DE102019110986A1 - Method and device for user-specific setting of vehicle parameters of a vehicle and vehicle - Google Patents

Abstract

For user-specific setting of vehicle parameters of a vehicle (1), at least one sensor system (3) of the vehicle (1) detects that a person (2) is in a predetermined area around the vehicle (1) and the vehicle (1 ) wants to use, - the person is recorded by means of a camera system (3c, 3d) which comprises at least one optical stereo camera (3c) and / or a thermal camera (3d), - using at least two images from the stereo camera (3c) and / or one Image of the thermal camera (3d) determined person-specific features and set the vehicle parameters based on the person-specific features.

Description

The invention relates to a method and a device for user-specific setting of vehicle parameters of a vehicle, as well as a vehicle.

Vehicles, in particular motor vehicles such as cars, motorcycles, trucks and buses, as a rule, have a large number of adjustable parameters which can be set with regard to the respective persons using the vehicle. The setting is made for the respective user, be it a vehicle driver or another passenger traveling with the vehicle or seated in the vehicle, according to their wishes, needs or person-specific properties or features. For example, in today's vehicles, the positions and angles of rear-view mirrors, or preferred radio program transmitters, can be set in a user-specific and, in particular, driver-specific manner. The setting of the respective seat, especially with regard to mechanical parameters such as the longitudinal position, which e.g. significantly determines the legroom, or various angles such as the angle of inclination of the backrest, the angle of inclination of the seat, or height position or angle of inclination of the headrest.

The parameters can conventionally be set by the respective user himself, for example by means of levers or rotary wheels, with the setting possibly being supported by means of electric drives. However, the setting can also be automated, for example in that corresponding values are stored electronically as a user profile and all the corresponding values are automatically read from the memory with a simple push of a button and used to adjust the seats by means of electric drives. Instead of pressing a button by the user, it can also be provided that the user is automatically recognized by sensors on the vehicle by means of a recognition method, for example by face recognition, and the corresponding settings are then made in the vehicle.

In the DE 10 2010 028 580 A1 describes a vehicle with a manually operated input unit via which an operator can enter vehicle occupant data such as body height, knee height, hip joint height, shoulder height and cervical vertebra length into the memory of a control unit. Algorithms or empirically determined map data are provided in a control unit, by means of which ergonomically optimal settings for vehicle or seat components are calculated from the vehicle occupant data. The actuators required for adjusting seat and / or vehicle components are then automatically activated via the control unit, depending on the calculation results.

In the WO 02/08022 A2 is described to identify a vehicle occupant by facial recognition for the automatic, user-specific setting of vehicle components.

Motor vehicles can increasingly be operated in an automated manner, with a distinction being made between different degrees of automation, in particular with regard to the longitudinal and lateral guidance of the vehicle, right up to fully autonomous driving. In the publication Todd Litman, Autonomous Vehicle Implementation Predictions - Implications for Transport Planning, Victoria Transport Policy Institute (2019), available at http://www.vtpi.org/avip.pdf (as of April 25, 2019), corresponding degrees of automation and Described possibilities that open up for passengers in the use of such vehicles.

The contents of the above publications are hereby incorporated into the present description by reference.

The object of the invention is to enable, in an automated manner, a setting of parameters for vehicle components that is as well adapted as possible to the needs of a vehicle user in a vehicle.

This object is achieved by the invention specified in the independent patent claims. Advantageous embodiments of the invention are specified in the dependent claims.

• - mittels mindestens eines Sensorsystems des Fahrzeugs detektiert wird, dass sich eine Person, in einem vorgegebenen Bereich um das Fahrzeug befindet und das Fahrzeug nutzen will,
• - die Person mittels eines Kamerasystems, das zumindest eine optische Stereokamera und/oder eine Thermokamera umfasst, aufgenommen wird,
• - anhand mindestens zwei Bildern der Stereokamera und/oder eines Bildes der Thermokamera personenspezifische Merkmale bestimmt werden und
• - anhand der personenspezifischen Merkmale die Fahrzeugparameter eingestellt werden.

According to the invention, for the user-specific setting of vehicle parameters of a vehicle, it is provided that

• - it is detected by means of at least one sensor system of the vehicle that a person is located in a predetermined area around the vehicle and wants to use the vehicle,
• - the person is recorded by means of a camera system that includes at least one optical stereo camera and / or a thermal camera,
• - Using at least two images of the stereo camera and / or one image of the thermal camera, person-specific features are determined and
• - The vehicle parameters are set based on the person-specific features.

The optical stereo camera is set up to generate a three-dimensional image of a three-dimensional scene. The optical stereo camera includes in particular two camera sensors that generate two partial images or half images for the three-dimensional scene by means of optical paths that are separate from one another. For this purpose, the stereo camera can in particular comprise two lenses attached next to one another and enable the simultaneous recording of the two, so-called stereoscopic half images, to be generated when the trigger is triggered.

Within the scope of the present invention, the optical stereo camera can also be used in particular to record the face of the person who is located in the predetermined area around the vehicle in a three-dimensional representation for face recognition in order to determine the person-specific features.

