Methods for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle, computer program, storage medium, data signal and automated vehicle

Abstract

Procedure for the protection of persons and / or for the protection of an automated vehicle (1), the procedure comprising the steps: Detection of a stopping area (7) of the automated vehicle (1), wherein the detection of the stopping area (7) is carried out by means of first sensors (2); Detection of objects within the holding area (7), wherein the detection of the objects is carried out by means of second sensors (2), wherein the objects include persons; Detection of objects in the automated vehicle (1), wherein the detection of the objects is carried out by means of third sensors (2), wherein the objects include persons; Determination of person-specific characteristics and derivation of an individual threat assessment; Classifying the objects based on the detections of the second and third sensors, with one object class classing threatening persons, one object class classing persons in need of protection and one object class classing persons capable of defending themselves; Determine the probability that classified persons will enter or exit the automated vehicle (1) at the detected stopping area (7); Determining the risk of an encounter between a threatening person and a vulnerable person without the presence of a person capable of defending themselves and / or the threatening person having access to the automated vehicle (1); Deriving measures when a threshold value of the determined risk is exceeded.

Description

[0001] The invention relates to a method for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle. Furthermore, the invention relates to a computer program, a storage medium, a data signal, and an automated vehicle.

[0002] The following definitions apply to the entire disclosure content.

[0003] Attacks by threatening individuals on passengers occur repeatedly in public transport. Passengers are often defenseless, for example, when there is no driver in the immediate vicinity of autonomous shuttles or other automated vehicles. Another problem is vandalism of the vehicle itself by threatening individuals.

[0004] In order to offer passengers in automated vehicles, especially in automated vehicles without on-site personnel or operators, a higher level of safety, it is therefore necessary to identify threatening persons and to derive appropriate measures if the risk is high.

[0005] The object of the invention was to provide a method for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle and an automated vehicle.

[0006] The subject matter of the independent claim and the subordinate claims each solve this problem.

[0007] According to one aspect, the invention provides a method for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle, the method comprising the steps: • Detecting a stopping area of ​​the automated vehicle, wherein the detecting of the stopping area is carried out by means of first sensors; • Detecting objects within the holding area, wherein the objects are detected by means of second sensors, wherein the objects include persons; • Detecting further objects in the automated vehicle, wherein the objects are detected by means of third sensors, wherein the objects include persons; • Recording of person-specific characteristics and deriving an individual threat assessment, whereby person-specific characteristics include threatening, vulnerable and / or defensive characteristics; • Classifying the objects based on the detections of the second and third sensors, wherein there are at least three object classes for persons, wherein one object class classifies threatening persons, one object class classifies vulnerable persons and one object class classifies defensive persons; • Determining a probability of whether classified persons will board or disembark from the automated vehicle at the detected stopping area; • Determining the risk of a confrontation between a threatening person and a vulnerable person without the presence of a defensive person and / or access by the threatening person to the automated vehicle; • Derivation of measures when a threshold value of the identified risk is exceeded.

[0008] An automated vehicle includes, for example, land vehicles, in particular vehicles for transporting people, for example a bus, a shuttle, or a rail-guided vehicle, wherein the rails can also be implemented virtually or via sensor-detectable reference elements that can be arranged next to, on, and / or in a roadway, for example via a magnetic track realized by magnets arranged in and / or on the roadway. Likewise, the automated vehicle can be, for example, a watercraft, for example a water taxi, or an aircraft, for example an air taxi. The present invention relates in particular to automated vehicles with a degree of automation of Level 4 or Level 5, but can also be applied to vehicles with other degrees of automation.

[0009] A stopping area is the area in which an automated vehicle intends to stop, for example, to let one or more passengers disembark or board. The stopping area is, in particular, a close area, preferably within a radius of 50 meters, particularly preferably within a radius of 30 meters, around the automated vehicle. The stopping area can, for example, be the area of ​​a bus stop.

