DE102020130646A1 - Method, system and computer program product for detecting a change in a wheel steering angle in a motor vehicle - Google Patents

Abstract

The invention relates to a method for detecting a change in a wheel steering angle of at least one wheel in a motor vehicle (10) using a data analysis device (40), comprising: - recording (S10) image and sensor data (32) using a camera and sensor device (30 ) of a second motor vehicle (20), the image and sensor data (32) representing the surroundings of the motor vehicle (20) with at least the first motor vehicle (10);- forwarding (S20) the image and sensor data (32) to the data analysis device (40), wherein the data analysis device (40) comprises a detection system (400) for detecting a change in a wheel steering angle of at least one wheel in the first motor vehicle (10), which uses algorithms from the field of artificial intelligence (AI) and machine image analysis; - Processing (S30) of the image and sensor data (32) in the data analysis device (40) by means of the detection system (400) to possible changes in the Radlenkwink els of the wheel;- Assigning (S40) the classified change in the wheel steering angle to at least one state (Sj) from a set of defined states (S1, S2, ..., Sn);- Generating (S50) output data (450 ) from the determined state (Sj) for further use in an automated driving function and/or for a user interface (70).

Description

The invention relates to a method, a system and a computer program product for detecting a change in a wheel steering angle of at least one wheel in a motor vehicle.

When a motor vehicle is driven by a person, the driver of the vehicle observes very closely and intuitively whether there are any possible sources of danger on the route. In particular, vehicles ahead are closely observed to obtain various information, such as the speed of the vehicle ahead or whether an overtaking maneuver is planned. A trained human driver makes these observations intuitively while driving and is often even aware of how he processes the information and assigns it to a possible dangerous situation.

In semi-autonomous and autonomous vehicles, camera systems and sensors are used to obtain information about the vehicle's surroundings. The development of highly automated driving (HAD) is therefore accompanied by an increase in the requirements for vehicle sensor systems to record suitable sensor data such as image data in particular. In addition, the recorded sensor data must be carefully interpreted in order to draw the right conclusions regarding a possible dangerous situation.

A particularly hazardous situation arises from parked vehicles in a parking space that a moving vehicle is driving past. It is often difficult for a driver to tell whether a parked vehicle wants to leave the parking space. Although the brake lights are often used as indicators for whether a vehicle wants to switch from the parking position to driving mode, they can be defective or they are overlooked due to the large number of information signals from the driver of a first vehicle driving past the parked second vehicle . A sure indication of a vehicle leaving a parking space, however, are changes in the steering angle of the wheels of the vehicle when the driver of the second parked vehicle has started the driving maneuver to leave the parking space. This applies in a similar way to an overtaking manoeuvre, in which a change in the wheel steering angle of the wheels also indicates a change in driving behavior.

An object of the present invention is therefore to create a method, a system and a computer program product for detecting a change in a wheel steering angle on at least one wheel of a motor vehicle, which is characterized by high reliability and efficient use of computing capacities.

This object is achieved according to the invention with regard to a method by the features of patent claim 1, with regard to a system by the features of patent claim 9, and with regard to a computer program product by the features of patent claim 15. The further claims relate to preferred embodiments of the invention.

• - Aufnehmen von Bild- und Sensordaten mittels einer Kamera- und Sensoreinrichtung eines zweiten Kraftfahrzeugs, wobei die Bild- und Sensordaten die Umgebung des zweiten Kraftfahrzeugs mit zumindest dem ersten Kraftfahrzeug repräsentieren;
• - Weitergeben der Bild- und Sensordaten an die Datenanalyseeinrichtung, wobei die Datenanalyseeinrichtung ein Erkennungssystem zur Erkennung einer Änderung eines Radlenkwinkels zumindest eines Rades des ersten Kraftfahrzeugs umfasst, welches Algorithmen aus dem Bereich der künstlichen Intelligenz (KI) und der maschinellen Bildanalyse verwendet;
• - Bearbeiten der Bild- und Sensordaten in der Datenanalyseeinrichtung mittels des Erkennungssystems, um mögliche Änderungen des Radlenkwinkels des Rades zu klassifizieren.;
• - Zuordnen der klassifizierten Änderung des Radlenkwinkels zu zumindest einem Zustand S_j aus einer Menge von definierten Zuständen S₁, S₂, ..., S_n;
• - Erzeugen von Ausgabedaten aus dem ermittelten Zustand S_j zur weiteren Verwendung in einer automatisierten Fahrfunktion und/oder für eine Benutzerschnittstelle.