The thermal camera can be used to generate thermographic images of a scene. It is sensitive to infrared radiation. In the context of the present invention, the thermal camera can in particular be used to record a thermographic image of a person. Such a recording can be more suitable than an optical recording, in particular in poor lighting conditions, for example at night. Furthermore, a thermographic recording makes it possible to record the blood flow condition of a person and from this, for example, to adjust vehicle parameters specifically to this condition, for example to automatically activate a heating device, seat heater, cooling device or massage device, in particular before the person gets into the vehicle or the vehicle starts moving.

With regard to the present invention, an optical camera, in particular a stereo camera, can have the advantages over a thermal camera that it has a higher image resolution and / or a higher depth resolution, that the recorded image is independent of the ambient temperature and that it has colors in the recorded scene can detect in the visible spectrum. Especially for taking pictures in good lighting conditions e.g. During the day, the image signals recorded by an optical stereo camera of a scene can therefore be better suited to recognizing a person and / or their approach to a vehicle and possibly the person-specific features than the corresponding image signals recorded by a thermal camera.

The sensor system can in particular comprise a camera system. The camera system can in particular comprise the optical stereo camera, the thermal camera and / or at least one further optical camera and / or thermal camera. The sensor system can also comprise other sensor units, such as a microphone, one or more ultrasonic sensors and / or a radar system with one or more radar sensors. Data from the sensor system and / or the camera system are processed in at least one data processing unit in order to recognize whether a person is in a predetermined area around the vehicle, to determine person-specific features and / or to use these features to determine the vehicle parameters for setting.

According to the invention, it was recognized that a combination of several sensor systems, which include an optical stereo camera and a thermal camera, makes it possible to recognize whether a person is approaching a vehicle and, if necessary, can also determine that the person wants to use the vehicle .

Use of a vehicle in the context of the present invention is understood to mean that at least one vehicle function is activated, in particular a locking device of the vehicle is activated and / or the vehicle is started and / or that a person gets into the vehicle. In the context of the use of a vehicle, non-driving-related, so-called non-driving forms of use, such as e.g. Reading, listening to music or the radio, watching TV, watching films, playing games or doing office work, and especially resting and relaxing in the vehicle. Such non-driving forms of use include in particular non-driving activities, i.e. Activities that are not aimed at driving the vehicle. In particular, they can be carried out in a highly automated or fully automated (autonomous) vehicle at least for a period of time specified by the automation level of the vehicle and in particular without a time limit, even if the passenger is assigned the role of a vehicle driver, or even if there is no driver in the vehicle is provided. Examples of such automation levels and further examples of non-driving forms of use are also described in the publication by Todd Litman mentioned above.

Vehicle components in the context of the present invention are in particular all components of a vehicle that are associated with one or more vehicle seats. Such components include in particular electromechanical components such as electric motors, electric linear drives and air conditioning components such as heating elements. You can, for example, to adjust a seat position in the longitudinal and / or transverse direction to a vehicle axis, in particular the vehicle longitudinal axis, a seat height and / or an inclination of seat elements such as a backrest, a seat surface or a seat length adjustment can be provided in the vehicle. The components can, in particular, also include components that affect a reclining position of the vehicle user and, for example, adjust seat elements such as a footrest, a headrest or a lumbar support. Such a component can also be a massage device integrated in particular in a seat.

With the invention, ergonomically optimal settings for vehicle or seat components can be made for the user and in particular the passenger of the vehicle.

In particular, a parameter of a vehicle seat can be set as the vehicle parameter. As a vehicle seat in the sense of the present invention, any type of device can be understood that is suitable for receiving a vehicle passenger including a vehicle driver while driving. This can be a conventional vehicle seat for one or more people, but it can also be some other furniture-like device, for example in the form of a couch or lounger. The vehicle seat and / or parts thereof can in particular be movably mounted in the vehicle, for example rotatable, displaceable along one or more axes and / or inclinable.

According to an advantageous exemplary embodiment of the invention, the sensor system is used to detect whether a person is approaching the vehicle. Furthermore, data relating to the person are recorded by means of the camera system and the person-specific features are determined by means of the data recorded by the camera system.

With regard to the detection of whether a person is approaching the vehicle, the sensor system fulfills in particular the function of a movement sensor and can have one or more corresponding sensors, for example an infrared or ultrasonic sensor. To detect whether a person is approaching the vehicle and / or to detect whether the person wants to use the vehicle, at least one microphone can also be provided on the vehicle and a system for speech and / or voice recognition coupled to it. In this way, for example, a person approaching the vehicle can use a code word such as e.g. “Hello BMW, I am Andreas” in front of the vehicle. Alternatively, the person can also use radio transmission technology such as RFID or NFC technology, in particular using a portable device such as e.g. of a smart phone to indicate to the vehicle when corresponding sensors, e.g. for RFID, NFC or the like. Are provided. It can also be recognized and possibly used as a process parameter, for example, that a person is carrying a backpack or a suitcase.