[0010] The first sensors, which are used to detect the stopping area, can be at least partially integrated into the automated vehicle. However, they can also be part of the infrastructure and transmit data to the automated vehicle via a communication interface. It can also be a combination. The first sensors can be, for example, imaging sensors, in particular cameras, lidar sensors, radar sensors, or other sensors.

[0011] To detect objects within the stopping area, the first sensors mentioned above can be used as second sensors, but also acoustic sensors, thermal imaging cameras, or other second sensors. Second sensors that are at least partially integrated into the automated vehicle or are part of the infrastructure can also be used to detect objects within the stopping area. A combination of these can also be used.

[0012] In addition to people, the detected objects can also be items that the person is carrying. For example, the detection of a stroller can indicate a baby in the stroller without the baby itself being detected by one of the sensors. Objects such as weapons, especially firearms, can also be detected. Other objects, such as umbrellas or sticks, can also further increase a potential threat from a potentially threatening person, as these objects can be used to inflict physical violence on other people.

[0013] To detect objects within the automated vehicle, third-party sensors integrated into the vehicle are used. These can be, for example, imaging third-party sensors, particularly cameras, but also acoustic sensors, odor sensors, or other third-party sensors.

[0014] Various factors can be used to determine whether a detected person exhibits threatening, vulnerable, or defensive characteristics. For example, loud or aggressive statements made by a person using an acoustic sensor can provide an indication of a person's threatening characteristics. Whether a statement is aggressive can be determined, for example, using a trained machine learning model, in particular an artificial neural network. However, objects described above and detected in connection with a person, such as weapons, can also provide an indication of a threatening characteristic. An odor sensor, for example, can provide an indication of whether a person is intoxicated and could therefore potentially acquire threatening characteristics.

[0015] For example, in the case of short persons, particularly those who are 150 cm or less tall, a vulnerable characteristic is likely because height is an indication that the person could be a child.

[0016] But other clues, such as toys or school bags or other objects that are recorded in connection with a person can also indicate that the person is a child and therefore highly likely to be a person in need of protection. If, for example, a walker is recorded in connection with a person, this is also an indication that this person is highly likely to be in need of protection. Adults without further clues can also be in need of protection. A person who displays defensive characteristics could, for example, be a police officer. Since defensive people, just like threatening people, often carry weapons, it can be determined whether a person is defensive using a trained machine learning model, in particular using an artificial neural network that correlates object characteristics with a person.In particular, a uniform, for example in combination with a bulletproof vest, can be an indication of a defensive person.

[0017] The classification of objects based on the detections from the second and third sensors can be performed using a trained machine learning model, in particular using an artificial neural network. However, the detection and classification of objects can also be performed with the help of an operator who, for example, receives sensor data from the second and third sensors in a control center. It is also possible for the operator to access recorded sensor data, for example, to gain a more precise impression of a person who has been in the stopping area for an extended period.

[0018] Passenger transport with the automated vehicle can, for example, be realized with the help of an electronic ticket system. Here, a passenger can, for example, book a ticket for a route between a first stopping area and a second stopping area using a smartphone. It is therefore possible to use an authentication function to determine whether a person intends to board or alight at a particular stopping area. The positioning of a person relative to an boarding point can also provide an indication that this person wishes to board the automated vehicle. If the automated vehicle is a bus or a shuttle, for example, getting up from a seat or pressing a stop request button can also be an indication that this person should leave the automated vehicle at the next stopping area.

[0019] Determining the probability of whether classified persons will enter or exit the automated vehicle at the detected stopping area can therefore be automated, for example with the help of algorithms, semi-automated, or manually, for example with the help of an operator.

[0020] Determining the risk of a confrontation between a threatening person and a vulnerable person without the presence of a defensive person is necessary because it is assumed that a threatening person would not threaten a vulnerable person in the presence of a defensive person, such as a police officer. It is assumed that a threat only occurs when no defensive person is present.

