DE102020005467A1 - Process for making anonymized, ADAS-relevant vehicle data available - Google Patents

Abstract

The invention relates to a method for making anonymized, ADAS-relevant vehicle data (11) available, comprising the steps of: - Acquisition of vehicle data (10) of a vehicle (1) within a time segment, the vehicle data (10) comprising data transmitted via a CAN Bus of the vehicle (1) can be transmitted; - Compression of the vehicle data (10) to form compressed vehicle data (11) in such a way that the compressed vehicle data (11) have a smaller amount of data than the vehicle data (10) and are anonymized; - Making the compressed Vehicle data (11) for ADAS (3).

Description

The invention relates to a method and a system for making available anonymized, ADAS (Advanced Driver Assistance Systems) relevant vehicle data and a computer program product.

The US2015300825 discloses a method and system for operating ADAS. The method includes generating a horizon based on vehicle data.

• - Erfassen von Fahrzeugdaten eines Fahrzeugs innerhalb eines Zeitabschnitts, wobei die Fahrzeugdaten Daten umfassen, welche über einen CAN Bus des Fahrzeugs übertragen werden können;
• - Komprimieren der Fahrzeugdaten zu komprimierten Fahrzeugdaten derart, dass die komprimierten Fahrzeugdaten eine kleinere Datenmenge gegenüber den Fahrzeugdaten aufweisen und anonymisiert sind;
• - Verfügbarmachen der komprimierten Fahrzeugdaten für ADAS.

A method according to the invention for making anonymized, ADAS-relevant vehicle data available comprises the steps:

• Acquisition of vehicle data of a vehicle within a time segment, the vehicle data including data which can be transmitted via a CAN bus of the vehicle;
• Compression of the vehicle data to form compressed vehicle data in such a way that the compressed vehicle data have a smaller amount of data than the vehicle data and are anonymized;
• - Making the compressed vehicle data available for ADAS.

The fact that data which can be transmitted via a CAN bus of the vehicle are included in the vehicle data makes it possible for the vehicle data to allow a detailed overview of a vehicle situation during the time segment. This enables a realistic reconstruction of the vehicle situation in order to then use it for training ADAS.

The vehicle data are preferably recorded for many time segments. This enables a large number of training data for ADAS to be obtained from the vehicle data.

The fact that the vehicle data are compressed in such a way that they are anonymized enables vehicles from different manufacturers to carry out the method according to the invention without the manufacturers having to fear that sensitive vehicle data can be traced back to them. This enables the method according to the invention to be used for a large number of vehicles worldwide, so that a large number of training data can be generated for ADAS.

The vehicle data preferably include sensor data from vehicle-internal sensors. In-vehicle sensors preferably include lidar sensors, radar sensors, cameras and / or other sensors such as steering angle sensors and / or acceleration sensors. The vehicle data preferably also include map data and in particular lane widths and number of lanes of lanes on which the vehicle is located at a point in time. The sensor data are preferably used to generate a realistic reconstruction. This reconstruction preferably includes a representation of the vehicle situation in a top view. This enables a better overview of the vehicle situation.

For the compression, the vehicle data is preferably converted into a barcode. Mathematical abstractions of the vehicle situation are converted in a standardized way into combinations of letters and numbers from ASCII space. This enables the compressed vehicle data to have only a small amount of data for comparison with the vehicle data. In addition, it is technically easy to anonymize the vehicle data. The barcode is preferably such that it can be converted into a simulation of the top view of the vehicle situation.

A system according to the invention for making anonymized, ADAS-relevant vehicle data available comprises a vehicle for acquiring and compressing vehicle data and a neural network for analyzing and making available the compressed vehicle data.

The fact that the system comprises a vehicle and a neural network makes it possible for the data to be recorded independently of the analysis. This enables an effective analysis to be carried out by a high-performance computer which operates spatially separate from the vehicle. This enables a reduction in the analysis time and an increase in the precision of the analysis.

A computer program product according to the invention comprises a program which, when it is executed by a computer, causes the computer to carry out a method according to the invention for selecting data for ADAS.