Using a self-recognition as described above, the determination of the person-specific parameters using the stereo and / or thermal camera within the scope of the invention can also be checked for plausibility by comparing these two recognition results, in particular with regard to a specifically recognized person . In the event of a discrepancy, only the result of the camera-based determination is preferably used, because e.g. the spoken code word and / or a recognized voice can relatively easily be misused or imitated by another person.

The camera system including the optical stereo camera and / or the thermal camera is preferably only switched on after it has been recognized that a person wants to use the vehicle.

Alternatively, the optical stereo camera and / or the thermal camera can also be used directly to detect whether a person is approaching the vehicle.

According to a further preferred exemplary embodiment of the invention, anthropometric data of the person are determined as person-specific features by means of data that were acquired by the camera system, in particular image data acquired by the optical stereo camera and / or the thermal camera. As a result, vehicle parameters for setting the vehicle seat intended for the person can be optimally adjusted to the needs of the person in terms of horizontal and vertical positions, lengths (for example a seat extension or a variable footrest) and the angle of inclination, especially for the backrest and seat. Anthropometric data of a person are in particular measurements of lengths and angles of the whole body or of body parts such as e.g. the body size, the knee height or the position of the body axes. Body types can be differentiated and population groups divided into percentiles that can be assigned to specific body types, in particular body sizes.

In a further preferred exemplary embodiment of the invention, situation-specific attributes that are used for setting the vehicle parameters are determined by means of data that have been recorded by the sensor system, in particular by the camera system. Such an attribute can be, for example, a season of the year and can be based on a recognized, worn by the person, Seasonal clothing such as shorts (summer) or a thick coat (winter) can be determined. In order to increase the accuracy, this result can, for example, also be compared with an outside temperature recorded by a thermal sensor and / or using the current date and location. On the basis of this attribute, for example, temperature parameters in the vehicle can be set, or the setting in a seat can be adjusted so that the seat position is set slightly lower due to thicker clothing such as a coat, so that the passenger, in particular the vehicle driver, is as independent of the height as possible to enable the eye-dependent field of vision. Depending on the clothing, one or more seat side bolsters can also be placed in a somewhat wider position in order to provide the passenger with a clothing-independent seating feeling that offers secure lateral support and is comfortable.

An attribute can, for example, also be determined with regard to the occupation of the passenger to be expected during the journey and used to set the vehicle parameters. For example, on the basis of a typical work item recognized by the camera system and carried by the passenger, it can be determined that the passenger is likely to carry out a work activity. Such a work item can for example be a briefcase or a portable computer for office work. Accordingly, e.g. Vehicle parameters for seat adjustment are controlled in such a way that a typical office chair shape is assumed for a relatively upright sitting position. If, on the other hand, it is recognized that the passenger is only carrying a conventional or electronic book (e-book reader), an attribute can be determined which characterizes relaxation and the seating parameters can then be set for a relatively relaxed lying position according to this attribute. For the most reliable assignment of the attributes, several pieces of information recognized by sensors can be used, for example information about the clothing of the passenger according to a characterization of “office clothing” or “casual clothing”.

A parameter of a vehicle seat that relates to a locking system of the vehicle can also be set as a vehicle parameter. In particular, the invention can be used to automatically open a vehicle after recognizing a specific, authorized person. Furthermore, it can be provided, in particular, that a seat that has already been assigned to the passenger in advance in the vehicle or assigned in the course of the person recognition process is adjusted before the locking system unlocks the vehicle. A time estimate and / or time recording can be carried out to determine how long the corresponding seat adjustment lasts. If a predetermined maximum duration value is exceeded, e.g. The vehicle can be unlocked automatically for 2 seconds, 5 seconds, 10 seconds or 30 seconds before the seat is fully adjusted. In this case it can also be provided that the seat is only adjusted in accordance with the automatically determined vehicle parameters after the passenger has got into the seat. It can again be provided that the seat is not adjusted until the user has confirmed the automatic adjustment separately, for example by pressing a button or by voice command.

For example, the absolute body size of the person can be determined from the data of the person to be recorded that is recorded by the camera system and in particular from the recorded anthropometric data. The stereo camera and a three-dimensional image of the person obtained therefrom can be used for this purpose. If, for example, all joints are visible in the picture, the absolute height of the person can be calculated using the Euclidean distance without the need for a reference measurement.

If, on the other hand, one or more joints and / or extremities are partially or completely not visible in the image, then a simulation for the one or more fully or partially invisible extremities can be carried out using one or more visible extremities. For example, the missing right leg can be simulated by the existing left leg. In addition, if more than two joints or extremities are not visible, at least one missing joint or at least one extremity can be simulated using methods known per se according to Leonardo Da Vinci and / or Neufert and / or an estimate (e.g. height by estimation) will. Such methods can in particular also be combined with one another.

The Neufert method is described, for example, in Abu-taieh, Evon & Alhaj, Abdullah & Abu-Tayeh, Alia & El Sheikh, Asim & Ghatasheh, Nazeeh, Taxonomy of Image File Formats, 74-81. 10.1109 / ECONF.2013.62 (2013).

