DE102019214713A1 - System for the automated actuation of a vehicle door, vehicle and method - Google Patents

Abstract

The invention relates to a system (1,1a, ..., 1c) for the automated actuation of a vehicle door (4a, ..., 4e), comprising: a memory unit (2), a computing unit (10) which is designed, among other things, for this purpose to perform a voice recognition in order to recognize a voice command of the authorized person to operate the vehicle door (4a, ..., 4e), and which is furthermore designed to transmit the recognized voice command to the relevant vehicle door (4a, ..., 4e) and which is further designed to generate an actuation signal based on the recognized voice command, a control unit (11) which is configured to receive the actuation signal and is further configured to based on the relevant vehicle door (4a, ..., 4e) to operate on the actuation signal. The invention also relates to a method and a vehicle.

Description

The invention relates to a system for the automated actuation of a vehicle door. The invention also relates to a vehicle with such a system and a method.

Most vehicles are currently equipped with central locking as standard, which can be activated by remote control to unlock the vehicle. To open the door, the person holding the ignition key usually has to pull a door handle to open the door.

Vehicle doors can also be opened using voice command data.

DE 10 2016 11 2597 A1 discloses a control system for a vehicle door, comprising: a servo device which is coupled to a hinge arrangement between a door and a body; a sensor; and a controller: receiving a signal from the sensor; Identifying a recognized user from the voice command data within the signal; Interpreting a control command from the voice command data within the signal; and causing the servo device to move the door in response to the control command.

It is an object of the invention to make the operation of vehicle doors, in particular from outside the vehicle, safer and more convenient.

This object is achieved by a system for the automated actuation of a vehicle door with the features of claim 1 and a vehicle with such a system with the features of claim 14.

The object is also achieved by a method having the features of claim 15.

The subclaims list further advantageous measures which can be suitably combined with one another in order to achieve further advantages.

• eine Speichereinheit in der zumindest ein personenspezifisches Authentifizierungskennzeichen wenigstens einer berechtigten Person zum automatisierten Betätigen einer Fahrzeugtür abgelegt ist,
• eine Erfassungseinrichtung zum Erkennen einer Annäherung einer Person an eine Fahrzeugtür, und zum Kennzeichnen der Fahrzeugtür als relevant und zum Erfassen des personenspezifischen Authentifizierungskennzeichens der sich nähernden Person,
• eine Recheneinheit, welche zum Abgleich des erfassten personenspezifischen Authentifizierungskennzeichens mit dem in der Speichereinheit hinterlegten zumindest einen berechtigten personenspezifischen Authentifizierungskennzeichen ausgebildet ist und welche ferner dazu ausgebildet ist, bei Erkennung des personenspezifischen Authentifizierungskennzeichens als berechtigt, eine Spracherkennung durchzuführen, um einen Sprachbefehl der berechtigten Person zum Betätigen der Fahrzeugtür zu erkennen, und welche ferner dazu ausgebildet ist, den erkannten Sprachbefehl zu der relevanten Fahrzeugtür zuzuordnen und welche ferner dazu ausgebildet ist, ein Betätigungssignal auf Basis des erkannten Sprachbefehls zu generieren,
• eine Steuerungseinheit, welche zum Empfangen des Betätigungssignals konfiguriert ist und ferner dazu konfiguriert ist, die relevante Fahrzeugtür basierend auf dem Betätigungssignal zu betätigen.

The task is solved by a system for the automated actuation of a relevant vehicle door, having:

• a memory unit in which at least one person-specific authentication identifier of at least one authorized person for the automated actuation of a vehicle door is stored,
• a detection device for detecting an approach of a person to a vehicle door, and for identifying the vehicle door as relevant and for detecting the person-specific authentication identifier of the approaching person,
• a computing unit which is designed to match the recorded person-specific authentication identifier with the at least one authorized person-specific authentication identifier stored in the memory unit and which is also designed to perform voice recognition when the person-specific authentication identifier is recognized as being authorized in order to operate a voice command from the authorized person the vehicle door, and which is also designed to assign the recognized voice command to the relevant vehicle door and which is also designed to generate an actuation signal on the basis of the recognized voice command,
• a control unit configured to receive the actuation signal and further configured to actuate the relevant vehicle door based on the actuation signal.