[0021] If the automated vehicle is transporting vulnerable persons, in the event that they wish to disembark at a stopping area, the risk relating to persons outside the automated vehicle must be determined, i.e. the risk of whether a threatening person is present in the stopping area.

[0022] If the automated vehicle is transporting vulnerable persons, in the event that they do not wish to disembark at a stopping area, the risk relating to persons outside the automated vehicle must be determined, i.e. the risk of whether a threatening person wishes to board the automated vehicle.

[0023] If the automated vehicle is transporting threatening persons, in the event that they wish to disembark at a stopping area, the risk relating to persons outside the automated vehicle must be determined, i.e. the risk of whether a vulnerable person wishes to board the automated vehicle or is in the stopping area for other reasons.

[0024] For example, there can be different risk levels. For example, there can be no risk, medium risk, high risk, and an unclear risk situation where, for example, an operator is consulted. For example, there is no risk if a vulnerable person wants to disembark at a stopping area and there are no people at the stopping area.

[0025] If no risk exists, it is not necessary to derive measures. However, an operator can be informed that there is no risk, for example, by means of a green light.

[0026] In the case of a medium risk, for example, if a threatening person is present at a stopping area, a possible derived measure could be to prevent a threatening person from boarding or a vulnerable person from disembarking at the stopping area. This could be achieved, for example, by keeping the doors closed or by continuing to drive and not stopping at the stopping area. In addition, the operator could be notified that a medium risk exists, for example, by means of a yellow or orange light. Furthermore, people inside the automated vehicle could be informed of the situation.

[0027] In the case of a high risk, for example if it turns out during a journey between two stopping areas that a threatening person is threatening a person in need of protection, an emergency call can also be made and an operator can be informed, for example by means of a red display.

[0028] Even if, for example, a threatening person damages or is very likely to damage at least parts of the automated vehicle, such as a sensor or a communication device, this can be considered a high risk. If vandalism by a threatening person is expected, a derived protective measure could be to prevent the automated vehicle from approaching the threatening person, deliberately avoid them, or even activate a protective system, such as a sensor's protective cover.

[0029] A derived measure, for example, starting at a medium risk, could also include an authentication requirement to authorize entry into the automated vehicle. A person's identity can be verified using a smartphone, for example. If the person orders an automated vehicle, the location of the smartphone can be shared.

[0030] When approaching the stopping area, the person's position at the stopping area can be checked to see if it matches the position transmitted by the smartphone. If there is a match, the corresponding identity is assumed. If the risk or situation is unclear, an operator can be consulted for verification.

[0031] In principle, it is possible to consult an operator at every step of the process, for example if the result is not clear or the automated vehicle does not have the necessary resources to perform the step automatically.

[0032] According to a further aspect, the invention provides a computer program for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle, the computer program comprising program instructions that cause a computer to carry out the steps of the method disclosed herein when the computer program is loaded onto the computer or executed.

[0033] The instructions of the computer program according to the invention comprise machine instructions, source code, or object code written in assembly language, an object-oriented programming language, for example, C++, or a procedural programming language, for example, C. The computer program is executed entirely or partially on an integrated circuit or in a remote system comprising a cloud. The integrated circuit can be, for example, a graphics processing unit (GPU).

[0034] According to a further aspect, the invention provides a storage medium on which the computer program for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle is stored.

[0035] The storage medium can be, for example, a non-volatile, permanent memory or a volatile memory. A storage medium can be, for example, a USB stick, a hard drive, a memory card, or a CD-ROM. However, the storage medium can also be, for example, a cloud in which the computer program is stored.

[0036] In another aspect, the invention provides a data signal that the computer program transmits for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle.

[0037] The computer program can be transferred from a cloud to a control unit of an automated vehicle via a data carrier signal, for example via a so-called over-the-air update.