1. 1. Die Länge der ausgestreckten Arme entspricht der Größe eines Mannes.
2. 2. Gesicht: vom Haaransatz bis zum Kinn ist es 1/10 der Körpergröße eines Mannes.
3. 3. Kopf: die Länge von unterhalb des Kinns bis zur Kopfspitze beträgt 1/8 der Körpergröße eines Mannes.
4. 4. Von oberhalb der Brust bis zur Kopfspitze entspricht es 1/6 der Körpergröße eines Mannes.
5. 5. Von oberhalb der Brust bis zum Haaransatz ist es 1/7 der Körpergröße eines Mannes.
6. 6. Schultern: Die maximale Breite der Schultern beträgt 1/4 der Körpergröße eines Mannes.
7. 7. Die Länge von der Brust bis zum Kopf entspricht 1/4 der Körpergröße eines Mannes.
8. 8. Unterarm: Der Abstand vom Ellenbogen zur Handspitze beträgt 1/4 der Körpergröße eines Mannes.
9. 9. Oberarm: Der Abstand vom Ellbogen zur Achselhöhle beträgt 1/8 der Körpergröße eines Mannes.
10. 10. Hand: Die Länge der Hand entspricht 1/10 der Höhe eines Mannes.
11. 11. Die Wurzel des Penis ist auf halber Höhe eines Mannes.
12. 12. Die Fußlänge beträgt 1/6 der Körpergröße eines Mannes.
13. 13. Bein: die Länge von unterhalb des Fußes bis unter das Knie entspricht 1/4 der Körpergröße eines Mannes.
14. 14. Oberschenkel: Die Länge von unterhalb des Knies bis zur Peniswurzel beträgt 1/4 der Körpergröße eines Mannes.
15. 15. Gesicht: Die Abstände von der Unterseite des Kinns zur Nase, von der Nase zu den Augenbrauen und von den Augenbrauen zu der Haarlinie betragen jeweils 1/3 des Gesichts.

Leonardo Da Vinci's method is based on a postulate about human proportions, the so-called Vitruvian man. Relationships between the dimensions of body segments are described and 15 rules of proportion are provided:

1. 1. The length of the outstretched arms is the size of a man.
2. 2. Face: from the hairline to the chin it is 1/10 the height of a man.
3. 3. Head: the length from below the chin to the tip of the head is 1/8 the height of a man.
4. 4. From above the chest to the tip of the head it is 1/6 the height of a man.
5. 5. It is 1/7 the height of a man from above the chest to the hairline.
6. 6. Shoulders: The maximum width of the shoulders is 1/4 the height of a man.
7. 7. The length from chest to head is 1/4 the height of a man.
8. 8. Forearm: The distance from the elbow to the tip of the hand is 1/4 the height of a man.
9. 9. Upper arm: The distance from the elbow to the armpit is 1/8 the height of a man.
10. 10. Hand: The length of the hand is 1/10 the height of a man.
11. 11. The root of the penis is halfway up a man.
12. 12. Foot length is 1/6 the height of a man.
13. 13. Leg: the length from below the foot to below the knee corresponds to 1/4 the height of a man.
14. 14. Thigh: The length from below the knee to the base of the penis is 1/4 the height of a man.
15. 15. Face: The distances from the bottom of the chin to the nose, from the nose to the eyebrows and from the eyebrows to the hairline are each 1/3 of the face.

Furthermore, within the scope of the invention, for example, the body size can also be approximately calculated by means of a simulation. For this purpose, for example, a reference value can be formed from measured values, which are obtained in particular with high-precision measurement methods, for forearm length, head length, knee height, wingspan, the so-called 52% method, forearm to fingertips, shoulder roof to navel or crown and / or ankle height.

Within the scope of the invention, recorded dimensions of individual body parts can also be further processed by computer and, in particular, added, subtracted and divided.

• - Verwendung von Referenzdaten, die mittels Verfahren der künstlichen Intelligenz ermittelt sind,
• - Gesichtserkennung, insbesondere unter Verwendung einer Gitterprojektion.

At least one of the following methods can advantageously be used to determine the person-specific characteristics and in particular the anthropometric data:

• - Use of reference data that are determined using artificial intelligence methods,
• - Face recognition, in particular using a grid projection.

In this case, empirical methods can in turn be used in which reference data are determined, from which calculation algorithms and in particular corresponding parameters for these algorithms are formed. Artificial intelligence methods can be used for this purpose. In the case of already existing database points, for example the method or an algorithm of machine learning can be used to replace missing data points.

With the methods described above, simulations can take place through the recognized joints and thus estimates can be made for properties, in particular dimensions such as lengths, widths, diameters and angles for other joints. If, for example, the distance from the floor to an ankle is known, then any other joint distance can be simulated using the Neufert method mentioned above.

Furthermore, the system for setting the vehicle parameters can be designed as a learning system in which e.g. the empirical data on which the setting is based are adjusted as soon as a person does not accept the set parameters, but instead changes the set parameters, or corrected or readjusted.