By actuating a vehicle door, for example, the opening and closing of the vehicle door can be understood.

Person-specific authentication identifiers denote all identifiers by means of which a person-specific authentication is possible. The authentication ensures that the identity of a person can be proven and verified against the system.

The system can hear, interpret and implement the voice command through the voice command of the authorized person from outside, for example.

With the system according to the invention, authorized persons can verbally give the command to unlock / open the vehicle door when they approach the vehicle. By combining the voice command with the position determination of the authorized person, the desired, relevant vehicle door can be determined and opened. As a result, activation of the automatic vehicle door unlocking and vehicle door opening via, for example, foot actuation / radio key can be dispensed with.

The computing unit can be, for example, the processor or processors of a vehicle.

The control unit can, for example, comprise at least the control unit of the vehicle, which can be designed as an electronic module.

Such a control unit can generate commands to open the relevant vehicle door based on the actuation signal. For this purpose, the control unit is connected to a door opening system, for example.

With the system according to the invention, the vehicle can hear, interpret and implement voice commands from the exterior of the vehicle as well as from the interior of the vehicle. When approaching the vehicle, an authorized person can verbally give the command to operate the relevant vehicle door, for example to unlock / open it. The system according to the invention cleverly uses the direction / position determination of the authorized person so that the voice command can be very short, for example “open door”, since the vehicle recognizes the position of the authorized person and thus the relevant vehicle door. This avoids long and cumbersome commands such as “open rear left side door”. Furthermore, vague terms such as left, which depend on the approaching side of the authorized person (for example from the front; from the rear), are avoided. Opening the wrong driver's door is thus avoided.

With the system according to the invention, for example, an authorized person who approaches the vehicle and carries objects and thus no longer has a free hand can load the vehicle at the desired location (trunk / passenger side ...) without putting the objects aside to have to open the relevant vehicle door.

The person-specific authentication identifier preferably comprises at least the voice profile of the at least one authorized person. The fact that a person's voice is just as unique and unmistakable as their fingerprint comes into play here. The human voice can be understood as the sound generated by the vocal folds of a person and modulated in the mouth, throat and nasal cavities. Each voice profile of a person has special properties so that the identity of the speaker can be assigned to the voice profile and vice versa.

Furthermore, the voice profile preferably comprises at least the voice biometrics. Voice biometrics indicates the characteristics of the voice of its owner. This voice biometry includes, for example, the pitch, the overtones, the average frequency of the first formants, or other measurable voice biometrics with regard to the voice of the owner. Such a voice biometric voice profile offers a protected and simple authentication of an authorized person.

The voice biometric voice profile thus enables reliable speaker authentication in a matter of seconds and is very convenient for authorized persons to use.

The voice profile can be recorded in advance, preferably by the same recording system with which the later voice command is recorded, so that the properties of the recording system are compensated for. The voice profile can then be stored in the memory unit.

Alternatively, voice profiles that are already present can also be transferred to the storage unit, for example by a communication unit, and stored there. These are then assigned to the authorized person.

The authentication of the voice profile and the speech recognition preferably take place by means of one or more artificial neural networks, in particular a deep learning network. These are particularly suitable for speech recognition and speech authentication.

In a further preferred embodiment, the artificial neural network is based on unsupervised or monitored learning.

The detection device preferably has at least one acoustic sensor for detecting the voice command. The at least one acoustic sensor is preferably designed as a directional microphone.

In this way, both the voice command and the direction from which the voice command is coming can be determined. As a result, the relevant vehicle door to which the voice command applies or to which the voice command is assigned can be determined.

The memory unit preferably has a code word, in particular the name of the authorized person, the code word being assigned to the associated voice profile, so that faster authentication is made possible on the basis of the code word.

By means of the code word, the voice command can be assigned more quickly to a - possibly - authorized person or it can be checked more quickly whether this person is authorized. This means that the system reacts faster to a voice command.

