Automatic door system and method for operating said system
Patent Information
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- ASSA ABLOY ENTRANCE SYST AB
- Filing Date
- 2024-07-09
- Publication Date
- 2026-05-20
AI Technical Summary
Existing automatic door systems face challenges in providing convenient, reliable, and safe operation, particularly in monitoring trigger and protection zones, due to limitations in sensor accuracy and reliability under environmental disturbances.
The implementation of a method that combines sensor data from different sensing principles, such as cameras and radar, to provide a more complete understanding of the environmental sensing fields, allowing for accurate door operation commands, and the use of a control unit that processes this data to generate commands for opening, closing, or stopping the door, potentially using self-learning models for improved performance.
This approach enhances the safety and reliability of automatic door systems by providing accurate object classification, tracking, and scene rating, reducing the risk of collisions and trapping, and allowing for energy-efficient operation, even in adverse environmental conditions.
Smart Images
Figure 00000023_0000 
Figure 00000023_0001 
Figure 00000024_0000
Abstract
Description
[0001] AUTOMATIC DOOR SYSTEM AND METHOD FOR OPERATING
[0002] SAID SYSTEM
[0003] Field of the invention
[0004] The invention relates to a method for operating an automatic door system, to a control unit for said automatic door system, to an automatic door system and to a computer program (product) relating to said method.
[0005] Background
[0006] Automatic door systems including an actuated door are widely known. Those door systems / doors are configured to open and close automatically using an actuator (e.g. an electro motor), a sensor and / or a remote control. These systems / doors are commonly used in public buildings, such as shopping centers, hospitals, airports, and hotels, as well as in private residences and commercial settings. Automated doors provide convenience and accessibility for people with disabilities, and they can also help to improve security and energy efficiency.
[0007] For providing a convenient and safe operation of the door typically a trigger zone and a protection zone are monitored individually. If an object, such as a pedestrian, is detected in the trigger zone the door is opened. The protection zone is monitored to prevent collisions with objects passing the door and / or being in close proximity to the door. Typically, this protection zone extends at least 20 cm around the door, the door’s main closing edge and / or the door’s secondary closing edge. The main closing edge of a door refers to the edge of a respective door leaf that comes into contact first with e.g. a door frame or a main closing edge of a further door leaf when the door is closing. The secondary closing edge of a door is typically the edge opposite to the main closing edge. In an actuated door, the main closing edge and the secondary closing edge may be surveilled with sensors or safety devices that detect obstructions or pressure and prevent the door from moving when an obstruction is detected. These sensors may e.g. be mounted to the closing edge(s), the secondary closing edge(s), or may be installed above the respective door. Due to the surveillance, the risk of trapping, damaging or injuring people / objects by the automatically moving door can be minimized.
[0008] Known automatic door systems provide individual sensors for monitoring the trigger zone and for monitoring the protection zone, including the monitoring of the main and / or secondary closing edges. For monitoring the trigger zone, typically radar sensors and / or cameras are used, wherein the protection zone is typically monitored using IR- sensors.
[0009] It is an object of the present invention to improve and facilitate the monitoring of automatically actuated doors. Particularly, the automated operation of those doors shall become more convenient, reliable and / or safe.
[0010] Summary
[0011] This object is achieved by a method for operating an automatic door system, by a control unit for said automatic door system, by an automatic door system and by a computer program (product), as specified in the independent claims. Further aspects of the invention are given in the dependent claims as well as in the following description.
[0012] Particularly, the object is achieved by a method for operating an automatic door system according to claim 1. The automatic door system comprises an actuated door, a first sensor and a second sensor, wherein the first sensor has a different sensing principle than the second sensor.
[0013] Further, the automatic door system comprises a control unit. The control unit may include software and / or hardware and is configured to control an operation of the door (e.g. opening, closing and / or stopping) based on sensor data acquired from the first and second sensors. Further data may also be considered, such as environmental parameters, a user input (e.g. a chosen configuration or specific application) a status of an emergency button and / or the like. In a particular aspect, the control unit may be a distributed control unit.
[0014] The door may be a swing door, a sliding door, a folding door, a pocket door, a revolving door and / or the like. The door may have a single door leaf, or multiple door leafs (at least two), such as a French door or a double sliding door.
[0015] The control unit may be configured to control the operation of a single door or of multiple doors (at least two). For example, the control unit may be a centralized control unit, being configured to control the entrance (or internal) doors of a particular building, such as a shopping center, a hospital, and airport, and / or the like.
[0016] The method according to the present invention comprises the following steps:
[0017] - acquiring sensor data of the first sensor, the first sensor being assigned to a first environmental sensing field;
[0018] - acquiring sensor data of the second sensor, the second sensor being assigned to a second environmental sensing field, wherein the first and second environmental sensing fields overlap at least partially;
[0019] - combining sensor data of the first and second sensors, thereby providing a more complete understanding of the environmental sensing fields being monitored; based on the combined sensor data, generating a door operation command, and
[0020] - outputting the door operation command for operating the door.