• - Geschlecht;
• - charakteristische Körperpartien wie z.B. Arme, Oberarme, Unterarme, Hände, Ellenbogen, Beine, Oberschenkel, Unterschenkel, Schulter, Nacken, Brust, Knie, Hüftgelenk, Sprunggelenk oder Kopf (insbesondere zur geeigneten Einstellung eines Kopfstützen-Abstands, wenn lange Haare zu einem Knoten hochgesteckt sind).

For example, the following additional properties of a person can be used to determine suitable anthropometric data, especially for automatic seat adjustment:

• - Gender;
• - Characteristic parts of the body such as arms, upper arms, forearms, hands, elbows, legs, thighs, lower legs, shoulders, neck, chest, knees, hips, ankles or head (especially for suitable adjustment of a headrest distance when long hair forms a knot are pinned up).

In particular, by means of artificial intelligence methods, it can also be achieved that optimized vehicle settings, in particular an optimized seat setting, also take place for a person with individual physical characteristics caused by illness or injury.

Within the scope of the invention, individual persons can also be stored in a data memory data and in particular setting parameters are also stored. These can be called up and used as part of the recognition of the person’s desire for use. As a result, it can be achieved, for example, that the setting takes place more quickly because one or more process steps for determining the vehicle parameters can be dispensed with.

• - A-Säule innen und/oder außen;
• - B-Säule innen und/oder außen;
• - C- Säule innen und/oder außen;
• - Dach innen und/oder außen;
• - Dachfinne;
• - Kühlergrill;
• - Kofferraumdeckel oder Heckklappe;
• - Stoßfänger vorne und/oder hinten;
• - Seitenspiegel rechts und/oder links;
• - Innenspiegel.

With regard to the sensor system, at least one sensor can preferably be used which is provided in at least one of the following positions of the vehicle:

• - A-pillar inside and / or outside;
• - B-pillar inside and / or outside;
• - C-pillar inside and / or outside;
• - roof inside and / or outside;
• - roof fin;
• - grille;
• - trunk lid or tailgate;
• - front and / or rear bumpers;
• - Right and / or left side mirror;
• - Inside mirror.

• - mindestens ein Sensorsystem, mit dem detektierbar ist, dass sich eine Person, in einem vorgegebenem Bereich um das Fahrzeug befindet und das Fahrzeug nutzen will,
• - ein Kamerasystem, das zumindest eine optische Stereokamera und/oder eine Thermokamera umfasst und mit dem die Person aufnehmbar ist,
• - eine Prozesseinheit, mittels der aus mindestens zwei Bildern der Stereokamera und/oder eines Bildes der Thermokamera personenspezifische Merkmale bestimmbar sind und anhand der personenspezifischen Merkmale die einzustellenden Fahrzeugparameter bestimmt werden.

In the context of the invention, a device for the user-specific setting of vehicle parameters of a vehicle can also be specified, which comprises:

• - At least one sensor system with which it can be detected that a person is located in a predetermined area around the vehicle and wants to use the vehicle,
• - a camera system which comprises at least one optical stereo camera and / or a thermal camera and with which the person can be recorded,
• a process unit, by means of which person-specific features can be determined from at least two images from the stereo camera and / or one image from the thermal camera and the vehicle parameters to be set are determined on the basis of the person-specific features.

The process unit can in particular include a computer, a microprocessor, processor, an electronic circuit, a volatile data memory and / or a non-volatile memory and be set up to process sensor data, in particular image data. It can also include computer software.

The device can also have a setting device with which process parameters for determining the person-specific features are set, set, changed, deleted and / or stored in a data memory. For this purpose, the setting device comprises in particular electronic hardware components, software components, an input and / or output interface and in particular a display device and / or buttons, in particular with a touch screen. A user who wants to make a setting to process parameters or a change to an automatically made setting (readjustment) can automatically be shown suggestions and in particular variants that are automatically offered by the system as an alternative to the vehicle parameters previously determined and used for setting, For example, as an alternative to a very flat relaxation lying position, a slightly more upright reading lying position. In the context of such a selection option, limit values can also be provided which the system automatically specifies, for example based on driving safety regulations.

Within the scope of the invention, a vehicle can also be specified which comprises such a device. The device can in particular be set up to carry out a method described in the context of the present description.

• 1 ein Fahrzeug, dem sich eine Person nähert;
• 2 eine Übersicht für die Verarbeitung von Eingangsdaten zur ausgangsseitigen Erzeugung von Fahrzeugparametern;
• 3 einen grundlegenden Prozessablauf;
• 4 einen detaillierteren Prozessablauf;
• 5 ein Funktionsdiagramm.

Further exemplary embodiments of the invention are explained in more detail below with reference to figures. Elements with the same meaning and in particular identical elements have the same reference symbols in each of the figures. Show it:

• 1 a vehicle approaching a person;
• 2 an overview for the processing of input data for the output generation of vehicle parameters;
• 3 a basic process flow;
• 4th a more detailed process flow;
• 5 a function diagram.

