WO2019192796A1

WO2019192796A1 - Determining the destination of a first person to be transported by a passenger transportation vehicle

Info

Publication number: WO2019192796A1
Application number: PCT/EP2019/055638
Authority: WO
Inventors: Daniel Scheffler; Jens Hoffmann
Original assignee: Zf Friedrichshafen Ag
Priority date: 2018-04-04
Filing date: 2019-03-07
Publication date: 2019-10-10
Also published as: DE102018205051A1

Abstract

An analysis device (10) for determining a destination of a first person (2) to be transported by a passenger transportation vehicle (1), the analysis device comprising a first interface (11) for obtaining at least one signal belonging to a detection of the first person (2), wherein the analysis device (10) is designed to determine first features of the first person (2) according to the at least one signal, and a second interface (12) for obtaining second features of second people (3), wherein the second features comprise identification features and at least one destination of the second person (3), wherein the analysis device (10) is designed to compare the first features and the second features, to identify the first person (2) according to said comparison and to determine at least one destination for the first person (2) according to said identification. The invention further relates to a corresponding central system, detection system, method and computer program product.

Description

Determining a transportation destination of a first person to be transported by a passenger transport vehicle

The invention relates to an evaluation device for determining a transportation destination of a first person to be transported by a passenger transport vehicle according to claim 1. Furthermore, the invention relates to a central system for determining a transportation destination of a first person to be transported by a passenger transport vehicle according to claim 9. Further relates The invention relates to a recognition system for determining a transportation destination of a first person to be transported by a passenger transport vehicle according to claim 10 and to a passenger transport vehicle with a recognition system according to claim 11. The invention further relates to a method for determining a transportation destination of a passenger transport vehicle conveying first person according to claim 12. The invention also relates to a computer program product for carrying out the method according to Ansp smell 15.

In many cities around the world, service providers offer online passenger transport services. For example, passengers are provided with rental cars with a human driver, private human drivers with their own vehicles, or taxis with a human taxi driver.

The ordering of a vehicle from a vehicle fleet of these service companies is usually done with a mobile device that is connected to the Internet, for example, with a smartphone. On the smartphone, an application, briefly App is running to order the vehicle. A person asks the vehicle with the app. A driver of the vehicle accepts this request. The app shows the estimated time of arrival of the vehicle and notifies the person when the vehicle arrives. The app displays the name of the driver, the vehicle and the license plate of the vehicle. When the person arrives at their destination, the journey is over. The fare is calculated automatically and charged a means of payment of the person in the app. Ordering the vehicle with a smartphone or other mobile device may be restricted. For example, the smartphone can not be used due to low battery capacity and / or poor reception. Many older people do not own a smartphone. Smartphones of many children are operated with a rechargeable account and therefore can not be used when the balance is too low. When a passenger transport vehicle approaches a person to be transported, there is generally little time to order the passenger vehicle during the approach by smartphone. If the passenger transport vehicle can no longer be ordered with the app during the approach, the passenger transport vehicle or a human driver of the passenger transport vehicle has no information about the person to be transported before the start of the journey, in particular the transport destination is not known prior to the commencement of the journey. In such cases, the transport destination will not be known until the person to be transported enters the passenger transport vehicle and mentions its destination. But time is lost.

This is where the invention starts. The invention has the object to improve the handling of trips with passenger vehicles.

The object is achieved by an evaluation device for determining a transportation destination of a first person to be transported by a passenger transport vehicle having the features of claim 1. Furthermore, the object is achieved by a central system for determining a transportation destination of a first person to be transported by a passenger transport vehicle with the features of Claim 9. Furthermore, the object is achieved by a recognition system for determining a transportation destination of a first person to be transported by a passenger transport vehicle having the features of claim 10 and a passenger transport vehicle having a recognition system according to the invention as claimed in claim 11. The object is further achieved by a method for Determining a transportation destination of a first person to be transported by a passenger transport vehicle with the features of claim 12 The object is also achieved by a computer program product for carrying out the method according to the invention with the features of claim 15.

Further developments and advantageous embodiments are specified in the subclaims.

The evaluation device according to the invention for determining a transportation destination of a first person to be transported by a passenger transport vehicle comprises a first interface for obtaining at least one signal belonging to a detection of the first person. The evaluation device is designed to determine first characteristics of the first person as a function of the data. Furthermore, the evaluation device comprises a second interface in order to obtain second features of second persons. The respective second features include identification features and at least one transportation destination of the respective second person. The evaluation device is designed to compare the first features and the second features. In addition, the evaluation device is designed to identify the first person as a function of this comparison and, depending on this identification, to determine at least one promotion destination for the first person.

