DE102020005474A1 - Method for data processing of data in a vehicle by means of a context management system and data processing system - Google Patents

Abstract

The invention relates to a method for data processing of data in a vehicle (1), wherein - the data are recorded by several recording units (3) of the vehicle (1), - the recorded data are transmitted to a data processing system (2) of the vehicle (1), and - the recorded data are evaluated with the data processing system (2) of the vehicle (1), the evaluated data being stored as context information in a memory unit (6) of the data processing system (2), - with various program functions (10, 11, 12) ) the data processing system (2) depending on the context information, each individual information for vehicle components (4) and / or vehicle systems (5) are generated, - the generated individual information is stored in the memory unit (6), and - the stored individual information is respective vehicle components (4) and / or vehicle systems (5) are provided. The invention also relates to a data processing system (2).

Description

The invention relates to a method for processing data in a vehicle, the data being recorded by a plurality of recording units in the vehicle. The recorded data are transmitted to a data processing system of the vehicle, and the recorded data are evaluated with the data processing system of the vehicle.

The invention also relates to a data processing system for a vehicle. The data processing system has a plurality of predetermined acquisition units of the vehicle for acquiring data. An interface unit is provided for transmitting the captured data to the data processing system. The data processing system also has a processing unit for evaluating the recorded data.

A wide variety of software architectures are used to create applications for web technologies (for example, websites, homepages or social networks). For example, the concepts "Flux Application Architecture" or "Redux concept" can be used here. The most diverse loT devices (Internet of Things devices) are networked with one another and with the special platforms and / or servers. Nowadays, loT devices serve as “edge computing platforms”. Cloud-based technology is used to enable greater performance of the individual components. Sensors and / or actuators can be managed using embedded mechatronic components.

With regard to the "Flux Application Architecture", this is used to generate mobile applications. This architecture can be implemented in particular in mobile applications, such as in the Android or IOS operating systems. In particular, the "Flux Application Architecture" in cloud-based domains serves as a mechanism for managing the interoperability of a micro-service architecture.
The US 9,866,637 B2 discloses techniques for the distributed processing of loT device data by edge systems. For example, a selected edge computing system is selected from a plurality of at least one of a plurality of edge computing systems that are located within the respective co-location facility. It is also disclosed that API endpoints (Application Programming Interface) for communication with the loT devices are provided on the selected edge computing system.

The object of the present invention is to optimize the data processing of onboard data and the associated software architecture in vehicles.

This object is achieved by a method and a data processing system according to the independent patent claims. Useful further developments result from the subclaims.

According to the invention, the object is achieved by a method for processing data in a vehicle. The data are recorded by several recording units in the vehicle, and the recorded data are transmitted to a data processing system in the vehicle. The recorded data is evaluated with the vehicle's data processing system. The evaluated data are stored as context information in a storage unit of the data processing system, with various program functions of the data processing system being used to generate individual information for vehicle components and / or vehicle systems depending on the context information. The generated individual information is stored in the storage unit and the stored individual information is made available to the respective vehicle components and / or vehicle systems.

With the proposed method, on-board data and software architectures in the vehicle can be optimized. In particular, the data processing and the software architecture in the vehicle are improved. Thus, for example, faster, leaner and more reliable software-based functions and / or applications based on onboard data can be integrated or activated in the vehicle.

In particular, the proposed method enables an onboard context management system to be integrated in a vehicle, thereby enabling edge computing in the vehicle. In other words, decentralized data processing is provided for the vehicle, with data and services being managed by central nodes. In particular, data streams can be processed in a more resource-efficient manner with the help of edge computing.

In particular, the proposed method enables the “Flux Application Architecture” to be integrated in a vehicle in order to manage the on-board data from connected loT devices.

In particular, this can be done in vehicles, motor vehicles, trucks, passenger cars, airplanes, boats or in any other system that processes sensor data and interacts with its environment as an Internet of Things device. In particular, the proposed method is used advantageously in the field of passenger cars.