[0038] According to a further aspect, the invention provides an automated vehicle, wherein the automated vehicle comprises at least one sensor, a communication interface and a computing unit or an interface to a computing unit for executing a method for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle.

[0039] The at least one sensor can be, for example, an imaging sensor, in particular a camera, a lidar, or radar sensor, for detecting a stopping zone, objects in the stopping zone, and / or objects within the automated vehicle. Alternatively or additionally, it can also be acoustic sensors, odor sensors, or other sensors. It is also possible for the automated vehicle to at least partially access sensors outside the automated vehicle and receive sensor data via an interface.

[0040] The communication interface can be used, for example, for communication between an operator located in a control center and passengers inside and / or outside the automated vehicle. Likewise, there can be a communication interface used for communication between the automated vehicle and an operator, for example, for consultation in an unclear risk situation. Likewise, there can be a communication interface to enable passengers, for example, to make an emergency call. Furthermore, there can be a communication interface to communicate with other vehicles and thus, for example, to receive information about threatening persons.It may be useful to use data from other vehicles that are currently in the vicinity of the automated vehicle, that have recently stopped at the stopping zone to which the automated vehicle intends to approach, or from vehicles that are currently further away from the automated vehicle but have detected and classified people approaching the stopping zone, such as a group of vandals. Likewise, there may be a V2X interface that allows communication between the automated vehicle and the infrastructure or sensors that are part of the infrastructure.

[0041] The communication interface can also be designed to receive information from at least one intelligent bus stop, wherein the at least one intelligent bus stop detects the surroundings of a stopping area using sensors and transmits information to the automated vehicle via the communication interface. The sensors can be various sensors as previously described for detecting objects within the stopping area. Ideally, the automated vehicle thus receives information from the intelligent bus stop before the automated vehicle reaches this stopping area. For example, the past behavior of people can be detected and transmitted, for example, that a certain person has behaved conspicuously within the past few minutes, for example due to staggering movements, aggressive behavior, or other conspicuous behavior.

[0042] A communication interface can also be used, for example, to alert local authorities or operators of stopping areas to threatening individuals, or to alert threatening individuals that their behavior is being recorded, for example, by video surveillance. This list of possible use cases for communication interfaces is not exhaustive, and there may be other or additional communication interfaces that are useful for executing a procedure to protect individuals inside and / or outside an automated vehicle and / or to protect the automated vehicle itself.

[0043] The computing unit can, for example, be a control unit. It is also possible to have two computing units, with a second computing unit being redundant in hardware and / or software to a first computing unit in order to increase the security of the system. According to one aspect, a watchdog can be executed on one of the computing units to monitor the system. This can be used to check whether the process is running reliably and, for example, to notify an operator in the event of an error. If, for example, a person tampers with a sensor or other component, or if a sensor or other component fails, the watchdog can detect this and notify an operator, who can then manually continue the process, for example by asking people in the automated vehicle for an assessment of the situation on site.

[0044] Further developments and advantageous embodiments emerge from the subclaims, the drawings and the description of preferred embodiments.

[0045] According to one aspect, the method comprises an additional step of executing at least one derived action.

[0046] Actions can be performed by the automated vehicle itself, such as keeping doors closed or not stopping at a stopping area. However, actions can also be performed by an operator, such as informing passengers about a situation or issuing instructions via a communication interface. Likewise, an operator can, for example, issue an emergency call as a measure. It is also possible for the operator to initiate the activation of actions, which the automated vehicle then executes.

[0047] It is also possible that one measure would involve bringing additional automated vehicles to a stopping area, for example, if several differently classified people or groups of people are to be transported. In this case, for example, vulnerable people can be transported separately in an automated vehicle.

[0048] According to a further aspect, the method comprises an additional step of consulting a person.

[0049] Consultation with a person may be necessary at various points within the process. Whenever a procedural step could not be performed correctly, or when a procedural step produces an unclear result, it may be necessary to consult a person. This person can be someone inside or outside the automated vehicle. It could be, for example, a passenger or an operator. Consultation may, for example, result in a procedural step being performed manually by the person.