An evaluation device is a device that processes incoming information and outputs a result resulting from this processing. In particular, an electronic circuit, such as a central processor unit or a graphics processor, an evaluation device.

A passenger transport vehicle is a land, sea or air vehicle designed to carry at least one person from a first location to a second location. In particular, a taxi or a bus is a passenger transport vehicle. In particular, minibuses for transporting persons on short journeys, for example in cities, airports or trade fairs, also referred to as people's vehicles, are passenger vehicles. Drones for transporting people are also passenger vehicles. Preferably, the passenger transport vehicle is operated automatically.

An automated passenger transport vehicle is a passenger transport vehicle having technical equipment to handle a driver task, including longitudinal and transverse guidance, the respective vehicle after activation of a corresponding automatic driving function, in particular a fully automatic or fully automated driving function according to the standard

SAEJ3016, with a vehicle control device.

A partially automated vehicle can take over certain driving tasks. The SAEJ3016 standard distinguishes between SAE Level 2 and SAE Level 3. Level 2 is defined by the driving mode specific execution of steering and acceleration / braking operations being performed by one or more driver assistance systems using information about the driving environment with the expectation in that the human driver performs all remaining aspects of the dynamic driving task. Level 3 is defined by performing the drive mode specific execution of all aspects of the dynamic driving task by an automated driving system with the expectation that the human driver will respond appropriately at the request of the system.

A fully automated vehicle replaces the driver. The SAEJ3016 standard distinguishes between SAE Level 4 and SAE Level 5 Level 4 is defined by the driving mode specific execution of all aspects of the dynamic driving task being performed by an automated driving system, even if the human driver does not respond adequately at the request of the system. Level 5 is defined as the consistent execution of all aspects of the dynamic driving task performed by an automated driving system under all driving and environmental conditions that can be managed by a human driver.

A pure assistance system assists the driver in carrying out a driving task. This corresponds to SAE Level 1. The invention is particularly intended for SAE levels 3, 4 and 5. In particular, in a transitional period to fully automated driving, the invention is applied to SAE levels 3 and 4, to be used subsequently to SAE level 5.

The first person is a person who wants to be transported by the passenger transport vehicle. For example, the first person is a person waiting on the side of the road who expresses his desire to travel by waving in combination with head orientation toward the passenger transport vehicle. The second persons are a lot of people, to which the first person can belong.

An interface is a device between at least two functional units at which an exchange of logical quantities, for example data or physical quantities, for example electrical signals, takes place, either only unidirectionally or bidirectionally. The exchange can be analog or digital. The exchange can also be wired or wireless.

A signal belonging to a detection of the first person is in particular an output signal of a sensor. The sensor detects, that is, detects, the first person or at least parts of the first person. From this detection arise first features that allow an actual recognition, known as recognition, the person in the sense of identification, that is, an assignment of the first person to the right second person. According to the invention, therefore, first a detection and then a recognition.

The first characteristics refer to the first person. The first features include first person identification features. Identification features are features that uniquely and unmistakably characterize a person. The evaluation device determines the first features from the at least one signal. For example, the evaluation device is designed to process image processing and / or speech recognition algorithms. Image processing algorithms recognize in image signals, for example, geometric shapes and / or arrangements which these forms. Speech recognition algorithms detect speech patterns, volume levels and / or frequency spectra in audio signals. With such algorithms, the first features are calculated.

The second features relate to the second person. The second features include identification features of the respective second person and transportation destinations, in particular probable transportation destinations, of this second person. The second features are stored for example in a database. The evaluation device accesses this database with the second interface. Advantageously, the database is a cloud, that is, a storage space provided as a service over the Internet. The second interface is advantageously a WLAN interface.

A first feature of the first person is compared to the corresponding second feature of the second person. For example, if the first feature is a first person's fingerprint, second fingerprints of the second person are compared to the first person's fingerprint, for example, by a corresponding query in the cloud. If the fingerprint of the first person matches the fingerprint of one of the second persons, the first person is identified as the respective second person. By the identification feature of the fingerprint of the second person whose destination is known. In a positive comparison, the first person's promotion goal is known.

The transport destinations can also be linked to a position specification. The position specification indicates the position at which the first person is detected. For example, the position indication is a specific stop at which the first person waits for the passenger transport vehicle. The position can be detected, for example, with a position detector based on the global positioning system. For example, for a second person Z, the transport destination station is linked to position X1. Or the transportation destination of the second person Z is airport linked to location X2. Now the first person E is detected at location X1 and gives a comparison the identification features that the person E is the person Z, then follows as the destination of transport due to the location X1 of the station.