By integrating the "Flux Application Architecture" into the onboard data processing functions, the individual processes can be decoupled. This makes it possible to create new applications depending on existing data and / or functions. This has the advantage that the software in vehicles can be developed more agile and faster with the help of cloud-based data processing.

In particular, the proposed method enables “true cloud-native” data processing with decoupled microservice architectures within edge computing devices.

In particular, there is the advantage that the data processing of sensor data can be carried out directly in the vehicle. This enables “true edge” data processing in the vehicle. Likewise, different functions that access the same sensor data can be loosely coupled with one another, which enables a microservice architecture in the vehicle. Furthermore, the proposed method can be used to develop cross-departmental concepts and applications for the vehicle, whereby the possible and various advantages of data domains can be used. In particular, an AI (artificial intelligence) technology can be combined in data processing at any time. For example, raw data from the plurality of detection units can be processed with the data processing system and the individual information for the vehicle components and / or vehicle systems can be generated in real time. The data processing of the data from the sensors of the vehicle can thus be carried out in the vehicle and then transmitted to the respective components and / or systems of the vehicle.

With the microservice architecture, complex application software from independent processes can be combined that communicate with each other with language-independent communication interfaces. The services are largely decoupled and application software can be structured in a modular manner.

Context is any information that can be used to characterize the situation of an entity. In particular, the context information is information that is made available by a wide variety of sources or sensors. This information is used to draw conclusions about the context. The determined context is used by the application to adapt its behavior, in particular the behavior of the user interface. Context can, for example, be a single piece of information or a combination of many pieces of information from different sources.

The data are in particular data objects.

The context describes in which specific roles the data objects are used. These are, for example, the use cases, scenarios and algorithms of the technical task of the respective application. Contexts are objects that represent one or more use cases. The individual data objects are identified with the aid of the context.

Context management means that a computer process that uses data in an application and also uses this in separate applications can be made more dynamic. Context management makes it possible to use data in an application and to take into account all further information from the other applications in this application.

According to the invention, a data processing system for a vehicle is also provided. An interface unit is provided to receive captured data. The data can be evaluated with a processing unit of the data processing system. The data processing system has a storage unit for storing the evaluated data as context information. Furthermore, the data processing unit has program functions for generating individual information for specified vehicle components and / or for specified vehicle systems as a function of the context information. The memory unit is also designed to store the individual information. The interface unit is designed to provide the individual information for the specified vehicle components and / or for the specified vehicle systems.

In particular, the above-mentioned method or a further development thereof can be carried out with the data processing system just described.

Further advantages, features and details of the invention emerge from the following description of preferred exemplary embodiments and on the basis of the drawing (s). The features and feature combinations mentioned above in the description as well as those below in the Features and combinations of features mentioned in the description of the figures and / or shown alone in the figures can be used not only in the respectively specified combination, but also in other combinations or alone, without departing from the scope of the invention.

• 1 eine schematische Darstellung eines in einem Fahrzeug integrierten Datenverarbeitungssystems;
• 2 eine beispielhafte Darstellung des Datenverarbeitungssystems aus 1; und
• 3 eine schematische Darstellung einer Kombination des Datenverarbeitungssystems aus 1 mit einer künstlichen Intelligenz.

The following figures show in:

• 1 a schematic representation of a data processing system integrated in a vehicle;
• 2 an exemplary representation of the data processing system 1 ; and
• 3 a schematic representation of a combination of the data processing system 1 with an artificial intelligence.

In the figures, functionally identical elements are provided with the same reference symbols. The 1 shows a schematic representation of a vehicle 1 , which is a data processing system 2 for processing data in the vehicle 1 having. In particular, the data processing system is 2 an integral part of the vehicle 1 . The vehicle is particularly important in this context 1 to be viewed as an Internet of Things device. In the data processing system 2 it is in particular an on-board context management system. With the vehicle 1 it is in particular a passenger car. The vehicle 1 may include a high performance computer system.