According to a first aspect, the invention relates to a method for detecting a change in a wheel steering angle of at least one wheel in a motor vehicle using a data analysis device. The procedure comprises the following procedural steps:

• - Recording of image and sensor data by means of a camera and sensor device of a second motor vehicle, the image and sensor data representing the surroundings of the second motor vehicle with at least the first motor vehicle;
• - Passing on the image and sensor data to the data analysis device, the data analysis device comprising a detection system for detecting a change in a wheel steering angle of at least one wheel of the first motor vehicle, which uses algorithms from the field of artificial intelligence (AI) and machine image analysis;
• - processing the image and sensor data in the data analysis device by means of the detection system in order to classify possible changes in the wheel steering angle of the wheel.;
• - Assigning the classified change in the wheel steering angle to at least one state S _j from a set of defined states S ₁ , S ₂ , ..., S _n ;
• - Generation of output data from the determined state S _j for further use in an automated driving function and/or for a user interface.

In a development of the invention, it is provided that the processing of the image and sensor data and the generation of output data take place in real time.

In one embodiment, the recognition system includes an analysis module and a classification module.

Advantageously, the recognition system uses deep learning with a neural network.

In particular, the neural network is designed as a folded neural network (convolutional neural network).

A further development provides that the classification _module contains features M ₁ , M ₂ , .

The image and sensor data are advantageously transmitted to the data analysis device by means of a mobile radio connection, with 5G radio modules being used in particular.

In one embodiment it is provided that the image and sensor device includes an action camera.

According to a second aspect, the invention relates to a system for detecting a change in a wheel steering angle of at least one wheel of a motor vehicle. The system includes an image and sensor device of a second motor vehicle for recording image and sensor data and a data analysis device. The image and sensor data represent the surroundings of the second motor vehicle with at least the first motor vehicle. The data analysis device includes a detection system for detecting a change in a wheel steering angle of at least one wheel in the first motor vehicle. The recognition system uses algorithms from the field of artificial intelligence (AI) and machine image analysis. The data analysis device is designed to process the image and sensor data using the recognition system, to classify possible changes in the wheel steering angle of the wheel, to assign at least one state S _j from a set of defined states S ₁ , S ₂ , ... to the classified change in the wheel steering angle. , S _n and generate output data from the determined state S _j for further use in an automated driving function and/or for a user interface.

In a development of the invention, it is provided that the processing of the image and sensor data and the generation of output data take place in real time.

In one embodiment, the recognition system includes an analysis module and a classification module.

Advantageously, the recognition system uses deep learning with a neural network, which is designed in particular as a folded neural network (convolutional neural network).

In particular, the classification _module contains features M ₁ , M ₂ , .

In a further development, the image and sensor data are transmitted to the data analysis device by means of a mobile radio connection, with 5G radio modules being used in particular.

In particular, the image and sensor device includes an action camera.

According to a third aspect, the invention relates to a computer program product, comprising an executable program code which is configured in such a way that when it is executed it carries out the method according to the first aspect.

The invention is explained in more detail below using an exemplary embodiment illustrated in the drawing.