[0021] Combining sensor data (also denoted as sensor fusion) allows to get a more complete understanding of the environmental sensing fields being monitored. Particularly, by combining sensor data from multiple sensors (here at least first and second sensors) allows to provide a more accurate, reliable, and / or informative representation of the environmental sensing fields being monitored. Accordingly, additional information can be generated, that goes beyond the information obtained from each sensor individually.
[0022] For example, sensor data from a camera (first sensor) and a radar (second sensor) can be combined not only to identify an object (e.g. a pedestrian) but to provide additional attributes of the identified object, such as position, velocity, acceleration and / or direction of movement.
[0023] Further, providing a radar and a camera allows monitoring and operating the door even if one of the sensors is disturbed, e.g. due to environmental disturbances, such as darkness, fog, smog, smoke, and the like. In case of such a disturbance, the door may be operated based on one type of sensor data only, to provide e.g. for an emergency exit.
[0024] Particularly, the sensor data of the first sensor may include image data, wherein this image data may be processed prior to the step of combining. The image data may be gathered in the different spectra, such as IR, visible and / or UV. Image processing includes the extraction of useful information from the captured image and / or improvement of image quality (such as contrast stretching, histogram equalization, filtering, and / or the like). The extracted information may include texture recognition, object recognition, object classification and / or the like. For example, image processing may provide information about persons and / or objects being in proximity to the door (i.e. in a first environmental sensing field). Those may be, but are not limited to, a pedestrian, a disabled person, a vehicle, an animal and / or the like. These type of objects may be further classified in persons with walking aids, wheelchair users, children, elderlies, passengers with luggage, vehicle types (e.g. a cleaning trolley, a conveyor vehicle, ...), baby carriages, a luggage carts, pets, wildlife and / or the like.
[0025] Even further, image processing may provide information about the door itself. For example, the movable parts of the door (e.g. the door leaf(s)) may be recognized as objects.
[0026] The sensor data of the second sensor may include movement data of an object such as a velocity, an acceleration, a position, a distance to the door and / or a direction of movement. This movement data may be acquired using a radar, a stereo camera, a Time of Flight Camera, a Lidar, and / or the like.
[0027] The combined sensor data may include an object classification, an object tracking and / or a scene rating. The door may then only be opened, if a predefined object is classified or tracked and / or a predefined scene is rated.
[0028] For example, movement data acquired by the second sensor may be assigned with an object being recognized and / or classified by processed image data. Further, individual movement data can be assigned to respective objects being recognized and / or classified by processed image data. Hence, a scene can be identified and rated. A possible scene includes a recognition of a pedestrian and its willingness to pass the door. In case it is determined that the pedestrian is willing to pass the door, e.g. based on the pedestrian’s position, direction of movement and / or velocity, a respective door operation command can be generated and outputted so that the door is opened and closed again after the pedestrian has passed the door. In case it is determined that the pedestrian seeks to pass by, the door can remain closed. Cross-traffic detection is therefore possible.
[0029] In a further exemplary scene it can be determined that the object coming towards the door is no pedestrian but e.g. a free-ranging animal, the door can remain closed, even though the animal’s movement data indicate that the animal is willing to pass the door. Thus, generating the door opening command may be dependent on the type of object being recognized.
[0030] Further, different objects and assigned movement data can be considered for the scene rating and a subsequent generation of a door operation command. Hence, the door can e.g. be left open for a desired amount of time to allow all objects passing the door, without undesired or disturbing door movements while the objects passing the door.
[0031] In an even further example, the door may be closed again between two objects passing the door, as it was determined that the first object moves significantly faster than the second object. Thus, energy can be saved as the time the door being open is reduced.
[0032] In case the door (e.g. movable parts of the door, such as door leaf(s)) are recognized as objects, the door, the door leaf(s) and / or the door’s main and / or secondary closing egdes can be tracked. Hence, actual movement data of the door is available, such as position, velocity, acceleration and / or direction of movement and can also be considered for generating a door operation command.
[0033] Further, object tracking and / or scene rating increase security of the door, as collisions can be effectively prevented as there is a better understanding of the door’s environment and the development over time of the door’s environment and / or of the door’s actual movement data. The method may further include comparing sensor data and / or combined sensor data with an environmental model for generating a door operation command. The environmental model may include predefined scenes and / or objects the sensor data and / or combined sensor can be compared with. The comparison of the sensor data and / or combined sensor data with an environmental model may include a probability analysis on whether the sensor data matches a predefined scene and / or object. Depending on the outcome of the probability analysis the door operation command may be an opening command, closing command and / or a stopping command.
[0034] Further, the model may be a self-learning model, which may be based on an artificial intelligence (Al) system that can learn and improve its performance without human intervention or explicit programming. Particularly, the selflearning model may be based on a deep learning and / or reinforcement learning technique, and may e.g. use large amounts of training data to iteratively improve the performance over time. The training data may be acquired from other doors being part of the automatic door system and / or from doors of other automatic door systems. Further, the training data may be based on historical data and / or may include newly acquired data.