With the evaluation device according to the invention probable destinations are already known before the promotion. This can be done faster than in conventional taxi rides, where the passenger must first tell the driver his transport destination, especially if the passenger can not order the passenger vehicle with an app, especially in a taxi ride. The identified passenger is proposed likely transportation destinations directly on boarding the passenger transport vehicle. This leads to logistical advantages and a better customer experience.

The at least one signal preferably comprises an image signal of an image recording, preferably of a face, of the first person and / or an audio signal of a voice recording of the first person. The image signal is advantageously an output signal of an imaging sensor arranged on a passenger transport vehicle. From this signal, the evaluation device calculates, for example, the size and arrangement of the eyes and / or the mouth in order to obtain a face recognition of the first person, via which the first person can be identified. For this calculation, the evaluation device preferably processes an image processing algorithm which recognizes forms, their size and arrangement from one another, from image signals.

The evaluation device is designed to detect objects which obscure the face, for example scarf, hat and / or spectacles, in order to be able to perform facial recognition even in the event of occlusion.

Preferably, the first features are biometric features. Biometric features are individual behavioral or physical characteristics of a person. The biometric features are measured by electronic systems and, like other data, can be stored, compressed, transmitted and compared with already stored biometric features by a computer. Biometric features For example, a person's deoxyribonucleic acid, a fingerprint, a gait style, a face geometry, a hand geometry, the iris, lip movements, ear shape, fundus and voice. The biometric features are calculated by the evaluation device from the at least one signal. For example, a face geometry is calculated from the image signal. From the voice signal, the voice is determined. By biometric features a person is identified.

Advantageously, the first features comprise arrangements, preferably position, distance and position relative to one another, and / or surface properties of at least the eyes, nose and / or mouth of the first person. Such features are in particular also biometric features. Eye, nose and / or mouth and also ear or chin are exemplary manifestations of visible features in the area of the head. These features allow facial recognition of the first person.

The first features preferably comprise at least voice frequency, volume and / or intonation of the first person. Intonation refers to the change in pitch when speaking. Such features are in particular also biometric features. Voice frequency, volume and / or intonation determine a person's voice. These features allow first person voice recognition.

The evaluation device is designed to determine from the at least one signal a biometric date. The evaluation device is preferably designed to execute a hash algorithm in order to obtain a hash value for the biometric datum. A hash algorithm specifies a hash function. A hash function maps a large input quantity, for example the biometric date, to a small target quantity, the so-called hash values, for example to a digital pattern. In particular, the hash algorithm identifies content, for example the biometric date, almost unambiguously, for example via the digital pattern, without revealing anything about the content. The hash value is used to anonymously manage personal data, in this case biometric data. Analogous to the first features, each of the second features can also be assigned a hash value. be pointed. It will then not the first features directly compared with the second features, but only the associated hash values.

Preferably, the second features comprise biometric features and person profiles of the second persons. This compares biometric features in a database. Such a comparison results in a secure identification. Person profiles include, in addition to a promotion goal of the respective second person, for example, a name of the respective second person. This not only determines the transport destination of the identified first person. In addition, for example, the name of the first person who belongs to their person profile, known. The first person will then be greeted with their name when entering the passenger transport vehicle. A personal greeting enhances the customer experience. The person profiles preferably include person related data, such as entertainment requests. Thus, the identified first person may, for example, listen to their preferred music while listening to the passenger transport vehicle or hear and / or read their favorite messages. The passenger transport vehicle is then equipped with appropriate infotainment systems at seating and / or standing room. This further improves the customer experience.

According to an advantageous embodiment of the invention, the evaluation device is designed to execute a computer program. The computer program includes software code sections. With the software code sections, first features of the first person are determined as a function of the at least one signal. The software code sections also compare the first features with the second features. Depending on the comparison, at least one promotion goal for the first person is determined with the software code sections. The evaluation device executes the computer program.

A computer is a device for processing data that can process data by means of programmable calculation rules. An evaluation device can be a computer. Software is a collective term for programs and related data. The complement to software is hardware. Hardware refers to the mechanical and electronic alignment of a data processing system.

The computer program calculates a transport destination from the at least one signal.

Preferably, the computer program comprises an artificial neural network architecture. The evaluation device is designed to feed the artificial neural network with the data. The artificial neural network is trained to determine the first features as a function of the data.