The code word is preferably the name of the authorized person. This can lead to a secure authentication, since proper names are usually pronounced with the same emphasis.

The detection device preferably comprises an optical sensor, for example a camera and / or a surround view camera system, which is designed to detect the approach of the at least one authorized person to a vehicle door of the vehicle in order to identify the relevant vehicle door and / or to use a movement trajectory Identification of the relevant vehicle door to be recorded. An existing surround view camera system or an existing camera can be used. The relevant vehicle door, for example, can be easily identified on the basis of such movement trajectories.

In a further preferred embodiment, the detection device comprises an optical sensor. Furthermore, the person-specific authentication identifier can be a part of the body, in particular the face. This is just as unmistakable as the voice and can be combined with the “voice profile” authentication code, for example.

The actuation signal is preferably designed in such a way that the control unit opens and closes the relevant vehicle door within the parameters set in advance. For this purpose, the control unit is connected, for example, to a door opening system / door actuation system for data transmission or for data transmission.

The object is also achieved by a vehicle with a system as described above. The system can, for example, be integrated into the vehicle as part of a driver assistance system. Subsequent installation can also be possible.

• - Bereitstellen einer Speichereinheit in der zumindest ein personenspezifisches Authentifizierungskennzeichen wenigstens einer berechtigten Person zum automatisierten Betätigen einer Fahrzeugtür abgelegt ist,
• - Erkennen einer Annäherung einer Person an eine Fahrzeugtür und kennzeichnen der Fahrzeugtür als relevant und erfassen des personenspezifischen Authentifizierungskennzeichens der sich nähernden Person durch eine Erfassungseinrichtung,
• - Abgleichen des erfassten personenspezifischen Authentifizierungskennzeichens mit dem in der Speichereinheit hinterlegten zumindest einen berechtigten personenspezifischen Authentifizierungskennzeichen und durchführen einer Spracherkennung bei Erkennung des personenspezifischen Authentifizierungskennzeichens als berechtigt, um einen Sprachbefehl der berechtigten Person zum Betätigen der Fahrzeugtür zu erkennen,
• - Zuordnen des Sprachbefehls zu der relevanten Fahrzeugtür und
• - Generieren eines Betätigungssignals auf Basis des erkannten Sprachbefehls,
• - Empfangen des Betätigungssignals durch eine Steuerungseinheit und betätigen der relevanten Fahrzeugtür basierend auf dem Betätigungssignal.

The object is also achieved by a method for the automated actuation of a vehicle door with the following steps:

• - Provision of a memory unit in which at least one person-specific authentication identifier of at least one authorized person for the automated actuation of a vehicle door is stored,
• - Detecting an approach of a person to a vehicle door and marking the vehicle door as relevant and recording the person-specific authentication identifier of the approaching person by a detection device,
• - Matching of the recorded person-specific authentication identifier with the at least one authorized person-specific authentication identifier stored in the memory unit and performing voice recognition when the person-specific authentication identifier is recognized as authorized in order to recognize a voice command from the authorized person to operate the vehicle door,
• - Assigning the voice command to the relevant vehicle door and
• - Generating an actuation signal based on the recognized voice command,
• - Receipt of the actuation signal by a control unit and actuation of the relevant vehicle door based on the actuation signal.

The advantages of the system can be transferred to the process.

• 1: ein erfindungsgemäßes System in einer ersten Ausgestaltung,
• 2: eine Anwendung des erfindungsgemäßen Systems,
• 3: eine weitere Ausgestaltung des erfindungsgemäßen Systems,
• 4: eine weitere Ausgestaltung des erfindungsgemäßen Systems,
• 5: eine weitere Ausgestaltung des erfindungsgemäßen Systems,
• 6: ein erfindungsgemäßes Verfahren.

Further features, properties and advantages of the present invention emerge from the following description with reference to the accompanying figures. They show schematically:

• 1 : a system according to the invention in a first embodiment,
• 2 : an application of the system according to the invention,
• 3 : a further embodiment of the system according to the invention,
• 4th : a further embodiment of the system according to the invention,
• 5 : a further embodiment of the system according to the invention,
• 6th : a method according to the invention.

