example by means of a fast Fourier Transformation (FFT) and / or a wavelet transformation.

According to a further optional embodiment, a plurality of frequency domain data are determined depending on the frequency data by means of predetermined bandpasses, which are each representative of a frequency range of the frequency spectrum. Respective signal energies of the frequency domain data are compared with each other. Depending on the comparison of the signal energies, it is determined whether the mobile radio device is in a non-automated vehicle.

The signal energies are very easy to determine. Thus, the proportion of high or low frequencies can be determined very quickly with the help of the signal energies.

According to a further optional embodiment, reference data are predetermined, which are representative of a predetermined reference frequency spectrum. Depending on the reference data and the frequency data, correlation data are determined which are representative of a correlation of the frequency spectrum with the reference frequency spectrum. Depending on the correlation data, it is determined whether the mobile radio device is in a non-automated vehicle.

Since it may be known how a frequency characteristic typically runs in a human driver and / or an automatically operated vehicle, it is very easy to determine by specifying a reference frequency spectrum whether the mobile radio device is in a non-automated vehicle.

According to another optional embodiment, a predetermined classifier is provided. Depending on a comparison of the classifier with the frequency data, it is determined whether the mobile radio device is in a non-automated vehicle.

The given classifier has been taught, for example, with typical frequency spectrums of human drivers and / or automated vehicles. Thus, it can be determined very reliably whether the mobile radio device is located in a non-automated vehicle.

According to another optional embodiment, the frequency data is divided into predetermined sections, which are representative of predetermined driving maneuvers. Depending on the divided frequency data, it is determined whether the mobile radio is in a non-automated vehicle.

The given driving maneuvers include in particular constant travel, acceleration and braking. In such driving maneuvers other frequency ranges are possibly decisive, by dividing the frequency data in such sections can thus be determined even more reliable, whether the mobile radio device is located in a non-automated operated vehicle.

According to a further optional embodiment, depending on the movement data, further movement data of a possible preceding vehicle are determined. Depending on the further movement data, expected movement data are determined. The expected movement data is compared with the movement data. Depending on the comparison of the expected movement data with the movement data, it is determined whether the mobile radio device is located in a non-automated vehicle.

In order to determine the possible preceding vehicle, for example, a position from the movement data is compared with positions of vehicles in the environment. For example, if the distance between the positions is less than a predetermined threshold, for example 200 m, then the vehicle is classified as a possible preceding vehicle.

Instead of a possible preceding vehicle, especially for lane departure warning systems, it is also possible to use curves of lane markings which can be determined in a similar manner to the possible preceding vehicle.

The expected motion data can be determined in particular based on a controller model, such as an ACC controller model. For this purpose, for example, initially a set of possible movement data are provided by preceding vehicles, for example, from a backend. On the basis of a situation analysis, such as an ACC situation analysis, the further movement data of the possible preceding vehicle are then determined from the set of possible movement data. Subsequently, for example, the further movement data is made available to a predefined model of controller and / or controller and vehicle, and the expected movement data is determined as a function thereof, that is to say in particular an expected course for the movement data to be evaluated with active ACC control.

If the movement of the preceding vehicle is known, it can be determined very accurately how well an automated vehicle and / or a human driver would drive behind. If the subsequent journey determined in this way corresponds approximately to the movement data, then it is highly probable that the vehicle will be operated automatically or not automatically.

According to a further optional embodiment, the movement data are representative of a course of speed data and / or position data and / or angle data of the mobile radio device. Depending on the movement data, frequency data are determined which are representative of a frequency spectrum of the course of speed data and / or position data and / or angle data. Depending on the movement data, further movement data of a possible preceding vehicle are determined. Depending on the further movement data, expected movement data are determined. Depending on the expected motion data, further frequency data are determined, which are representative of a frequency spectrum of the expected motion data. The frequency data are compared with the other frequency data. Depending on the comparison of the frequency data with the other frequency data, it is determined whether the mobile radio device is in a non-automated vehicle.