[0050] According to a further aspect, the measure includes denying entry and / or exit to at least one person.

[0051] Denying entry and / or exit to at least one person includes keeping doors closed and / or continuing to drive and not stopping at a stopping area.

[0052] According to another aspect, the measure includes making an emergency call. Making an emergency call is necessary, for example, if the procedure has identified a high risk of a confrontation between a threatening person and a vulnerable person without the presence of a defensive person and / or without the threatening person having access to the automated vehicle. It is possible for an operator to make the emergency call, but it is also possible for persons inside or outside the automated vehicle to be prompted by the automated vehicle or the operator to make an emergency call or at least to communicate with an emergency call center if the automated vehicle itself makes the emergency call.

[0053] The invention is explained by way of example using the following exemplary embodiments. They show: Fig. 1 a flowchart of a method for protecting persons inside and / or outside an automated vehicle and / or for protecting the automated vehicle, Fig. 2 an example of a schematic representation of an automated vehicle, Fig. 3a a tabular representation of exemplary situations when exiting at a stopping area, Fig. 3b a tabular representation of exemplary situations when boarding at a stopping area, Fig. 4a a schematic exemplary representation of an automated vehicle and a stopping area, Fig. 4b shows a further schematic exemplary representation of an automated vehicle and a stopping area, Fig. 5 a schematic exemplary representation of an automated vehicle, a stopping area and an operator.

[0054] In the figures, identical reference symbols designate identical or functionally similar objects. To avoid repetition, only the relevant objects are identified by reference symbols.

[0055] The Fig. The method shown in Figure 1 for protecting persons inside and / or outside an automated vehicle 1 and / or for protecting the automated vehicle 1 comprises a step V1 of detecting a stopping area 7 of the automated vehicle 1, wherein the detecting of the stopping area 7 takes place by means of first sensors 2. Furthermore, the method comprises a step V2 of detecting objects within the stopping area 7, wherein the detecting of the objects takes place by means of second sensors 2, wherein the objects comprise persons. The method also comprises a step V3 of detecting objects in the automated vehicle 1, wherein the detecting of the objects takes place by means of third sensors 2, wherein the objects comprise persons, and a step V4 of determining a probability as to whether one of the detected persons has predominantly threatening, vulnerable, or defensive characteristics.The method also includes a step V5 of classifying the objects based on the detections of the second and third sensors, wherein there are at least three object classes for people, wherein one object class classifies threatening people 5, one object class classifies vulnerable people 6, and one object class classifies defensive people. Furthermore, the method includes a step V6 of determining a probability of whether classified people will enter or exit the automated vehicle 1 at the detected stopping area 7, and a step V7 of determining a risk of an encounter between a threatening person and a vulnerable person without the presence of a defensive person and / or the threatening person having access to the automated vehicle 1.Furthermore, the method comprises a step V8 of deriving measures if a threshold value of the determined risk is exceeded.

[0056] The Fig. The example of a schematic representation of an automated vehicle 1 shown in Figure 2 shows, by way of example, a bus as an automated vehicle 1. According to Fig. 2, the automated vehicle 1 comprises a sensor 2, a communication interface 3 and a computing unit 4. The sensor 2 is located according to Fig. 2 in the interior of the bus in the area behind the windshield. Therefore, the sensor 2 could, for example, be a camera that can capture data from the bus interior as well as the surroundings and the stopping area 7. The communication interface 3 is in Fig. 2 is shown as an example as a radio interface and the computing unit 4 as a control unit.