• 1 eine schematische Darstellung eines erfindungsgemäßen Systems zur Erkennung einer Änderung eines Radlenkwinkels zumindest eines Rades bei einem Kraftfahrzeug;
• 2 ein Ausführungsbeispiel eines Erkennungssystems gemäß der Erfindung;
• 3 ein Flussdiagramm zur Erläuterung der einzelnen Verfahrensschritte eines erfindungsgemäßen Verfahrens;
• 4 ein Computerprogrammprodukt gemäß einer Ausführungsform der des dritten Aspekts der Erfindung.

It shows:

• 1 a schematic representation of a system according to the invention for detecting a change in a wheel steering angle of at least one wheel in a motor vehicle;
• 2 an embodiment of a detection system according to the invention;
• 3 a flowchart to explain the individual steps of a method according to the invention;
• 4 a computer program product according to an embodiment of the third aspect of the invention.

Additional features, aspects and advantages of the invention or embodiments thereof will become apparent from the detailed description coupled with the claims.

In 1 the system 100 according to the invention for detecting a change in a wheel steering angle in at least one wheel of a motor vehicle 10 is shown. In particular, the motor vehicle 10 is a parked motor vehicle. A camera and sensor device 30 of a second motor vehicle 20 records the surroundings of the motor vehicle 20 and thus also the parked motor vehicle 10 when the second motor vehicle 20 drives past the first motor vehicle 10, for example. The camera and sensor device 30 forwards the recorded image data and measured values to a data analysis device 40 for further processing.

The camera and sensor device 30 captures image and sensor data 32 in a recording area. The camera and sensor device 30 includes in particular an RGB camera 35 in the visible range with the primary colors blue, green and red. However, there can also be an additional UV camera 37 in the ultraviolet range and/or an IR camera 39 in the infrared range be provided. The cameras 35, 37, 39, which differ in terms of their recording spectrum, can thus image different lighting conditions in the recording area.

The recording frequency of the camera of the camera and sensor device 30 is designed for fast speeds of the wheel movements and can record image data 32 with a high image recording frequency. Furthermore, the camera and sensor device 30 can be equipped with a microphone for the detection of acoustic signals. As a result, rolling noises from the tires or engine noises can be recorded.

Provision can also be made for the camera and sensor device 30 to automatically start the image recording process when there is a significant change in terms of area in the recording area of the camera and sensor device 30, for example when a vehicle 10 appears in the recording area of the camera and sensor device 30. This enables a selective data acquisition process and only relevant image and sensor data 32 are processed by the data analysis device 40 . This allows computing capacities to be used more efficiently.

In addition, a GPS connection is advantageously provided in order to determine the geographic location and to assign this to the recorded image and sensor data 32 .

In particular, it is provided to use a weatherproof action camera as the camera, which can be arranged in particular in the exterior of the vehicle 20 . Action cameras have wide-angle fisheye lenses, which makes it possible to achieve a visible radius of around 180°. Action cameras can usually record videos in Full HD (1,920 x 1,080 pixels), but action cameras in Ultra HD or 4K (at least 3,840 x 2,160 pixels) can also be used, resulting in a significant increase in image quality. The frame rate is typically 60 frames per second in 4K and up to 240 frames per second in Full HD. An integrated image stabilizer can also be provided. In addition, action cameras are often equipped with an integrated microphone. In addition, methods of differential signal processing can be used to selectively suppress background noise.

The image and sensor data 32 recorded by the camera and sensor device 30 are preferably forwarded to the data analysis device 40 via a wireless mobile phone connection.

The data analysis device 40 preferably has a processor 41 which processes the image and sensor data 32 using a recognition system 400 . The processor 41 or another processor is also designed to control the camera and image recording device 30 . However, it is also conceivable that the image and sensor data 32 are initially stored in a memory unit 50 or a software module 55 and only processed by the data analysis device 40 at a later point in time. The data analysis device 40 and the processor 41 can be integrated in the vehicle 20 or can be embodied as a cloud-based solution that is connected to the vehicle 20 via a mobile radio connection.