In the 1 is a vehicle 1 represented, to which a person 2 approximates. Using sensors 3 of the vehicle 1 be the approach of the person 2 recognized, possibly their intention the vehicle 2 to use and specific parameters of the person 2 determined. The sensors 3 include one on the right front fender 4th provided ultrasonic sensor 3a , one on the right B-pillar 5 of the vehicle 1 attached microphone 3b and on the right side mirror 6 a stereo camera 3c and a thermal camera 3d . On the left side of the vehicle as well as on the front and rear of the vehicle 2 can corresponding, in the 1 sensors not shown may be provided. The person 2 holds a tablet computer 7th in hand and is wearing shorts 8th clothed.

In 2 is shown as an overview, as from a group 9 of input data by means of a process unit 10 a group 11 is generated from output data. The input data include data of a person approaching a vehicle and by means of sensors, in particular the sensors 3 Data obtained that are recorded on the person, such as data on their anthropometry, voice, gender, clothing worn and face. The data, in particular sensor data, can, however, also include person-independent data such as, for example, about objects, about temperature, about location, about weather, and about date and / or time. The output data can include, for example, data to adjust one or more vehicle seats, adjust air conditioning elements for ventilation, heating or cooling, or to operate the locking system of the vehicle. The output data thus form vehicle parameters that can be used to adjust vehicle components.

In the following, an example scenario is used to describe how a person who wants to use a vehicle is recognized as such by the vehicle and how vehicle components are set according to the use of the vehicle desired by the person.

In a sensor system of the vehicle assigned to the recognition, only one microphone with a speech recognition device connected to it is activated in the standby state (with the ignition switched off). The person in front of the vehicle activates further sensor components, in particular the entire sensor system, by means of their voice with the code word “Hello BMW, Andreas wants to drive with you”. Any other word would not activate this system. After the system has been activated, the position from which the voice is coming is determined using a directional microphone or several microphones by triangulation, and a search is made for a person in this area. The person's voice color can also be recorded and evaluated. If no person is detected in the area, the sensors are switched back to the standby state. Otherwise, more precise person recognition is carried out by means of speech, anthropometry, face and / or gender recognition. In particular, a camera system with a stereo camera and / or a thermal camera is also used to record the anthropometric data such as the lengths of the person's arms, legs, hands, upper body, etc.

The data determined by means of the sensors and, in particular, cameras are compared with a database and either, if the person is known, a corresponding identification signal is generated and, if necessary, further identification data determined are stored. The recognition signal can be fed to a feedback system in which, for example, recognition rates are recorded or a recognition is visualized. The signal can then, for example, be sent to a computer system, in particular to a portable electronic unit such as e.g. a tablet or smart phone.

In particular, if the person was not recognized, anthropometric data are determined by assigning the recorded person to a percentile, i.e. the personal data initially recorded by sensors are assigned to a statistical percentile of groups of persons or characteristics of persons.

A sitting position that is aimed at relaxation is influenced by several factors, on the one hand by the person's specific body size and on the other hand by circumstances that are independent of the person. These circumstances can relate, for example, to the objects carried by the person or present in the vicinity of the seat and to special sensor data.

A data record can be generated from the sensor data, which includes recorded objects, temperature, location, weather, time, voice, face, gender and anthropometric data. This data set can be subjected to a correlation analysis with stored, previous data. From this, probabilities for desired non-driving activities such as Reading or resting can be determined. This wish of a recorded person can, for example, be mirrored onto the coordinate system of the vehicle with the seat setting parameters linked to it. This then results in control values in the vehicle for moving and in particular moving seat elements, as well as possible positions for the human-seat-vehicle system. Previously made manual changes to an automatically generated suggestion for the seat settings can serve as a control variable at this point before the seat is brought into its final position. After the seat has been optimally adjusted, at least one vehicle door is also released for the passenger and, in particular, a door locking mechanism of the vehicle is automatically opened.

In 3 describes a basic process sequence for setting vehicle parameters using the example of seat parameters, in which in one step S1 a person accesses a vehicle that is in a stand-by state. In step S2 if the person speaks a code word, e.g. "Hello BMW, I'm Andreas" and the vehicle picks it up The spoken word is recorded by means of a microphone and speech recognition system that is ready for recording even in the stand-by state and recognizes the code word. The vehicle then starts in step S3 a camera system. If the person is within a distance of between 0.5 meters and 3.5 meters from the vehicle, they are recorded by the camera system and the image data generated are used to measure the person anthropometrically. Furthermore, the camera system recognizes whether the person is carrying an item and, if applicable, what type of item it is or what properties it has. If the person is within a distance of between 0.5 meters and 2 meters from the vehicle, step S5 using the camera system and images recorded by it, in particular at least one stereo image and / or at least one thermographic image, facial recognition is carried out for the person. Then, on the basis of the properties determined (anthropometric data, object recognition and / or face recognition) and an algorithm, it is determined which seating parameters are suitable for the person under the given circumstances. In step S6 the seat is automatically adjusted accordingly. Then in step S7 the vehicle opened so that the person in step S8 can get in. In step S9 the camera system is deactivated and the setting process is ended.