1 shows a vehicle 3 with different vehicle doors 4a , 4b , 4c , 4d , 4e with a system according to the invention 1 in a first embodiment. The trunk door is also used as a vehicle door 4e designated.

The system according to the invention 1 has a storage unit 2 in which several person-specific authentication identifiers, which are designed here as voice profiles, are stored for a respective authorized person.

Each voice profile is based on a voice biometry that indicates the properties of the user's voice. This voice biometry includes, for example, the pitch, the overtones, the average frequency of the first formants, or other measurable ones Voice biometrics in relation to the user's voice. Such a voice biometric voice profile offers a protected and simple authentication of an authorized person. The voice biometric voice profile thus enables reliable speaker authentication in a matter of seconds and is very convenient to use for the calculated people.

A voice biometric profile usually contains more than 100 unique characteristics of a person related to his or her voice. These are determined on the one hand by physical factors such as the length of the vocal tract and / or the length of the nasal passage. Fluctuations in the voice, such as with a cold, usually do not cause the system to fail 1 , since there are enough other special features that can contribute to authentication.

The voice profile can be saved in the vehicle beforehand 3 be included. The voice profile is preferably made with the same sensor system on the vehicle 3 recorded, which later also receive the respective voice commands of the authorized persons. This allows special features in the sensor system to be compensated for. Alternatively, existing voice profiles that have already been recorded for smart devices, for example, can be adopted. This can be done, for example, through a wireless communication interface on the vehicle 3 be imported.

Furthermore, a detection device is provided which here has one or more optical sensors and / or one or more optical cameras 6th comprises, for detecting an approach of a person to a vehicle door 4a, ..., 4e as well as to identify the relevant vehicle door 4a, ..., 4e .

The optical sensors are here, for example, on each side of the vehicle 7a , 7b arranged to detect the side view of the vehicle 3 looked like. The optical camera 6th Can be used as a front camera to capture the front 8th and the back 9 be trained. However, a different arrangement and a different number of optical sensors / cameras can also be provided, or an already existing surround view system can be used. The optical sensors and the camera 6th recognize an approach or even movement trajectory of an approaching person to the vehicle 3 and recognize from it the relevant vehicle door 4a, ..., 4e .

Furthermore, the detection device comprises 1 multiple microphones 5a , 5b for the directed detection of the voice command of the approaching person.

There are the microphones 5a , 5b preferably designed as directional microphones and each on both sides of the vehicle 7a , 7b or for example on the front 8th and at the back 9 of the vehicle 3 arranged. Optionally, fewer or more microphones can also be arranged at different locations. For example, a microphone which is arranged on the roof antenna is sufficient to detect the direction and the voice command of the approaching person. The voice command is the relevant vehicle door 4a, ..., 4e assigned.

The system also includes 1 another arithmetic unit 10 which are used to compare the recorded voice profile of the voice command with those in the memory unit 2 stored authorized voice profiles.

This comparison can be carried out, for example, by means of artificial intelligence, in particular an artificial neural network, the artificial neural network learning to pay attention to different aspects during training.

The artificial neural network can be based on monitored machine learning or unsupervised machine learning; in particular, the comparison can be carried out on the basis of deep learning methods / algorithms. These are particularly suitable for speech recognition and speech authentication.

If the voice profile is recognized as authorized, voice recognition is carried out in order to receive a voice command to operate the relevant vehicle door 4a, .., 4e to recognize. This speech recognition can also be carried out on the basis of artificial neural networks, in particular deep learning methods. The arithmetic unit then generates 10 an actuation signal based on the recognized voice command and the recognized relevant vehicle door 4a, ..., 4e . If the authorized person approaches as in 1 the passenger door, here as a vehicle door 4b and if, for example, the command “open door off” is spoken, an actuation signal is generated, which a control unit 11 prompts you to take the necessary actions to open the passenger door.