An artificial neural network is an intelligent algorithm. Intelligent algorithm means that the algorithm means artificial intelligence, in English as artificial intelligence, abbreviated AI, the branch of computer science, which deals with the automation of intelligent behavior and machine learning includes. An intelligent algorithm, for example, is an artificial neural network.

The artificial neural network is executed on an electronic circuit and is programmed on the model of the neural network of the human brain. Functional units of an artificial neural network are artificial neurons whose output is generally evaluated as the value of an activation function over a weighted sum of the inputs plus a systematic error, the so-called bias. By testing several predetermined inputs, also called target training data, with different weighting factors and / or activation functions, artificial neural networks similar to the human brain are trained. Target training data is positive training data with which the artificial neural network learns true information. Target training data are labeled with the meaning of the information, that is marked, so that the artificial neural network can capture information semantically. The training of an artificial neural network using target training data is called machine learning. A subset of machine learning is deep learning, in which a series of hierarchical layers of neurons called hidden layers are used to perform the machine learning process. An artificial neural network with multiple hidden layers is a deep neural network.

In the training phase, weighting factors are used to evaluate connections between neurons. Feed forward, meaning forward propagation, means that information is fed into the input layer of the artificial neural network, traverses the following layers and outputs it in the output layer. Backfeed, meaning backward propagation, means that information is input to the output layer and output in the input layer. By successively feeding back the error from the output layer into the respective previous layer up to the input layer, the errors of the respective layers are obtained. The errors are a function of the weighting factors. Minimizing the error in the training phase changes the weighting factors. As a result, an approach to the desired output is achieved when re-feeding input. Backfeed is described in detail in Michael A. Nielsen, Neural Networks and Deep Learning, Determination Press, 2015.

The transportation destinations associated with the second features are preferably learned during the training phase of the artificial neural network during a transportation journey. The trained artificial neural network knows then already with the identification of the first person their probable transport destination, without having to resort to the second characteristics deposited promotion goals.

A trained artificial neural network is characterized by purposely responding to new information. Examples of artificial neural networks are convolutional or recurrent neural networks. Convolutional artificial neural networks are characterized by a two- or three-dimensional arrangement of neurons and shared weighting factors and are used in particular for image recognition, in which the input is a gray image or an image in a three-dimensional color space and thus a two- or three-dimensional matrix. A recurrent, also feedback, neural network is a network which is characterized by connections of neurons of a layer to neurons of the same or a preceding layer. Practical applications commonly find artificial recurrent neural networks in problems requiring the processing of sequences. Examples include text recognition, speech recognition and machine translation.

The artificial neural network is particularly preferably trained to identify the first person as a function of the comparison of the first features with the second features and to issue at least one delivery destination as a function of this identification.

In one development of the invention, the evaluation device comprises a third interface in order to output the at least one transport destination. The third interface is preferably an audio interface and / or an optical interface for audibly and / or visibly outputting the transportation destination to a human driver of the passenger transport vehicle. In a development of the invention, the third interface is an interface to a control device of the passenger transport vehicle in order to automatically control the passenger transport vehicle to the transport destination.

The evaluation device is preferably an AI platform which can be used for the automotive sector and has a high-performance graphics processor which is designed to execute driver tasks for automated driving.

The evaluation device is preferably designed to determine a state of health of the identified first person as a function of the detected and compared biometric features. Thus, the journey of the passenger be adapted to the state of health of the first person. For example, with a detected nausea, braking torques, choice of gears and / or adjustment of the damping system are selected such that the ride is relatively comfortable and gentle, unlike a sports mode.

The evaluation device is advantageously designed to refuse problematic passengers, for example extremely drunk passengers, the entry into the passenger transport vehicle in a diplomatic manner, for example by a voice message to the passenger that the passenger transport vehicle is already reserved. Such passengers are detected by appropriate signals, for example beer bottle.

According to a development of the invention, the evaluation device is designed to respond appropriately to recognized wheelchair users or persons with prams, for example by opening a tailgate and / or extending a ramp. Preferably, the evaluation device is designed to recognize injured and / or needy people and initiate appropriate action.

The central system according to the invention for determining a transportation destination of a first person to be transported by a passenger transport vehicle comprises a data memory. The data store comprises second features of second persons. The respective second features include identification features and at least one transportation destination of the respective second person. In addition, the central system comprises an evaluation device according to the invention. The first interface of the evaluation device is designed to exchange the at least one signal associated with the detection of the first person with the passenger transport vehicle. The second interface of the evaluation device is designed to obtain the second features from the data memory. The evaluation device is designed to be executed on the central system and to determine, depending on the passenger transport vehicle, at least one signal belonging to the detection of the first person, first characteristics of the first person. The central system is designed to determine the destination via the first interface to the passenger transport vehicle.