The dependent claims describe advantageous embodiments of the invention.

• 1 schematisch ein Ausführungsbeispiel eines erfindungsgemäßen Systems zum Verfügbarmachen von anonymisierten, ADAS relevanten Fahrzeugdaten und
• 2 ein weiteres Ausführungsbeispiel eines erfindungsgemäßen Systems zum Verfügbarmachen von anonymisierten, ADAS relevanten Fahrzeugdaten.

Preferred exemplary embodiments are explained in more detail with reference to the following figures. It shows

• 1 schematically an embodiment of a system according to the invention for making available anonymized, ADAS-relevant vehicle data and
• 2 Another exemplary embodiment of a system according to the invention for making anonymized, ADAS-relevant vehicle data available.

1 shows an embodiment of a system according to the invention 4th to make anonymized, ADAS-relevant vehicle data available 11 . The system 4th includes a vehicle 1 for capturing and compressing vehicle data 10 and a neural network 2 for analyzing and making available the compressed vehicle data 11 .

The vehicle 1 comprises an in-vehicle computing device for acquiring and compressing the vehicle data 10 . In doing so, data is recorded by vehicle-internal sensors. In-vehicle sensors include lidar sensors, radar sensors, cameras, steering angle sensors, and acceleration sensors. In addition, data is collected from external sources. This data includes map data. The vehicle data 10 also include information about the sensors, for example manufacturer information and / or operating parameters of the sensors.

The acquisition of vehicle data 10 of the vehicle 1 takes place within a period of time, with the vehicle data 10 Include data sent via a vehicle's CAN bus 1 can be transferred.

The time segment is 5 minutes in this exemplary embodiment, and this step is carried out repeatedly for a plurality of time segments. This enables the vehicle data 10 A variety of ADAS relevant traffic situations may include those of the vehicle 1 are recorded. The time segment with a length of 5 minutes is suitable for mapping, for example, when driving on motorways. For inner-city journeys, a period of a few seconds is selected as an alternative.

The vehicle data 10 are then compressed into compressed vehicle data 11 . The compression takes place in such a way that the compressed vehicle data 11 a smaller amount of data compared to the vehicle data 10 have and are anonymized.

The data is anonymized by the fact that all information which relates to the driver of the vehicle 1 and / or sensitive operating parameters of the vehicle 1 close, when the compression is removed.

The vehicle data are used for compression 10 converted into a barcode.

In this exemplary embodiment, the barcode is designed as a QR code. This is generated in that previously selected relevant sensor data and / or information from sensor data are abstractly added one after the other. For example, the barcode includes information about the distance to a vehicle in front.

The compressed vehicle data 11 are for ADAS 3 exposed. In this exemplary embodiment, this method step is carried out chronologically after the aforementioned method steps.

In this exemplary embodiment, the barcodes become ADAS via mobile networks 3 transfer. The barcodes are used for this ADAS 3 converted into a simulation of the time period. Based on this simulation, ADAS 3 be tested. The simulation includes a top view of the vehicle situation for the time segment.

2 shows a further embodiment of a system according to the invention 4th to make anonymized, ADAS-relevant vehicle data available 11 .

After the first implementation of the steps mentioned in the first embodiment, an external source provides ADAS with relevant categories 12 ready.

In this exemplary embodiment, a first category is driving around a curve.

The neural network 2 categorizes the compressed vehicle data 11 based on the ADAS relevant categories 12 to categorized, compressed vehicle data 13 . This step is carried out repeatedly after the ADAS relevant categories have been made available. The category is the compressed vehicle data 11 as metadata.

This enables all the time periods in which the vehicle 1 drives around a bend to be categorized by the first. By having this step from the neural network 2 is carried out, a precise categorization of all time periods is made possible.

Alternatively, all possible categories are set as ADAS relevant categories.

The neural network 2 adds the categorized, compressed vehicle data 13 to a library as a library entry. This step is performed repeatedly after the compressed vehicle data 13 have been categorized.