Also, two or more of the above ways of determining whether the mobile radio is in a non-automated vehicle may be combined. This can be done, for example, by determining, for each of the determinations to be combined, whether the mobile radio is in a non-automated vehicle, a probability and / or evidence representative of whether the mobile radio is in a non-automated vehicle located. Subsequently, the probabilities and / or evidences determined in this way can be determined by means of an arithmetic and / or weighted arithmetic mean filter and / or a suitable probabilistic method, such as, for example, B. a Bayesian filter or the Dempster-Shafer theory.

Embodiments of the invention are explained in more detail below with reference to the schematic drawings. It shows:

1 a first flow chart for the detection of a non-automated vehicle,

2 another flowchart for the detection of a non-automated operated vehicle.

Elements of the same construction or function are provided across the figures with the same reference numerals.

The 1 shows a flowchart of a program for the detection of a non-automated operated vehicle. The program can be processed, for example, by a control device. For this purpose, the control device BE has in particular a computing unit, a program and data memory and, for example, one or more communication interfaces. The program and data memory and / or the arithmetic unit and / or the communication interfaces of the control device BE can be formed in a structural unit and / or distributed over a plurality of structural units. The control device BE is arranged, for example, in a server and / or in a backend.

The control device BE can also be referred to as a device for detecting a non-automated operated vehicle.

For this purpose, the program is stored in particular on the data and program memory of the control device BE.

The program is started in a step S1 in which variables can be initialized if necessary.

In a step S3, movement data BD of a mobile radio device is provided. The movement data BD are for example representative of a course of speed data and / or position data and / or angle data of the mobile radio device.

In a step S5 it is determined, depending on the movement data BD, whether the mobile radio device is in a non-automated vehicle. For example, the determination of whether the mobile radio device is in a non-automated vehicle is made by determining a likelihood of whether the mobile radio device is in a non-automated vehicle. Subsequently, the determined probability is compared, for example, with a predetermined threshold value.

2 shows an exemplary detailed flow of step S5. The step S5 may therefore include in particular a plurality of sub-steps.

In a sub-step S501, depending on the movement data BD, frequency data FD are determined which are representative of a frequency spectrum of the course of speed data and / or position data and / or angle data.

In a sub-step S503, the frequency data FD are subdivided into predetermined sections AS which are representative of given driving maneuvers.

In a sub-step S505 a plurality of frequency domain data FBD are determined depending on the frequency data FD by means of predetermined bandpasses, which are each representative of a frequency range of the frequency spectrum. Subsequently, the program is continued in a sub-step S511.

In the sub-step S511, respective signal energies SE of the frequency domain data FBD are compared with each other. Depending on the comparison of the signal energies SE, it is determined whether the mobile radio device is in a non-automated vehicle. Subsequently, the program is continued in a sub-step S513.

In a sub-step S507, which can be carried out in parallel to the sub-step S505, reference data RD are given, which are representative of a predetermined reference frequency spectrum. Depending on the reference data RD and the frequency data FD, correlation data are determined which are representative of a correlation of the frequency spectrum with the reference frequency spectrum. Depending on the correlation data, it is determined whether the mobile radio device is in a non-automated vehicle. Subsequently, the program is continued in the sub-step S513.

In a sub-step S509, which can be carried out in parallel with the sub-step S505, a predetermined classifier KL is provided and, depending on a comparison of the classifier KL with the frequency data FD, it is determined whether the mobile radio device is in a non-automated vehicle. Subsequently, the program is continued in the sub-step S513.

In sub-step S513, the results from sub-steps S511 and / or S507 and / or S509 are merged.

For this purpose, the individual results include, for example, probabilities and / or evidences as to whether the mobile radio device is in a non-automated vehicle. Subsequently, the probabilities and / or evidences by means of an arithmetic and / or weighted arithmetic mean filter and / or a suitable probabilistic method, such. B. a Bayesian filter or the Dempster-Shafer theory combined.

In a sub-step S515, which can be executed in parallel to the sub-step S501, further movement data BD_2 of a possible preceding vehicle are determined.

In a sub-step S517, depending on the further movement data BD_2, expected movement data BD_3 are determined. Subsequently, the program is continued in a sub-step S519 and / or in a sub-step S521.

In a sub-step S519, depending on the expected movement data BD_3, further frequency data are determined, which are representative of a frequency spectrum of the expected movement data BD_3. The frequency data FD are compared with the other frequency data. Depending on the comparison of the frequency data FD with the other frequency data, it is determined whether the mobile radio device is in a non-automated vehicle.