2 shows a further application of the system according to the invention 1 . Here the authorized person approaches the rear from behind 9 of the vehicle 3 and thus to the trunk with the trunk door, here as a vehicle door 4e , marked, on.

This is done by the system 1 registered. With the command "Open door", the trunk door is now opened automatically.

With the system according to the invention 1 can the vehicle 3 Hear, interpret and implement voice commands from outside the vehicle as well as from inside the vehicle.

With the system according to the invention 1 the authorized person can approach the vehicle 3 verbally give the command to unlock / open. With the system according to the invention 1 the direction determination is cleverly used so that the voice command can be very brief "open the door" because the vehicle 3 the position of the authorized person and the relevant vehicle door 4a, ..., 4e recognizes.

This avoids long, cumbersome commands such as “open rear left side door”. Furthermore, vague terms such as left, which depend on the approaching side of the authorized person (from the front from the back), are avoided. Opening the wrong vehicle door 4a, ..., 4e is thus avoided.

With the system according to the invention 1 for example, an authorized person who is driving the vehicle 3 approaches and carries objects, and thus no longer has a free hand, the vehicle 3 Load in the desired place (trunk / passenger side ...) without having to put the items to one side around the relevant vehicle door 4a, ..., 4e to open. The objects can be a purchase / child, for example.

The vehicle doors 4a, .., 4e can for example be designed as a stop or pivot door or as a sliding door. Under operating the vehicle door 4a, .. ,, 4e can be understood as the relevant, desired vehicle door 4a, .., 4e is unlocked and is moved, for example, around a hinge or along a rail that opening is accomplished within preset parameters, for example, in order to get in or load the objects. The authorization can open or close the relevant vehicle door 4a, .., 4e include.

Using the voice biometric profile enables one-to-one recognition, similar to a person's genetic fingerprint. Such a voice biometric voice profile causes a protected authentication of the authorized person by the system 1 .

3 shows a further embodiment of a system 1a according to the invention. This shows as already at 1 the storage unit 2 in which the person-specific authentication IDs of several authorized persons are stored. The person-specific authentication identifiers are also designed here as a voice biometric voice profile.

Furthermore, the system 1 a according to the invention also comprises a detection device, which here as microphones 5a , 5b , in particular directional microphones, which, due to their direction-dependent microphone sensitivity, allow interference signals coming from lateral and rearward directions to be excluded.

This allows the relevant vehicle door 4a, ... 4e can only be determined on the basis of the direction from which the voice or the voice command is coming. The inventive system 1a has as in 1 an arithmetic unit 10 to receive the recorded voice commands with a voice biometric profile.

The arithmetic unit is also used 10 to compare the recorded voice biometric profile with those in the storage unit 2 stored authorized voice biometric voice profiles and to carry out - if the recorded voice biometric voice profile is recognized as authorized - the voice recognition of the voice command. Based on the recognized voice command and the relevant vehicle door 4a, ..., 4e an actuation signal is generated.

The system 1a according to the invention further comprises a control unit 11 configured to receive the actuation signal and further configured to open the relevant vehicle door 4a, ..., 4e to open or close based on the actuation signal.

4th comprises a further embodiment of a system 1b according to the invention. The system 1b according to the invention has the memory unit 2 in which the person-specific authentication IDs of the authorized persons are stored. The person-specific authentication identifiers are designed here as face recognition.

Face recognition describes the analysis of the expression of visible features in the area of the head, given by a geometric arrangement of, for example, eyes, mouth and nose, the upper edges of the eye sockets, the areas around the cheekbones and the side areas of the mouth and texture properties of the skin as well as the assignment of the face of a particular person. Face recognition can be counted among the biometric processes. A reference image or a plurality of reference images can be stored beforehand in the storage unit 2 can be saved.

The detection system also includes a surround view camera 12th with which an approaching person can be recorded. The movement trajectories of the approaching person can also be seen using the surround view camera 12th and thereby a relevant vehicle door 4a , ..., 4e be identified.

The detection system also includes directional microphones 5a , 5b for detecting the voice command of the approaching person. On the basis of the direction of the voice command, the relevant vehicle door can, for example, alternatively or additionally 4a, ..., 4e be determined.