In the 4th is a compared to that of the 3 somewhat more complex process for seat adjustment described, with a process step group 15th with process steps S15 to S22 essentially the process steps of 3 corresponds. In step S15 in turn, a person approaches a vehicle and speaks at a step S16 a code word, here “Hello BMW, Camera on”. Then in step S17 the camera system is switched on by an in-vehicle control.

In step S18 it is checked whether the person is in the measuring range of the camera system or within a specified distance range, so that the measurement of the person with regard to anthropometric values such as body size and possibly other properties such as gender start and in the crotch S19 can be done. In step S20 the determined values are displayed, the corresponding vehicle parameters are determined and in step S20 the vehicle seat is automatically adjusted accordingly. Then can in step S21 in turn the door locking mechanism and thus the vehicle door are opened.

Look at the step S19 Captured image data can not only capture personal data, but also other process parameters can be determined, such as in one step S10 an item from which in the step S11 a use case / non-driving activity (English. Use case) is derived or extracted, which characterizes a use of the vehicle desired by the recorded person, for example relaxation, entertainment, office work etc. From the recorded use case in step S12 it can be determined whether there is an unusual situation or an error has occurred and in step S13 a corresponding acoustic or graphic message can then be output, for example by means of a loudspeaker and / or a display device. A so-called red border can also be displayed.

In step S14 can be checked automatically or by the person whether there is an error and a corresponding output signal "Ok" or "not ok" can be output. In the case of the OK, data characteristic of the use case for determining the vehicle parameters that are used to adjust the seat can be incorporated, for example immediately in the course of the step S19 (Measurement of the person), or later in the step S21 the seat adjustment. In the event of a "not OK", an error message is generated and the process can be stopped.

Based on the step S19 The person can use the captured image data in one step S23 for example, can also be identified and / or the gender of the person can be recognized. To this end, an optical image, in particular a 3D image, can be captured or generated, in particular, of the person's face and this in step 24 recognized and in step S25 being checked.

To illustrate the face recognition, an electronic display that is visible from the outside and in particular for the person approaching the vehicle and can be specifically controlled for this purpose can be provided on the vehicle. It can for example be provided on or in an outer pane and / or a strongly light-scattering pane (frosted glass) on the vehicle. A face symbol (so-called FaceCloud) can then be displayed on the illustrative surface or the display, on which the person can intuitively see whether they have been recognized by the vehicle sensors including the downstream data processing and any existing personal data records, for example by a color symbolic design of the face or the edge of the face with eg green color or red color. The person can also be notified of the result of the recognition by means of one or more other sensory channels, for example acoustically and in particular verbally, e.g. by means of a personalized welcome text.

In step S26 the body dimensions assigned to the person are recognized and based on this On the basis of the step, the setting parameters for the seat setting, such as seat angle, length and / or height setting parameters, are determined, in particular calculated. Based on the setting parameters, the seat is in step S21 set and in step S22 in turn ensured that the door can be opened.

From the crotch S19 captured image data or the 3D image of the face generated therefrom can in step S28 it can be determined whether there is an unusual situation or an error has occurred and in step S29 a corresponding acoustic or graphic message can then be output, for example by means of a loudspeaker and / or a display device. A so-called red border can also be displayed.

In step S30 can be checked automatically or by the person whether there is an error and a corresponding output signal "Ok" or "not ok" can be output. In the case of OK, the step can be face examination S25 take place, in case of a "not OK" an error message is generated and the process can be stopped.

In 5 A functional diagram is shown in which the structure of functional elements of a vehicle is shown, with which, for example, the processes described above can be carried out. Functional elements can be sensors and actuators, data memories, data transmission elements, in particular electronic data buses and data processing elements such as

Include computer program modules or a microprocessor. In the process unit 20th that the in 2 process unit shown 10 such functional elements are provided.

With the functional element F1 active microphones are used to check whether spoken text contains a code word. For this purpose, there are corresponding microphone sensors according to the function block F2 active, with which it can also be recognized from which direction the person is approaching the vehicle. If the code word is not recognized, the function block F3 via a feedback function block F4 a function block F5 activated, which in turn puts the respective sensors in a standby state.

In a function block F6 Personal data determined by vehicle sensors, in particular acoustic data (for voice) and optical data (for example for the face of the person) are processed and with data from data stored in a data memory in a personalized manner (function block F7 ) and carried out to identify a person or their characteristics such as anthropometric characteristics or gender. In this case, data relating to a person's face can also be processed and used as part of a first face recognition. In the function block F8 it is checked whether in a data memory, in particular in the data memory in the function block F7 has been read out, corresponding data of a person are already stored and, if necessary, the corresponding person is thereby identified. Existing personal data can be updated or supplemented (function block F10 ) and a second, possibly refined, face recognition can also be carried out. If in the function block F8 no corresponding personal data are determined, the data recorded by sensors or data derived therefrom are stored in the data memory as new personal data (function block F9 ).