In a sub-step S521, the expected movement data BD_3 are compared with the movement data BD. Depending on the comparison of the expected movement data BD_3 with the movement data BD, it is determined whether the mobile radio device is in a non-automated vehicle.

In a sub-step S523, the results from sub-steps S519 and / or S521 are merged. This can be done in the same way as in step S513.

In a sub-step S525, the results from sub-steps S513 and / or S523 are merged. This can be done in the same way as in step S513.

In a step S7 (see 1 ), the program is ended and may optionally be restarted in step S1.

Claims (10)

 Method for detecting a non-automated vehicle in which: Providing movement data (BD) of a mobile radio device, - Is determined depending on the movement data (BD), whether the mobile radio device is located in a non-automated vehicle. Method according to Claim 1, in which - the movement data (BD) are representative of a course of speed data and / or Position data and / or angle data of the mobile radio device, - be determined depending on the movement data (BD) frequency data (FD), which are representative of a frequency spectrum of the course of speed data and / or position data and / or angle data, - is determined depending on the frequency data whether the mobile radio device is in a non-automated vehicle.  The method of claim 2, wherein - Depending on the frequency data by means of predetermined bandpasses several frequency domain data (FBD) are determined, each of which is representative of a frequency range of the frequency spectrum and - Respective signal energies (SE) of the frequency domain data (FBD) are compared with each other and - Is determined depending on the comparison of the signal energies (SE), whether the mobile radio device is located in a non-automated vehicle.  Method according to one of claims 2 or 3, in which - Reference data (RD) are given, which are representative of a predetermined reference frequency spectrum, Be determined depending on the reference data (RD) and the frequency data (FD) correlation data, which are representative of a correlation of the frequency spectrum with the reference frequency spectrum and - Is determined depending on the correlation data, whether the mobile radio device is located in a non-automated vehicle. A method according to any of claims 2 to 4, wherein A predetermined classifier (KL) is provided and, depending on a comparison of the classifier (KL) with the frequency data (FD), it is determined whether the mobile radio device is in a non-automated vehicle.  Method according to one of Claims 2 to 5, in which the frequency data (FD) are subdivided into predetermined sections (AS) which are representative of predetermined driving maneuvers and - is determined depending on the divided frequency data (FD), whether the mobile radio device is located in a non-automated vehicle.  Method according to one of the preceding claims, in which Be determined depending on the movement data (BD) further movement data (BD_2) of a possible preceding vehicle, - depending on the further movement data (BD_2) expected movement data (BD_3) are determined - the expected transaction data (BD_3) are compared with the transaction data (BD) and - Is determined depending on the comparison of the expected movement data (BD_3) with the movement data (BD), whether the mobile radio device is located in a non-automated vehicle.  Method according to one of the preceding claims, in which The movement data (BD) are representative of a course of speed data and / or position data and / or angle data of the mobile radio device, Be determined dependent on the movement data (BD) frequency data (FD) which are representative of a frequency spectrum of the course of speed data and / or position data and / or angle data, Be determined depending on the movement data (BD) further movement data (BD_2) of a possible preceding vehicle, - depending on the further movement data (BD_2) expected movement data (BD_3) are determined Depending on the expected movement data (BD_3) further frequency data are determined, which are representative of a frequency spectrum of the expected movement data (BD_3), - The frequency data (FD) are compared with the other frequency data and - Is determined depending on the comparison of the frequency data (FD) with the other frequency data, whether the mobile radio device is located in a non-automated vehicle.  A method according to any one of claims 1 to 8, wherein two or more of the ways of determining claims 1 to 8, whether the mobile radio is in a non-automated vehicle, are combined, By determining, for each of the determinations to be combined, whether the mobile radio is in a non-automated vehicle, a probability and / or evidence representative of whether the mobile radio is in a non-automated vehicle and - The probabilities and / or evidences thus determined are combined by means of an arithmetic and / or weighted arithmetic mean filter and / or a predetermined probabilistic method. Device for the detection of non-automated operated vehicle, wherein the device is designed to carry out a method according to one of claims 1 to 9.

Images