This makes it possible that a large number of time segments can be saved and that the individual time segments can be easily searched for and found on the basis of the categorization.

The neural network 2 checks all library entries of an ADAS relevant category 12 for the purpose of detecting missing library entries. This step is carried out after the library entries have been added for the first time.

In this exemplary embodiment, the neural network checks 2 all library entries that are categorized with driving around a curve. This shows that some of the library entries represent time segments during which the vehicle 1 drives around a curve with a curve radius of 50 m. For other entries, the curve radius is 30th m.

The neural network 2 determines an area as curve radii from all library entries, which occurs in reality. In this exemplary embodiment, this area is between 5 m and 500 m. The neural network 2 checks the library entries of the category Drive around a curve and determines that library entries are available for a curve radius resolution of 5 m for all possible curve radii. There are only no library entries for a curve radius of 55 m.

The neural network 2 then creates missing library entries based on existing library entries.

The neural network 2 creates a library entry for driving around a curve with a curve radius of 55 m. The neural network 2 was trained to generate new library entries based on existing library entries. The training took place on the basis of the categorized, compressed vehicle data 11 . This enables the missing library entry that is created to be realistic.

The system 4th comprises a computing device for making available a library of categorized, compressed vehicle data 13 .

Relevant categories from the set of ADAS-relevant categories are provided for the computing device. This step is carried out after missing library entries have been created. Alternatively, it can also be carried out after adding the categorized, compressed vehicle data 13 to the library conceivable.

In this embodiment, the relevant categories are provided by an external source. The external source here is a user of the system 4th , the test for ADAS 3 want to determine. In this embodiment, the user wants ADAS 3 test for cornering. It therefore provides the relevant relevant category.

The computing device outputs all library entries of the extended library which are categorized with the relevant category. This step takes place after the relevant categories have been made available.

The computing device outputs all library entries which are categorized as driving around a curve. The missing library entries created are also output. This enables the ADAS 3 can be fully tested for the relevant category.

A computer program product (not shown) comprises a program which, when it is executed by a computer, causes the computer to carry out an exemplary embodiment of a method according to the invention according to one of the.

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• US 2015300825 [0002]

Claims (10)

Method for making anonymized, ADAS-relevant vehicle data (11) available, comprising the steps: - Acquisition of vehicle data (10) of a vehicle (1) within a period of time, the vehicle data (10) including data which can be transmitted via a CAN bus of the vehicle (1); - Compressing the vehicle data (10) to form compressed vehicle data (11) in such a way that the compressed vehicle data (11) have a smaller amount of data than the vehicle data (10) and are anonymized; - Making the compressed vehicle data (11) available for ADAS (3). Procedure according to Claim 1 , wherein the vehicle data (10) are converted into a barcode for the compression. Method according to one of the preceding claims, comprising the steps: - Provision of ADAS relevant categories (12); - Categorizing the compressed vehicle data (11) using the ADAS-relevant categories (12) into categorized, compressed vehicle data (13). Procedure according to Claim 3 comprising the step of: adding the categorized, compressed vehicle data (13) to a library as a library entry. Procedure according to Claim 4 , comprising the step: - checking all library entries of an ADAS-relevant category (12) for the purpose of detecting missing library entries. Procedure according to Claim 5 , comprising the steps of: generating missing library entries on the basis of existing library entries; - Expand the library with the library entries created to create an expanded library. Procedure according to Claim 6 , comprising the steps: providing a relevant category from the set of ADAS relevant categories; - Output of all library entries of the extended library which are categorized with the relevant category. Computer program product, comprising a program which, when executed by a computer, causes the computer to implement a method according to one of the Claims 1 to 7th perform. System (4) for making anonymized, ADAS-relevant vehicle data (11) available, comprising a vehicle (1) for capturing and compressing vehicle data (10) and a neural network (2) for analyzing and making available the compressed vehicle data (11). System (4) according to Claim 9 comprising a computing device for making available a library of categorized, compressed vehicle data (13).

Images