From the in the function blocks F9 or. F10 determined personal data is in the function block F11 the person recorded by sensors on the vehicle is assigned to a percentile of the population on the basis of predetermined statistical population data. In function block F 12, this percentile information is processed with further data, namely with further data recorded by sensors on the vehicle, such as according to the function block F13 Data on an identified object that the captured person is carrying, identified clothing that the captured person wears and, if applicable, other data. In addition, data of typical use cases are read from a memory (function block F14 ) and in the function block F12 processed, which correspond, for example, to corresponding objects (e.g. a briefcase that the person carries with him and that represents the application of doing office work). Representative sensor data can also be read out and processed, in particular correlated with the other data (function block F15 ).

In the function block F12 control data are calculated from the input data just mentioned (function block F16 ), on the one hand to control vehicle components, for example such as 5 shown, a vehicle seat can be used (function block F17 ) and, on the other hand, can also be used to create new data for other applications if necessary (function block F18 ). In the function block F17 Current setting data of the vehicle component to be set, here coordinate data for positions of vehicle seat elements, are also used as input data (function block F19 ). Using the in the function block F17 The control data obtained can then be stored in a processor core (kernel), function block F20 , if necessary additionally using manual change values, new setting values for the seat components are calculated and, for example, for Control of corresponding actuators of the seat (function block F21 ), for example an electric motor, can be used. During or after the seat is adjusted accordingly, the locking system of the vehicle can also be activated and thus the vehicle can be opened for boarding (function block F22 ). The devices and system components described can in particular be controlled by means of computer programs and / or electronic control devices. For this purpose, they can have further elements of computers and digital control devices known per se, such as a microprocessor, volatile and non-volatile memories, interfaces, etc. The invention can therefore also be implemented entirely or partially in the form of a computer program product which, when loaded and executed on a computer, completely or partially effects a method according to the invention. It can be provided, for example, in the form of a data carrier such as a CD / DVD or also in the form of one or more files on a server from which the computer program can be downloaded.

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

• DE 102010028580 A1 [0004]
• WO 0208022 A2 [0005]

Claims (13)

Method for user-specific setting of vehicle parameters of a vehicle (1), in which - it is detected by means of at least one sensor system (3) of the vehicle (1) that a person (2) is located in a predetermined area around the vehicle (1) and wants to use the vehicle (1), - the person is recorded by means of a camera system (3c, 3d) which comprises at least one optical stereo camera (3c) and / or a thermal camera (3d), - Using at least two images from the stereo camera (3c) and / or one image from the thermal camera (3d), person-specific features are determined and - The vehicle parameters are set based on the person-specific features. Procedure according to Claim 1 , whereby a) the sensor system detects whether a1) a person (2) is approaching the vehicle (3) and a2) whether the person (2) wants to use the vehicle (3), b) using the camera system (3c, 3d) data relating to the person (2) are recorded and c) the person-specific features are determined by means of the data recorded by the camera system (3c, 3d). Procedure according to Claim 1 or 2 , with anthropometric data of the person (2) being determined as person-specific features by means of data that were recorded by the camera system (3c, 3d). Method according to one of the Claims 1 to 3 , situation-specific attributes that are used for setting the vehicle parameters are determined by means of data that have been recorded by the sensor system (3). Method according to one of the preceding claims, wherein a parameter of a vehicle seat is set as the vehicle parameter. Method according to one of the preceding claims, the absolute body size of the person (2) being determined from the data of the person (2) recorded by the camera system (3c, 3d). Method according to one of the preceding claims, wherein at least one of the following methods is used to determine the person-specific features: - Use of reference data that are determined using artificial intelligence methods, - Face recognition, in particular using a grid projection. Method according to one of the preceding claims, wherein at least one sensor of the sensor system (3) is used which is provided in at least one of the following positions of the vehicle (1): - A-pillar inside and / or outside; - B-pillar inside and / or outside; - C-pillar inside and / or outside; - roof inside and / or outside; - roof fin; - grille; - trunk lid or tailgate; - front and / or rear bumpers; - Right and / or left side mirror; - Inside mirror. Method according to one of the preceding claims, wherein at least one sensor of the sensor system (3) is a microphone (3b) and a speech recognition unit is used to recognize text from signals output by the microphone (3b). Device for user-specific setting of vehicle parameters of a vehicle (1), comprising - At least one sensor system (3) with which it can be detected that a person (2) is in a predetermined area around the vehicle (1) and wants to use the vehicle (2), - a camera system (3c, 3d) which comprises at least one optical stereo camera (3c) and / or a thermal camera (3d) and with which the person (2) can be recorded, - A process unit (10) by means of which person-specific features can be determined from at least two images from the stereo camera (3c) and / or one image from the thermal camera (3d) and the vehicle parameters to be set are determined on the basis of the person-specific features. Device according to Claim 9 or 10 , further comprising a setting device for setting process parameters for determining the person-specific features. Device according to Claim 10 or 11 , set up a method according to one of the Claims 1 to 9 execute. Vehicle comprising a device according to one of the Claims 10 to 12 .

Images