In addition, the data analysis device 40 can access one or more other databases 60 . For example, classification parameters for analyzing the recorded image and sensor data 32 or other images and/or parameters can be stored in the database 60 . Furthermore, target variables and target values that define a safety standard can be stored in the database 60 . Furthermore, a user interface 70 can be provided for entering further data and for displaying the calculation results created by the data analysis device 40 . In particular, the user interface 70 is designed as a display with a touch screen.

In the context of the invention, a “processor” can be understood to mean, for example, a machine or an electronic circuit or a powerful computer. A processor can in particular be a central processing unit (CPU), a microprocessor or a microcontroller, for example an application-specific integrated circuit or a digital signal processor, possibly in combination with a memory unit for storing program instructions, etc . A processor can also be understood to mean a virtualized processor, a virtual machine or a soft CPU. It can also be a programmable processor, for example, which is equipped with configuration steps for executing the method according to the invention or configured with configuration steps in this way is that the programmable processor implements the inventive features of the method, the component, the modules, or other aspects and/or partial aspects of the invention. In addition, highly parallel computing units and powerful graphics modules can be provided.

In the context of the invention, a “memory unit” or “memory module” and the like can mean, for example, a volatile memory in the form of random-access memory (RAM) or a permanent memory such as a hard disk or a data carrier or z. B. be understood as a removable memory module. However, the storage module can also be a cloud-based storage solution.

In connection with the invention, a “module” can be understood to mean, for example, a processor and/or a memory unit for storing program instructions. For example, the processor is specially set up to execute the program instructions in such a way that the processor executes functions in order to implement or realize the method according to the invention or a step of the method according to the invention.

In the context of the invention, recorded image and sensor data 32 is to be understood as meaning both the raw data and already processed data from the recording results of the image and sensor device 30 .

In particular, the image and sensor device 30 can have mobile radio modules of the 5G standard. 5G is the fifth-generation mobile communications standard and, compared to the 4G mobile communications standard, is characterized by higher data rates of up to 10 Gbit/sec, the use of higher frequency ranges such as 2100, 2600 or 3600 megahertz, increased frequency capacity and thus increased data throughput and real-time data transmission because up to one million devices per square kilometer can be addressed simultaneously. The latency times are a few milliseconds to less than 1 ms, so that real-time transmission of data and calculation results is possible. The image and sensor data 32 determined by the image and sensor device 30 are sent in real time to a cloud platform, where the corresponding analysis and calculation is carried out. The analysis and calculation results are sent back to the vehicle 20 in real time and can therefore be quickly integrated into instructions for the driver or into automated driving functions. This data transmission speed is required if cloud-based solutions are to be used for processing the image and sensor data 32 .

Cloud-based solutions offer the advantage of high and therefore fast computing power.

When the data analysis device 40 is integrated in the vehicle, an AI hardware acceleration such as the Coral Dev Board is advantageously used for the processor 41 in order to enable real-time processing. This is a microcomputer with a tensor processing unit (TPU) that allows a pre-trained software application to evaluate up to 70 frames per second.

2 shows the detection system 400 according to the invention, which is designed as a software application for analyzing and processing the captured and/or stored image and sensor data 32 in order to detect a change in the wheel steering angle on the vehicle 10 . In particular, the recognition system 400 processes the captured image and sensor data 32 using artificial intelligence algorithms and machine image analysis in order to select and classify them. The recognition system 400 advantageously uses algorithms from the field of machine learning, preferably deep learning with, for example, folded neural networks (convolutional neural networks) to analyze the captured image and sensor data 32.

A neural network consists of neurons that are arranged in several layers and connected to each other in different ways. A neuron is able to receive information from outside or from another neuron at its input, to evaluate the information in a certain way and to pass it on to another neuron in a changed form at the neuron output or to output it as the end result. Hidden neurons are located between the input neurons and output neurons. Depending on the network type, there can be several layers of hidden neurons. They ensure the forwarding and processing of the information. Finally, output neurons deliver a result and output it to the outside world. Different types of neural networks such as feedforward networks, recurrent networks or convolutional neural networks are created by the arrangement and linking of the neurons. The networks can be trained through unsupervised or supervised learning

In particular, the recognition system 400 has an analysis module 410 configured as a convolutional neural network (CNN). The image and sensor data 32 of the camera and sensor device 30 ver turns. In addition, data from the database 60 can also be used. The data formats of the input data are preferably in the form of tensors. In addition, different image formats can advantageously be used.