The advantage of the central system is that the evaluation device does not have to be operated in an on-board network of the passenger transport vehicle when in use. Thus, the evaluation consumed in operation no energy of the passenger transport vehicle, which is particularly advantageous if the passenger vehicle is an electrically powered passenger transport vehicle. Rather, the evaluation device is executed on a central infrastructure component, for example a cloud server. Via the first interface, which is preferably a WLAN interface, the passenger transport vehicle communicates with the central system. That is, the passenger vehicle transmits the at least one signal to the center system and receives the destination of transportation determined in the center system.

The detection system according to the invention for determining a transportation destination of a first person to be transported by a passenger transport vehicle comprises at least one sensor which can be arranged on the passenger transport vehicle for detecting the first person. In addition, the recognition system includes a data storage. The data store comprises second features of second persons. The respective second features include identification features and at least one transportation destination of the respective second person. Furthermore, the recognition system comprises an evaluation device according to the invention. Furthermore, the recognition system comprises a utilization device. The first interface of the evaluation device is an interface to the sensor. The second interface of the evaluation device is an interface to the data memory. The third interface of the evaluation device is an interface to the utilization device.

A user interface, referred to as a human machine interface, is the place or action that a person uses to contact a machine. The contact then takes place with the use device. The utilization device is, for example, a screen and / or a loudspeaker. In contrast to the central system, the calculation of the transport destination within the recognition system takes place. The data memory is arranged in the passenger transport vehicle. Alternatively, the data store is an online data store. The recognition system is a self-contained, fully functional system. Passenger vehicles can be retrofitted with the detection system. For example, taxis of a taxi fleet can be retrofitted with the recognition system. The recognition system searches for and recognizes potential passengers. This relieves the driver of the taxi. The driver can concentrate fully on controlling the vehicle and does not have to search for passengers at the same time.

The sensor may be one or more imaging sensors, such as a camera, and / or sound transducers, such as a microphone. With a camera visually visible properties are detected. With a microphone acoustically perceptible properties are detected. The signals of these acquisitions then give rise to the biometric features of a face geometry and / or a voice. Thus, a face recognition and / or voice / speech recognition of the first person can be performed. Preferably, the evaluation device is designed to fuse signals from a plurality of sensors in order to achieve improved detection. By merging several sensor signals, disturbing factors in the detection, for example illumination and / or distortion, are detected and corrected at a distance.

With at least two cameras which can be arranged on the passenger transport vehicle, a first camera being arranged on a first side and a second camera being arranged on a second side of the passenger transport vehicle, advantageously both street sides are searched for first persons.

According to the invention, a passenger transport vehicle comprises a detection system according to the invention. Passenger vehicles, especially passenger vehicles with driver assistance systems, are usually already equipped with environment detection sensors, for example cameras. The method according to the invention for determining a transportation destination of a first person to be transported by a passenger transport vehicle comprising the following method steps: a. Obtaining at least one signal associated with a detection of the first person,

b. Determining first features as a function of the at least one signal,

c. Obtaining second features from second persons, the respective second features comprising identification features and at least one transportation destination of the respective second person,

d. Comparing the first features and the second features,

e. Identifying the first person as a function of this comparison and f. Determining at least one transportation destination for the first person, wherein the method steps a) to f) take place during a journey of the passenger transport vehicle to the first person in order to determine the at least one transportation destination already before the beginning of the journey.

It is essential to the invention that the method steps a) to f) take place during a journey of the passenger transport vehicle to the first person. This will save time in the handling of the transport and the resulting logistical benefits.

If the first person is interested in traveling with the passenger transport vehicle by means of a corresponding head and / or body orientation, this person is considered in detail in method step a). The passenger transport vehicle drives the first person for further scans, preferably autonomously.

During the approach, the first person is considered in detail and method step b) is carried out, that is, a biometric date of the first person, for example a face recognition and / or a body recognition, is determined. Process steps b) to f) can also be carried out until the passenger transport vehicle is in front of the first person. At smaller the distance the sensors capture more details. This will preserve biometric features in detail.

Preferably, prior to method step a), a transport request of the first person for carriage with the passenger transport vehicle is detected as a function of the data on the detection of the first person, and depending on this recognition, method steps a) to f) are executed. During a journey of the passenger transport vehicle, the sensors of the passenger transport vehicle scan the surroundings of the passenger transport vehicle for first persons who signal a request for transport.