The system 1b according to the invention also includes the computing unit 10 . This is designed to use the captured face to make a comparison with that in the storage unit 2 to match stored reference images from authorized persons.

This can also be done, for example, with the aid of an artificial neural network. The artificial neural network is preferably designed as a convolutional neural network. These are mainly used for processing image data. At their core, they are classic neural networks, but they have a convolution and a pooling layer. The convolution layer processes the recorded facial data several times in a row, but always only a section of it; the pooling layer reduces the detail data to reduced information.

The relevant vehicle door 4a, .., 4e can in this case through the movement trajectory recorded by the surround view camera 12th has been detected, or by the direction of the directional microphones 5a , 5b recorded voice command.

Then, when an authorized person is recognized, the voice command is analyzed by means of artificial intelligence and the actuation signal is generated. This becomes the control unit 11 transmits which actions are generated for executing the actuation signal.

5 shows a further embodiment of a system 1c according to the invention.

As in 1 the storage unit 2 in which the person-specific authentication identifiers of the authorized persons are stored, the person-specific authentication identifiers being designed as voice profiles. Each voice profile is based on a voice biometry that indicates the properties of the user's voice.

Furthermore, in addition to the voice profile, a code word, for example the name of the authorized person, is stored and linked to the person-specific authentication identifier and to the corresponding authorized person.

By means of the code word, the voice command can be assigned more quickly to a - possibly - authorized person or it can be checked more quickly whether this person is authorized. This means that the system 1c reacts faster to a voice command.

The code word can also be combined with one of the other embodiments.

Authentication by means of face or body recognition can also be used in combination with authentication by means of a voice profile.

6th shows the method according to the invention.

In a first step S1 are in a storage unit 2 ( 1 ) biometric voice profiles, which are linked to an authorized person, are saved.

Alternatively or in addition, for example, the face / or facial features can also be stored and / or a code word or the voice profile of a code word, for example the name, can be stored. In this context, stored means that the biometric voice profile, the code word and / or the face / facial features are always associated with a person who is referred to as an authorized person.

This person is authorized to use vehicle doors 4a, ..., 4e ( 1 ) to operate, for example to open.

In a second step S2 becomes an approach of a person to a vehicle door 4a, ..., 4e ( 1 ) recognized and this vehicle door 4a, ..., 4e ( 1 ) classified as relevant.

This can be accomplished, for example, with the aid of the optical sensor and the detection of the person's movement trajectory.

Furthermore, this can also be achieved by an acoustic sensor, for example a directional microphone. This assigns a direction to a received voice. This allows the relevant vehicle door 4a, ..., 4e ( 1 ) be recognized.

In a third step S3 the biometric voice profile and / or the face / facial features and possibly also the code word as a person-specific authentication identifier is recorded either by the optical sensor or by the directional microphone.

In a fourth step S4 the biometric voice profile and / or the face / facial features and possibly also the code word with the corresponding in the memory unit 2 ( 1 ) matched authorized person-specific authentication identifier.

When an authorization is recognized, the voice command of the authorized person is recognized.

In a fifth step S5 the recognized voice command is sent to the relevant vehicle door 4a, ..., 4e ( 1 ) assigned.

In a sixth step S6 an actuation signal is generated based on the recognized voice command and the recognized relevant vehicle door 4a, ..., 4e ( 1 ) generated.

In a seventh step S7 the generated actuation signal is sent to the control unit 11 transmitted and received by this. The control unit generates based on the actuation signal 11 now an actuation of the relevant vehicle door 4a, ..., 4e ( 1 ).