The convolutional neural network is a special form of an artificial neural network. It has multiple layers of convolution and is well suited for machine learning and artificial intelligence (AI) applications in the field of image and speech recognition. The functionality of a convolutional neural network is to a certain extent modeled on biological processes and the structure is comparable to the visual cortex of the brain. The training of a convolutional neural network usually takes place under supervision. Conventional neural networks consist of fully or partially meshed neurons in several levels. However, these structures reach their limits when processing images, since there would have to be a number of inputs corresponding to the number of pixels. The convolutional neural network is made up of different layers and is based on the basic principle of a partly locally meshed neural feedforward network. The individual layers of the CNN are the convolutional layer, the pooling layer and the fully meshed layer. The pooling layer follows the convolutional layer and can exist multiple times in this combination. Since the pooling layer and the convolutional layer are locally meshed subnets, the number of connections in these layers remains limited and manageable, even with large amounts of input. The final step is a fully meshed layer. The convolutional layer is the actual convolution layer and is able to recognize and extract individual features in the input data. In image processing, these can be features such as lines, edges or specific shapes. The input data is processed in the form of tensors such as a matrix or vectors. The pooling layer, also known as the subsampling layer, compresses and reduces the resolution of the recognized features using appropriate filter functions. In particular, a maxpool function is used for this, which calculates the maximum value for a (usually) non-overlapping sub-area of the data. In addition to maximum pooling, mean pooling can also be used. Pooling discards superfluous information and reduces the amount of data. This does not reduce machine learning performance. The reduced data volume increases the calculation speed.

At the end of the convolutional neural network is the fully connected layer. It is the result of the repeated sequencing of the convolutional and pooling layers. All features and elements of the upstream layers are linked to each output feature. The fully connected neurons can be arranged in multiple levels. The number of neurons depends on the input data that the convolutional neural network is supposed to process.

The convolutional neural network (CNN) therefore offers numerous advantages over conventional non-convoluted neural networks. It is suitable for machine learning and artificial intelligence applications with large amounts of input data, such as in image recognition. The network works reliably and is not sensitive to distortions or other optical changes. CNN can process images captured under different lighting conditions and from different perspectives. It still recognizes the typical features of an image. Since the CNN is divided into several local, partially meshed layers, it requires significantly less storage space than fully meshed neural networks. The convolutional layers drastically reduce storage requirements. The training time of the convolutional neural network is also greatly reduced. With the use of modern graphics processors, CNNs can be trained very efficiently. The CNN uses filters to detect and extract features from the input images. First of all, the CNN recognizes simple structures such as lines, color features or edges in the first levels. In the further levels, the convolutional neural network learns combinations of these structures such as simple shapes or curves. With each level, more complex structures can be identified. The data is repeatedly resampled and filtered in the layers.

The image and sensor data 32 are thus preferably processed in the analysis module 410 by means of a convolutional neural network. In addition, a classification module 430 is provided, which contains features M ₁ , M ₂ , . . . , M _n of the wheel movements and the wheel steering angle of a vehicle. In addition, certain states S ₁ , S ₂ , . . . S _{n of} the vehicle can be assigned to these features M ₁ , M ₂ , . Thus, certain values of the wheel steering angles of the wheels of the vehicle 10 can indicate that the vehicle 10 has started to pull out of the parking space and will soon leave the parking space. The speed of the wheel movements can also be used to determine how quickly vehicle 10 will probably leave the parking space. Security levels can in turn be assigned to the states S ₁ , S ₂ , . . . , S _n . These features M ₁ , M ₂ , . . . , M _n and/or _states S ₁ , S ₂ , phase determined by the CNN or they have been predefined and submitted to the classification module.