Typical transportation signals that signal ride-in are, for example, waving in combination with head orientation and / or whistling with the person's lips and / or fingers toward the passenger vehicle. Pipes with lips or other call signals, especially voice signals, are also transport signals taken on their own. Such ringing signals can be detected with microphones or other sound transducers for further processing.

Before method step a), for example, initially only one head and / or body orientation is analyzed for the passenger transport vehicle. Analogous to face recognition and / or body recognition, speech recognition is preferably carried out. The first person not only identifies with his or her face, but alternatively or additionally with the use of their voice. Speech recognition is preferably carried out on a scan of a conveyance request, for example when the first person calls a taxi.

Advantageously, an evaluation device according to the invention is used to carry out the method.

The method is preferably carried out on a central system according to the invention or in a passenger transport vehicle according to the invention. Thus, the method can be realized for example in a cloud or with the recognition system in the passenger transport vehicle. The computer program product according to the invention is designed to be loaded into a memory of a computer and comprises software code sections with which the steps of the method according to the invention are carried out when the computer program product is running on the computer.

Computer program products typically include a sequence of instructions that cause the hardware, when the program is loaded, to perform a particular procedure that results in a particular result. When the program concerned is used on a computer, the computer program product causes a technical effect, namely, obtaining a transportation destination from the processing of at least one signal associated with a first person's acquisition.

A memory is a medium for backing up data.

The computer program product according to the invention is platform independent. That is, it can run on any computing platform. Preferably, the computer program product is executed on an evaluation device according to the invention.

If the first person signaling a promotion request is not included in the amount of second persons, the first person may identify himself with the passenger transport vehicle. The evaluation device is designed to store this identification as a new second person in the amount of existing second persons.

In summary, a passenger is recognized due to its external features such as face or voice. This happens because the passenger is detected by the cameras / microphones of the passenger transport vehicle and the person profile created therefrom is compared with the person profiles stored in a database. The advantage is that this is done very early before the passenger vehicle has reached the passenger. The resulting Benefits include a better customer experience, such as a personal greeting or faster processing, and logistical benefits because likely transportation goals are known earlier.

The invention will be explained by way of example with reference to the following figures. Show it:

1 a is a schematic view of an embodiment of an evaluation device according to the invention,

1 b is a schematic view of an embodiment of a face geometry,

2 is a schematic view of an embodiment of a passenger transport vehicle with an embodiment of a detection system according to the invention,

Fig. 3 is a schematic view of an embodiment of a central system according to the invention and

4 shows a schematic sequence of a method according to the invention.

In the figures, unless otherwise stated, like reference numerals designate like or functionally similar reference parts. In the figures, the respective relevant reference parts are marked.

FIG. 1 a shows an evaluation device 10. The evaluation device 10 is designed to execute an artificial neural network 20. The artificial neural network 20 is a convolutional neural network. The artificial neural network 20 is available as a computer program product 50. The computer program product 50 is executable independently of a computing platform. The evaluation device 10 receives signals from a camera via the detection of a face 2a of a first person 2 via a first interface 11. The first person 2 stands at the edge of a road and signals a passenger transport vehicle 1 to request a transport. The artificial neural network 20 is trained to classify eyes 2b, nose 2c and mouth 2d of the face 2a depending on the signals of the camera. The face 2a is shown in detail in FIG. 1b. From this classification, the artificial neural network 20 calculates the facial geometry and thereby determines first biometric features of the first person 2. The artificial neural network 20 is implemented, depending on signals from a microphone via a first person 2 voice recording, a voice recognition as a further biometric feature first person 2 to charge.

The evaluation device 10 receives second features of second persons 3 via a second interface 12. The second features are biometric features of the second persons 3 and comprise a person profile of the respective second person 3. In the person profiles are preferred transportation destinations, names and favorite music of the respective second Persons 3 deposited.

The artificial neural network 20 compares the biometric features of the first person 2 with corresponding biometric features of the second persons 3. In a match, the first person 2 is identified as the corresponding second person 3. Since the second features of the second person 3 also include preferred transportation destinations, the first person 2 is identified and their preferred transportation destinations are known.