Using the data processing system 2 can receive data from the vehicle 1 recorded in the vehicle 1 processed by yourself.

In particular, the data from several acquisition units 3 of the vehicle 1 detected. With the registration units 3 it is in particular on-board sensors, such as sensors for recording video and / or audio sequences or radar sensors. In particular, the multiple registration units deliver 3 Raw data that must be processed before they can be used as vehicle components 4th and / or vehicle systems 5 can be provided.

Like in the 1 can see the captured data in the data processing system 2 are transmitted so that the data processing system 2 can evaluate the data. In particular, the data are stored as context information in a storage unit 6th of the data processing system 2 saved. For example, the data is evaluated with a processing unit 7th of the data processing system 2 . The data can also be transmitted to the processing unit 7th converted accordingly as context information.

The data processing system can also 2 via an interface unit 8th the data of the registration units 3 recall. The evaluated data, which as context data in the storage unit 6th are stored, can using the data processing system 2 as individual information for the vehicle components 4th and / or vehicle systems 5 converted or generated. The individual information is also stored in the storage unit 6th stored and can from the vehicle components 4th and / or vehicle systems 5 can be accessed. For example, it can be with the vehicle components 4th and / or vehicle systems 5 be driver assistance systems and / or infotainment systems. In the end, the individual information is used by the vehicle user 1 and can use a user interface 9 to be provided.

The 2 shows a generic exemplary overview of the data processing system 2 . It can be seen that the generation of the individual information is dependent on a predetermined sequence of various program functions 10 , 11 , 12 can be carried out. For example, it is the program function 10 to function A and to the program function 11 with function B and with the program function 12 the function C.

The data processing mechanism of the data processing system 2 will now be briefly explained. Any program function 10 to 12 accesses the same data in the storage unit 6th to. The data is the context information. The context information is used for storage in the storage unit 6th each assigned a read access as the standard access type. In particular, the various program functions are dependent on the specified sequence 10 to 12 the context information of the storage unit 6th updated and / or additional context information supplied. The various program functions 10 to 12 accessed by means of sequential access. Like in the 2 can be seen, the data processing proceeds along the data stream 17th . In other words, the respective program function must 10 to 12 wait for her turn to update or add contextual information.

For example, with the activated program function 10 the respective data from the storage unit 6th in condition 13 the storage unit 6th retrieved. Once the program function 10 has completed its processing, the results are stored in the storage unit 6th stored, this now in the second state 14th located. The following program function 11 puts their results in the memory unit 6th by doing third state 15th from. The other processes run analogously. The next program function saves the results in the state 16 the storage unit 6th from. In particular, the data is processed sequentially. Whenever the individual functions are in the memory unit 6th are stored, the state of the context information is defined and every write operation or write access to the context information is consistent, since this is exclusive.

As an example, it can be the vehicle system 5 be a driver assistance system for recognizing traffic lights. The registration units 3 the current signal of the traffic light is recorded and when the traffic light changes to green, an acoustic signal is generated. In particular, the acquisition and processing take place in real time. For example, AI (artificial intelligence) technology can be used for this, in order to be able to evaluate the recorded raw data more efficiently and to provide corresponding information about the objects around the vehicle 1 to detect.

The 3 Figure 3 shows a detailed embodiment of the data processing system 2 in combination with several groupings of an artificial intelligence (AI). Several layers ("Layers") 18 to 21 are shown in a row one above the other. With the columns are multiple domains 20th to 25th shown.

In particular, the data processing system 2 automatically machine-trained with a machine-learning algorithm when generating the individual information.

Thus, the data processing system 2 be trained more efficiently using a functionality of an artificial intelligence. Thus, individual information can be made more efficient and better for the vehicle components 4th and / or vehicle systems 5 to be generated.