The image and sensor data 32 processed in this way are integrated into an automated driving function as output data 450 and/or transmitted to the user interface 70 . They can be output there as recommendations for action or warnings to the driver of the vehicle 20 . For example, a warning tone or an optical indication can be output via the user interface 70 . With an automated driving function, the driving speed can also be reduced automatically.

• In einem Schritt S10 werden Bild- und Sensordaten 32 von einer Kamera- und Sensoreinrichtung 30 eines zweiten Kraftfahrzeugs 20 aufgenommen, wobei die Bild- und Sensordaten 32 die Umgebung des Fahrzeugs 20 mit zumindest dem ersten Kraftfahrzeug 10 repräsentieren.
• In einem Schritt S20 werden die Bild- und Sensordaten 32 an eine Datenanalyseeinrichtung 40 weitergegeben, wobei die Datenanalyseeinrichtung 40 ein Erkennungssystem 400 zur Erkennung einer Änderung eines Radlenkwinkels zumindest eines Rades bei dem ersten Kraftfahrzeug umfasst, welches Algorithmen aus dem Bereich der künstlichen Intelligenz (KI) und der maschinellen Bildanalyse verwendet.
• In einem Schritt S30 werden die Bild- und Sensordaten 32 in der Datenanalyseeinrichtung 30 mittels des Erkennungssystems 200 bearbeitet, um mögliche Änderungen des Radlenkwinkels des Rades zu klassifizieren.
• In einem Schritt S40 wird der klassifizierten Änderung des Radlenkwinkels ein Zustand S_j aus einer Menge von definierten Zuständen S₁, S₂, ..., S_n zugeordnet.
• In einem Schritt S50 werden aus dem ermittelten Zustand S_j Ausgabedaten 280 zur weiteren Verwendung in einer automatisierten Fahrfunktion und/oder für eine Benutzerschnittstelle 70 erzeugt.

A method for detecting a change in a wheel steering angle of at least one wheel in a motor vehicle according to the present invention therefore includes the following steps:

• In a step S10, image and sensor data 32 are recorded by a camera and sensor device 30 of a second motor vehicle 20, the image and sensor data 32 representing the surroundings of vehicle 20 with at least first motor vehicle 10.
• In a step S20, the image and sensor data 32 are forwarded to a data analysis device 40, the data analysis device 40 including a detection system 400 for detecting a change in a wheel steering angle of at least one wheel in the first motor vehicle, which algorithms from the field of artificial intelligence (AI) and machine image analysis.
• In a step S30, the image and sensor data 32 are processed in the data analysis device 30 using the recognition system 200 in order to classify possible changes in the wheel steering angle of the wheel.
• In a step S40, a state S _j from a set of defined states S ₁ , S ₂ , . . . , S _n is assigned to the classified change in the wheel steering angle.
• In a step S50, output data 280 for further use in an automated driving function and/or for a user interface 70 are generated from the ascertained state S _j .

A detection system 400 that uses algorithms from the field of artificial intelligence (AI) and machine image analysis can thus analyze images from the surroundings of a second vehicle 20 with regard to changes in the wheel steering angle of another first vehicle 10 in real time. With the present invention, the automatic detection of changes in the wheel steering angle on at least one wheel of the vehicle 10, in particular on a parked vehicle 10, is made possible. The driving behavior of the vehicle 10 can be derived from the classified changes in the wheel steering angle. The result of the analysis is output, for example, as a warning signal on a user interface 70 of the second vehicle 20 if the change in the wheel steering angle indicates a critical driving situation for the second vehicle 20 . The safety when driving can thus be further increased by the present invention.