[0057] The Fig. The tabular representation of exemplary situations when exiting at a stopping area 7 shown in Figure 3a shows people who can be in the stopping area 7 and people who are in an automated vehicle 1 and who want to leave the automated vehicle 1 at the stopping area 7. People who are in the stopping area 7 can either stay at the stopping area 7 or enter the automated vehicle 1. The density of the Fig. The points shown in Figure 3a reflect the probability of a threatening situation. As the density increases, the threatening nature of the situation changes from non-critical to slightly critical to highly critical. If there is neither a person in the stopping area 7 nor a person in the automated vehicle 1, no threatening situation is to be assumed. If a threatening person 5 is in the stopping area 7 or in the automated vehicle 1, a threatening situation can arise, for example, due to vandalism by the threatening person 5. A risk can also exist for vulnerable persons 6, as, for example, the emotional state of a vulnerable person can change depending on the situation and a vulnerable person 6 can also become a threatening person 5. It is also possible that a vulnerable person 6 was incorrectly classified because no external threatening characteristics were recognizable.

[0058] The Fig. The tabular representation of exemplary situations when boarding at a stopping area 7 shown in Figure 3b shows people who are in the stopping area 7 and want to board the automated vehicle 1 as well as people who are in the automated vehicle 1. People who are in the automated vehicle 1 can either get off at the stopping area 7 or remain in the automated vehicle 1. The density of the Fig. The points shown in Figure 3b reflect the probability of a threatening situation. As the density increases, the threatening nature of the situation changes from non-critical to slightly critical to highly critical. If there is neither a person in the stopping area 7 nor a person in the automated vehicle 1, no threatening situation is to be assumed. If a threatening person 5 is in the stopping area 7 or in the automated vehicle 1, a threatening situation can arise, for example, due to vandalism by the threatening person 5. A risk can also exist for vulnerable persons 6, since, for example, the emotional state of a vulnerable person can change depending on the situation and a vulnerable person 6 can also become a threatening person 5. It is also possible that a vulnerable person 6 was incorrectly classified because no external threatening characteristics were recognizable.

[0059] The Fig. The schematic exemplary representation of an automated vehicle 1 and a stopping area 7 shown in Fig. 4a shows, by way of example, only persons 6 in need of protection within the automated vehicle 1. According to Fig. 4a, the persons in need of protection 6 are, for example, a child and a pregnant woman. At the stopping area 7, Fig. 4a shows a threatening person 5, who is depicted using a baseball bat as a weapon. According to Fig. 4a, a threatening situation may occur if either a vulnerable person 6 gets out at the stopping area 7 or the threatening person 5 gets into the automated vehicle 1. The possible entry and exit is Fig. 4a is visualized with a double arrow. As already shown in Fig. 2 also includes the Fig. 4a, the automated vehicle 1 comprises a sensor 2, a communication interface 3 and a computing unit 4. The sensor 2 is also located according to Fig. 4a in the interior of the bus in the area behind the windshield. Therefore, the sensor 2 could, for example, be a camera that can capture data from the bus interior as well as the bus surroundings and the stopping area 7. The communication interface 3 is in Fig. 4a is also shown as an example as a radio interface and the computing unit 4 as a control unit.

[0060] The Fig. 4b shows a further schematic exemplary representation of an automated vehicle 1 and a holding area 7, which shows, for example, a threatening person 5 within the automated vehicle 1, who is shown with a baseball bat as a striking weapon. According to Fig. 4b exclusively for persons in need of protection 6. According to Fig. 4b, the persons in need of protection 6 are, for example, a child and a pregnant woman. Fig. 4b, a threatening situation may occur if either the threatening person 5 gets out at the stopping area 7 or a person in need of protection 6 gets into the automated vehicle 1. The possible entry and exit is Fig. 4b is visualized with a double arrow. As already shown in Fig. 2 also includes the Fig. 4b, the automated vehicle 1 comprises a sensor 2, a communication interface 3 and a computing unit 4. The sensor 2 is also located according to Fig. 4b in the interior of the bus in the area behind the windshield. Therefore, the sensor 2 could, for example, be a camera that can capture data from the bus interior as well as the bus surroundings and the stopping area 7. The communication interface 3 is in Fig. 4b is also shown as an example of a radio interface and the computing unit 4 as a control unit.