Furthermore, the evaluation device 10 has a third interface 13. Via the third interface 13, the preferred transportation destinations of the first person 2 are outputted to the passenger transport vehicle 1 during the journey to the first person 2. Thus, the first person 2 does not have to be asked about his transportation destination when entering the passenger transport vehicle 1. The preferred destinations are already known in advance. The artificial neural network 20 may also learn the preferred transportation goals depending on position and / or time of day. If, for example, the first person 2 rises at the Station X arrives at passenger transport vehicle 1 at 4:00 pm every Tuesday for transportation to its home, the trained artificial neural network 20 is aware of the transportation destination of that first person 2 entering at the given conditions, and the transportation destination may be directly after identification of the person first person 2 and getting into the passenger vehicle are controlled.

FIG. 2 shows the passenger transport vehicle 1 equipped with a recognition system 40. The passenger transport vehicle 1 is a minibus whose interior offers, for example, eight seats and five standing places. For example, the minibus is five meters long, two meters wide and two and a half meters high. The minibus is powered, for example, by an electric drive system with a capacity of about 150 kilowatts. The battery of the drive system has a capacity that ensures a service life of about ten hours. The curb weight of the minibus is for example about 2000 kilograms. Such a minibus is particularly well suited as a people mover.

The detection system 40 comprises a sensor 41 in the form of a camera and a data memory 31 on board the passenger transport vehicle 1. Alternatively, the data memory 31 is an online data memory. The data memory comprises the second features of the second persons 3. Furthermore, the recognition system 40 comprises the evaluation device 10. The evaluation device 10 is connected to the sensor 41 via the first interface 11. Via the second interface 12, the evaluation device 10 is connected to the data memory 31. The evaluation device is connected via the third interface 13 with a use device 42 in the form of a screen. Via the screen, the passenger transport vehicle 1 and / or the first person to be transported receives preferred transport destinations.

3 shows a central system 30 in the form of a cloud server. The central system 30 comprises the data memory 31 and the evaluation device 10. The data memory 31 comprises the second features and person profiles of the second persons 3. The data memory 31 is connected via the second interface 12 to the ejector. teeinrichtung 10 connected. The evaluation device 10 is executed on the central system 30. The evaluation device 10 exchanges information with the passenger transport vehicle 1 via the first interface 1 1. The passenger transport vehicle 1 sends to a detection of the first person 2 belonging signals to the central system 30. About the first interface 1 1, these signals are received. As a function of these signals and the second features from the data memory 31, the evaluation device 10 calculates probable delivery destinations of the first person 2. These transportation destinations are sent by the central system 30 via the first interface 11 of the evaluation device 10 to the passenger transport vehicle 1. The passenger transport vehicle 1 is executed To receive and further develop transport destinations.

4 shows the method for determining a transportation destination of the first person 2 to be transported by the passenger transport vehicle 1. In a first method step V1, at least one signal belonging to the detection of the first person 2 is received. The at least one signal is, for example, an audio signal of a microphone via detection of the voice of the first person 2. In a second method step V2, first features are determined as a function of the at least one signal. For example, the trained artificial neural network 20 in an input layer receives the at least one signal. In an output layer, the artificial neural network 20 outputs the first features. The first features are biometric features, such as volume and frequency response of the first person's voice 2. In a third method step V3, second features 3 are obtained from second persons. The respective second features comprise identification features and at least one transport destination of the respective second person 3. The second features are stored, for example, in a database. In a fourth method step V4, the first features and the second features are compared. In a fifth method step V5, the first person 2 is identified as a function of this comparison as a function of a match of the first features with the second features of one of the second persons 3. Depending on this identification and the destination associated with the respective second person 3 is in a sixth Process step determines the transport destination for the first person 2. The process steps V1 to V6 take place during a journey of the passenger transport vehicle 1 to the first person 2. Thus, the transport destination is determined before the start of the transport.

reference numeral

A passenger service vehicle

first person

a face

b eye

c nose

the mouth

second person

0 evaluation device

1 first interface

2 second interface

3 third interface

0 artificial neural network

0 central system

1 data store

0 recognition system

1 sensor

2 utilization device

0 computer program product

Claims

claims

An evaluation device (10) for determining a transportation destination of a first person (2) to be transported by a passenger transport vehicle (1)

A first interface (1 1) for obtaining at least one signal belonging to a detection of the first person (2),

wherein the evaluation device (10) is designed to determine first characteristics of the first person (2) as a function of the at least one signal,

A second interface (12) for obtaining second features of second persons (3), the respective second features comprising identification features and at least one transport destination of the respective second person (3),

wherein the evaluation device (10) is executed,

To compare the first features and the second features

• to identify the first person (2) as a function of this comparison and

• determine at least one transport destination for the first person (2) depending on this identification.