List of reference symbols

1

system

2

Storage unit

3

vehicle

4a, .., 4e

Vehicle door

5a, 5b

Directional microphone

6th

camera

7a, 7b

Vehicle side

8th

front

9

back

10

Arithmetic unit

11

Control unit

12th

Surround view camera

S1,, S7

Procedural steps

Claims (15)

System (1,1a, .., 1c) for the automated actuation of a vehicle door (4a, ..., 4e), comprising: a memory unit (2) in which at least one person-specific authentication identifier of at least one authorized person for the automated actuation of a vehicle door (4a, ..., 4e) is stored, a detection device for detecting an approach of a person to a vehicle door (4a, ..., 4e) and for identifying the vehicle door (4a, ..., 4e) as relevant and for detecting the person-specific authentication identifier of the approaching person, a computing unit (10) which is designed to match the recorded person-specific authentication identifier with the at least one authorized person-specific authentication identifier stored in the memory unit (2) and which is also designed to perform voice recognition when the person-specific authentication identifier is recognized as authorized to perform a Recognize voice command of the authorized person to operate the vehicle door (4a, ..., 4e), and which is also designed to assign the recognized voice command to the relevant vehicle door (4a, ..., 4e) and which is also designed to generate an actuation signal based on the recognized voice command, a control unit (11) which is configured to receive the actuation signal and is further configured to actuate the relevant vehicle door (4a, ..., 4e) based on the actuation signal. System (1,1a, .., 1c) according to Claim 1 , characterized in that the person-specific authentication identifier comprises at least the voice profile of the at least one authorized person. System (1,1a, .., 1c) according to Claim 2 , characterized in that the voice profile comprises at least the voice biometrics. System (1,1a, .., 1c) according to Claim 2 or 3 , characterized in that the authentication of the voice profile as well as the speech recognition takes place by means of one or more artificial neural networks. System (1,1a, .., 1c) according to Claim 4 , characterized in that the artificial neural network is based on unsupervised or monitored learning. System (1,1a, .., 1c) according to one of the preceding claims, characterized in that the detection device has at least one acoustic sensor for detecting the voice command. System (1,1a, .., 1c) according to Claim 6 , characterized in that the at least one acoustic sensor is designed as a directional microphone (5a, 5b). System (1,1a, .., 1c) according to one of the preceding claims, characterized in that the memory unit (2) has a code word, the code word being assigned to the associated voice profile, so that faster authentication is made possible on the basis of the code word . System (1,1a, .., 1c) according to Claim 8 , characterized in that the code word is the name of the authorized person. System (1,1a, .., 1c) according to one of the preceding claims, characterized in that the detection device comprises an optical sensor which is designed to allow the at least one authorized person to approach a vehicle door (4a, ..., 4e ) of the vehicle (3) to identify the relevant vehicle door (4a, ... 4e) and / or to record a movement trajectory to identify the relevant vehicle door (4a, ..., 4e). System (1,1a, .., 1c) according to one of the preceding claims, characterized in that the detection device comprises an optical sensor and the person-specific authentication identifier is a part of the body. System (1,1a, .., 1c) according to Claim 11 , characterized in that the body part is the face. System (1,1a, .., 1c) according to one of the preceding claims, characterized in that the actuation signal is designed such that the control unit (11) opens and closes the relevant vehicle door (4a, ..., 4e) within the parameters set in advance. Vehicle (3) with a system (1,1a, .., 1c) according to one of the preceding claims. Method for the automated actuation of a vehicle door (4a, ..., 4e) characterized by the steps: - Provision of a memory unit (2) in which at least one person-specific authentication identifier of at least one authorized person for the automated actuation of a vehicle door (4a, ..., 4e) ) is stored, - Detection of a person approaching a vehicle door (4a, ..., 4e) and marking the vehicle door (4a, ..., 4e) as relevant and recording the person-specific authentication identifier of the approaching person by a detection device, - Matching of the recorded person-specific authentication identifier with the at least one authorized person-specific authentication identifier stored in the storage unit (2) and performing a voice recognition when the person-specific authentication identifier is recognized as authorized in order to receive a voice command from the authorized person to operate the vehicle gtür (4a, ..., 4e), - assignment of the recognized voice command to the relevant vehicle door (4a, ..., 4e) and - generation of an actuation signal based on the recognized voice command, - reception of the actuation signal by a control unit ( 11) and actuate the relevant vehicle door (4a, ..., 4e) based on the actuation signal.

Images