[0061] The Fig. The schematic exemplary representation of an automated vehicle 1, a holding area 7 and an operator 8 shown in Figure 5 shows, by way of example, a person 6 in need of protection within the automated vehicle 1. According to Fig. 5, the person in need of protection 6 is, for example, a wheelchair user. At the stopping area 7, Fig. 5 a threatening person 5, who is depicted with a baseball bat as an example of a weapon. There are also other people in the holding area who were neither classified as needing protection nor as threatening nor as capable of defending themselves. As already mentioned in Fig. 2 also includes the Fig. 5, the automated vehicle 1 comprises a sensor 2, a communication interface 3 and a computing unit 4. The sensor 2 is also located according to Fig. 5 in the interior of the bus in the area behind the windshield. Therefore, the sensor 2 could, for example, be a camera that can capture data from the bus interior as well as the bus surroundings and the stopping area 7. The communication interface 3 is in Fig. 5 is also shown as an example of a radio interface and the processing unit 4 as a control unit. According to Fig. 5, the stopping area 7 includes an intelligent bus stop, which is Fig. 5 is shown as an example with a camera that can communicate with both the automated vehicle 1 and the operator 8. Likewise, in Fig. 5 shows by way of example that the automated vehicle 1 can communicate with the operator 8 and the operator 8 can also communicate with the automated vehicle 1 and the intelligent bus stop. Reference symbol V1 - V8 process steps 1 automated vehicle 2 sensors 3 Communication interface 4 Computing unit 5 threatening person 6 vulnerable person 7 Holding area 8 operators

Claims

[1] A method for protecting persons inside and / or outside an automated vehicle (1) and / or for protecting the automated vehicle (1), the method comprising the steps: • Detecting a stopping area (7) of the automated vehicle (1), wherein the detecting of the stopping area (7) is carried out by means of first sensors (2); • Detecting objects within the holding area (7), wherein the objects are detected by means of second sensors (2), wherein the objects include persons; • Detecting objects in the automated vehicle (1), wherein the objects are detected by means of third sensors (2), wherein the objects include persons; • Determination of person-specific characteristics and derivation of an individual threat assessment, whereby person-specific characteristics include threatening, vulnerable and / or defensive characteristics; • Classifying the objects based on the detections of the second and third sensors, wherein there are at least three object classes for persons, wherein one object class classifies threatening persons (5), one object class classifies persons in need of protection and one object class classifies defensive persons; • Determining a probability of whether classified persons will enter or exit the automated vehicle (1) at the detected stopping area (7); • Determining a risk of an encounter between a threatening person and a vulnerable person without the presence of a defensive person and / or access by the threatening person to the automated vehicle (1); • Derivation of measures when a threshold value of the identified risk is exceeded. [2] The method of claim 1, wherein the method comprises an additional step of executing at least one derived action. [3] A method according to any one of the preceding claims, wherein the method comprises an additional step of consulting a person. [4] Method according to one of the preceding claims, wherein the measure comprises denying entry and / or exit of at least one person. [5] Method according to one of the preceding steps, wherein the measure comprises making an emergency call. [6] Computer program for protecting persons inside and / or outside an automated vehicle (1) and / or for protecting the automated vehicle (1), the computer program comprising program instructions which cause a computer to carry out the steps of a method according to one of the preceding claims when the computer program is loaded onto the computer or executed. [7] Storage medium on which the computer program according to claim 6 is stored. [8] Data signal transmitting the computer program according to claim 6. [9] Automated vehicle (1), wherein the automated vehicle (1) comprises at least one sensor (2), a communication interface (3) and a computing unit (4) or an interface to a computing unit (4) for carrying out the method according to one of claims 1 to 5.

Images