2. evaluation device (10) according to claim 1, wherein the at least one signal comprises an image signal of an image recording, preferably a face (2a), the first person (2) and / or an audio signal of a voice recording of the first person (2).

3. evaluation device (10) according to claim 1 or 2, wherein the first features are biometric features.

4. evaluation device (10) according to claim 2 and 3, wherein the first features arrangements, preferably position, distance and position to each other, and / or surface properties of at least eyes (2b), nose (2c) and / or mouth (2d) of the first Person (2) include.

5. evaluation device (10) according to claim 2 and 3, wherein the first features at least voice frequency, volume and / or intonation of the first person (2).

6. evaluation device (10) according to any one of the preceding claims, wherein the second features include biometric features and person profiles of the second person (3).

7. evaluation device (10) according to one of the preceding claims, wherein the evaluation device (10) is executed,

To execute a computer program comprising software code sections with which first features of the first person (2) are determined in dependence on the at least one signal, the first features are compared with the second features and with which at least one transport destination for the first person (2) is determined as a function of this comparison when the evaluation device (10) executes the computer program,

Wherein preferably the computer program comprises an architecture of an artificial neural network (20) and the evaluation device (10) is adapted to feed the artificial neural network (20) with the data, wherein the artificial neural network (20) is trained, depending on the data to determine the first characteristics

Wherein, particularly preferably, the artificial neural network (20) is trained to identify the first person (2) as a function of the comparison of the first features with the second features and to issue at least one transportation destination in dependence on this identification.

8. evaluation device (10) according to one of the preceding claims, wherein the evaluation device (10) comprises a third interface (13) to output the at least one transport destination.

A central system (30) for determining a transportation destination of a first person (2) to be transported by a passenger transport vehicle (1) • a data memory (31) comprising second features of second persons (3), the respective second features comprising identification features and at least one transport destination of the respective second person (3), and

• an evaluation device (10) according to any one of claims 1 to 7, wherein o the first interface (1 1) of the evaluation device (10) is executed, the at least one to the detection of the first person (2) associated signal with the passenger transport vehicle (1 o) the second interface (12) of the evaluation device (10) is designed to receive the second features from the data memory (31), o wherein the evaluation device (10) is designed to be executed on the central system (30) and determining first features of the first person (2) as a function of the passenger transport vehicle (1), at least one signal belonging to the detection of the first person (1),

wherein the central system (30) is designed to output the transport destination determined by the evaluation device (10) via the first interface (11) of the evaluation device (10) to the passenger transport vehicle (1).

A recognition system (40) for determining a transportation destination of a first person (2) to be transported by a passenger transport vehicle (1)

At least one sensor (41), which can be arranged on the passenger transport vehicle (1), for detecting the first person (2),

A data memory (31) comprising second features of second persons (3), the respective second features comprising identification features and at least one transport destination of the respective second person (3),

• An evaluation device (10) according to any one of claims 1 to 8, and

A utilization device (42),

wherein the first interface (1 1) of the evaluation device (10) an interface to the sensor (41), the second interface (12) of the evaluation device (10) an interface to the data memory (31) and the third interface (13) of the evaluation (10) is an interface to the utilization device (42).

1. A passenger transport vehicle (1) comprising a recognition system (40) according to claim 10.

12. A method for determining a transportation destination of a first person (2) to be transported by a passenger transport vehicle (1), comprising the following method steps:

a. Receive at least one signal (V1) belonging to a detection of the first person (2),

b. Determining first features as a function of the at least one signal (V2),

c. Obtaining second features from second persons (3) (V3), the respective second features comprising identification features and at least one transportation destination of the respective second person (3),

d. Comparing the first features and the second features (V4),

e. Identifying the first person (2) as a function of this comparison (V5) and

f. Determining at least one transportation destination for the first person (2) (V6), wherein the method steps a) to f) take place during a journey of the passenger transport vehicle (1) to the first person (2) to determine the at least one transportation destination already before the beginning of the journey ,

13. The method according to claim 12, wherein prior to method step a), a transport request of the first person (2) for carriage with the passenger transport vehicle (1) is detected as a function of the data on the detection of the first person (2) (V0) and in dependence this recognition process steps a) to f) are executed.

14. The method of claim 12 or 13, wherein the method on a central system (30) according to claim 9 or in a passenger transport vehicle (1) according to claim 1 1 is executed.

A computer program product (50) configured to be loaded into a memory of a computer and comprising software code portions that perform the steps of the method of any one of claims 12 to 14 when the computer program product (50) is run on the computer ,