The functionality is now explained more clearly below. The bottom layer 18th is used to read the sensor data from the registration units 3 . Several functions can work sequentially and access the raw data in order to generate the individual information. For example, is in the first domain 20th a function for detecting objects, individuals and their behavior is given. The example of traffic light detection can be used again. For example, the function 26th evaluate the traffic lights so that an object can be generated which contains the active color of the traffic lights, the direction of the traffic lights and the distance to the traffic lights. The results can be returned as context information in the storage unit 6th get saved. Here again, a sequential sequence of domains is first used 22nd to 25th carried out. The data is then evaluated again using the function 27 . This is done cyclically, for example. Independent of this further processing step, the context information of the second layer 19th to be provided. The knowledge that is contained in the context information is generated. For example, the function 19th Detect special patterns of traffic lights or take correlations with the current position and / or past data into account. In particular, a change in the current signal of the traffic light can be predicted for the next few seconds. This information can be found in the context information in the storage unit 6th get saved. In particular, this will be done in the next layer 20th saved. In the shift 20th in other words, the knowledge of the context information is generated as “wisdom”. The last step is the context information in the layer 21st for special applications (vehicle components 4th and / or vehicle systems 5 ) provided. In the example of the traffic light, the function 28 Depending on the context information, an acoustic signal can be generated to the user of the vehicle 1 to inform about a signal change to green.

Further analogous examples can be found in the domain 23 a weather forecast or in the domain 24 be a traffic situation. In particular, these are domains shown 22nd to 25th only shown as an example. A wide variety of functions and / or applications can be taken into account.

In the shift 21st are the boundaries of the domains 22nd to 25th not exclusive. In particular, using the data processing system 2 in combination with an artificial intelligence application developed that take advantage of the most diverse domains 22nd to 25th include. In particular, the domains 22nd to 25th can be combined as required.

List of reference symbols

1

vehicle

2

Data processing system

3

Registration units

4th

Vehicle components

5

Vehicle systems

6th

Storage unit

7th

Processing unit

8th

Interface unit

9

User interface

10

Program function

11

Program function

12

Program function

14th

State of the storage unit

15th

State of the storage unit

16

State of the storage unit

17th

Data stream

18th

layer

19th

layer

20th

layer

21st

layer

22nd

domain

23

domain

24

domain

25th

domain

26th

function

27

function

28

function

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 9866637 B2 [0004]

Claims (6)

A method for data processing of data in a vehicle (1), wherein - the data are recorded by several recording units (3) of the vehicle (1), - the recorded data are transmitted to a data processing system (2) of the vehicle (1), and - the recorded data are evaluated with the data processing system (2) of the vehicle (1), characterized in that - the evaluated data are stored as context information in a storage unit (6) of the data processing system (2), - with various program functions (10, 11, 12) ) of the data processing system (2) depending on the context information, each individual information for vehicle components (4) and / or vehicle systems (5) are generated, - the generated individual information is stored in the memory unit (6), and - the stored individual information is respective vehicle components (4) and / or vehicle systems (5) are provided. Procedure according to Claim 1 , characterized in that the context information for storage in the memory unit (6) is assigned a read access as the standard access type. Procedure according to Claim 1 or 2 , characterized in that the various program functions (10, 11, 12) are called up by means of sequential access, the individual information being generated as a function of a predetermined sequence of the various program functions (10, 11, 12). Procedure according to Claim 3 , characterized in that the context information in the memory unit (6) is updated and / or further context information is added as a function of the predetermined sequence of the various program functions (10, 11, 12). Method according to one of the preceding claims, characterized in that the data processing system (2) is automatically, machine-trained with a machine learning algorithm when generating the individual information. Data processing system (2) for a vehicle (1) with - An interface unit (8) for receiving captured data, and - A processing unit (7) for evaluating the recorded data, characterized by - A storage unit (6) for storing the evaluated data as context information, wherein - the processing unit has program functions (10, 11, 12) for generating individual information for specified vehicle components (4) and / or for specified vehicle systems (5) as a function of the context information, - The memory unit (6) is designed to store the individual information, and - The interface unit (8) is designed to provide the individual information for the specified vehicle components (4) and / or for the specified vehicle systems (5).

Images