Reference List

10

first motor vehicle

20

second motor vehicle

30

Camera and sensor setup

32

Image and sensor data

35

RGB camera

37

UV camera

39

IR camera

40

data analysis facility

41

processor

50

storage unit

55

software module

60

Database

70

user interface

100

system

400

detection system

410

analysis module

430

classification module

450

output data

500

computer program product

550

program code

Claims (15)

Method for detecting a change in a wheel steering angle of at least one wheel in a motor vehicle (10) using a data analysis device (40), comprising: - recording (S10) image and sensor data (32) using a camera and sensor device (30) of a second motor vehicle (20), wherein the image and sensor data (32) represent the surroundings of the second motor vehicle (20) with at least the first motor vehicle (10); - Forwarding (S20) of the image and sensor data (32) to the data analysis device (40), wherein the data analysis device (40) has a detection system system (400) for detecting a change in a wheel steering angle of at least one wheel in the first motor vehicle (10), which uses algorithms from the field of artificial intelligence (AI) and machine image analysis; - Processing (S30) the image and sensor data (32) in the data analysis device (40) by means of the detection system (400) in order to classify possible changes in the wheel steering angle of the wheel.; - Assigning (S40) the classified change in the wheel steering angle to at least one state (S _j ) from a set of defined states (S ₁ , S ₂ , ..., S _n ); - Generating (S50) output data (450) from the determined state (Sj) for further use in an automated driving function and/or for a user interface (70). procedure after claim 1 , wherein the processing of the image and sensor data (32) and the generation of output data (450) takes place in real time. procedure after claim 1 or 2 , wherein the recognition system (400) comprises an analysis module (410) and a classification module (430). procedure after claim 3 , where the recognition system (400) uses deep learning with a neural network. procedure after claim 4 , wherein the neural network is designed as a folded neural network (convolutional neural network). Procedure according to one of claims 3 until 5 , wherein the classification module (430) contains features (M ₁ , M ₂ , ..., M _n ) of the wheel movements and the wheel steering angle of a vehicle, which were determined or previously defined in a training phase of the classification module (430). procedure after claim 5 , wherein the image and sensor data (32) are transmitted to the data analysis device (40) by means of a mobile radio connection, with 5G radio modules being used in particular. Procedure according to one of Claims 1 until 7 , wherein the image and sensor device (30) comprises an action camera. System (100) for detecting a change in a wheel steering angle of at least one wheel in a motor vehicle (10), comprising - an image and sensor device (30) of a second motor vehicle (20), the image and sensor data (32) relating to the environment of the second Motor vehicle (20) with at least the first motor vehicle (10); and - a data analysis device (40), wherein the data analysis device (20) comprises a detection system (400) for detecting a change in a wheel steering angle of at least one wheel in the first motor vehicle (10), which algorithms from the field of artificial intelligence (AI) and machine image analysis is used, and the data analysis device (40) is designed to process the image and sensor data (32) using the detection system (400), to classify possible changes in the wheel steering angle of the wheel, to assign at least one state (p _j ) from a set of defined states (S ₁ , S ₂ , ..., S _n ) and output data (450) from the determined state (Sj) for further use in an automated driving function and/or for a user interface (70 ) to create. system (100) after claim 9 , wherein the processing of the image and sensor data (32) and the generation of output data (450) takes place in real time. system (100) after claim 9 or 10 , wherein the recognition system (400) comprises an analysis module (410) and a classification module (430). system (100) after claim 11 , wherein the recognition system (400) uses deep learning with a neural network, which is designed in particular as a folded neural network (convolutional neural network). system (100) after claim 11 or 12 , wherein the classification module (430) contains features (M ₁ , M ₂ , ..., M _n ) of the wheel movements and the wheel steering angle of a vehicle, which were determined or previously defined in a training phase of the classification module (430). System (100) according to one of claims 9 until 13 , wherein the image and sensor data (32) are transmitted to the data analysis device (40) by means of a mobile radio connection, in particular 5G radio modules being used, and/or the image and sensor device (30) comprises an action camera. Computer program product (500), comprising an executable program code (550), which is configured so that when it is executed it performs the method according to any one of Claims 1 until 8th executes

Images