[0035] The acquired sensor data may be combined using at least one of the following: Kalman filtering, a Bayesian network, Dempster-Shafer theory, an artificial neural network, such as a convolutional neural network, a Gaussian process, and / or a Fussy logic.
[0036] Further, the method may include the step of determining based on the combined sensor data, whether the door can be operated safely and whether the door needs to be operated. By evaluating both conditions, i.e. safe operation and needed operation, the automated operation of the door becomes more convenient, reliable and / or safe. Particularly trapping of objects / persons and / or collisions can be effectively prevented, so that no separate monitoring of a protection zone and / or the closing edges (main closing edge and / or secondary closing edge) of the door is required. Hence, the system and the operation of the door can be facilitated.
[0037] In an aspect, the door operation command may include at least one of the following: an opening width of the door, an opening position of the door, an opening speed of the door, a keep-open time of the door, a not-opening command and / or a stop-command.
[0038] Further, in case a stop-command is generated, at least one of the following measures may be taken by the control unit:
[0039] - a warning is outputted at the door, in particular the warning being an acoustic warning and / or a visual warning,
[0040] - a notification is sent to a supervisor of the owner of the door system,
[0041] - a visual indication is displayed in a control interface for the automatic door system,
[0042] - a maintenance instruction is sent to a technician of the owner or the manufacturer of the door system, and / or
[0043] - a notification is sent to the manufacturer of the door system.
[0044] Dependent on the event having caused the stop-command, the door can immediately be operated again or may need inspection.
[0045] Dependent on the combined sensor data (e.g. an object tracking and / or a scene rating) the opening position of the door can be chosen. For example, a sliding door with multiple door leafs may be operated in way that the door leafs do not open symmetrically but so that the door opens at a desired position, only. For example, only one door leaf is moved, or the multiple door leafs are moved independently from one another. Further the opening speed can be chosen, dependent on the combined sensor data, including e.g. type of object, distance to the door and velocity of the object, so that a reliable opening, without collision can be achieved. As already outlined above, the time the door is kept open (keep-open time) can be determined, depending on the scene as well as the decision not to open the door (not-opening command). Further, in case combined sensor data predict a collision, a stop-command can be generated in order to prevent the predicted collision.
[0046] According to a further aspect, the first sensor may be an optical sensor, particularly a camera (e.g. IR, visible range, or UV), and wherein the second sensor is a radar sensor, particularly a Frequency Modulated Continuous Wave, FMCW, radar sensor.
[0047] A FMCW radar sensor is a type of radar-based sensor that emits a continuous radio signal with a varying frequency. By measuring the frequency shift of the reflected signal, the sensor can calculate movement data, such as a distance and / or a velocity of object(s) in its field of view, i.e. the second environmental sensing field.
[0048] The automatic door system may further include at least a third sensor. The at least third sensor may have a different sensing principle than the first sensor and the second sensor, or may be of the same type as the first or second sensor while being assigned to a third environmental sensing field. The method may include further following steps:
[0049] - acquiring sensor data of the third sensor, the third sensor being assigned to a third environmental sensing field, wherein the third environmental sensing field overlaps at least partially with the first and / or second environmental sensing field, and combining sensor data of the at least third sensor with sensor data of the first and / or second sensors. In a particular example, the third sensor may include a photocell, an infrared sensor, a laser scanner, a light curtain, a radar, an electromechanical contact sensor, in particular an electromechanical safety edge, a button, a switch, in particular an emergency stop switch, a lever, in particular an emergency brake release lever, and / or an interface for receiving a signal indicating the open or closed state of another door.
[0050] The first environmental sensing field and the second environmental sensing field may be arranged on an internal area and / or external area, wherein the door separates the internal area from the external area, when being closed. Accordingly, objects may be tracked and scenes of both sides of the door may be rated for operating the door of the automatic door system.
[0051] Further, a first set of first and second sensors may be assigned to an internal area, and a second set of first and second sensors may be assigned to an external area, wherein the door separates the internal area from the external area, when being closed. Hence, separate sensors, respectively sets of sensors may be provided for the internal area and the external area. For each of the sets, i.e. the first set and / or the second set, the method described above may be performed. Further, sensor data of the first set may be combined with sensor data of the second set, thereby providing a more complete understanding of the environmental sensing fields being monitored. For example, the door may be opened if an identified object comes near the door and may be closed when it was verified that the object has left the door again.
[0052] In a particular aspect of the invention, there are at least two first sensors (e.g. cameras), wherein first one of the first sensors is assigned to a first internal environmental sensing field and a second one of the first sensors is assigned to a first external environmental sensing field. Further, there may be provided at least one second sensor. The second sensor, e.g. a radar sensor, may be assigned to a second environmental sensing field, spanning the internal and external area of the door. For example, the second sensor may have a substantially vertical sensing orientation, wherein the sensing orientation of the at least two first sensors may be angled (e.g. 45°).
[0053] The object is further achieved by a control unit for an automatic door system, wherein the control unit is configured to carry out the method described above. The control unit may include software and / or hardware and is configured to control an operation of the door. Further, the control unit may be located in a single spot (e.g. in a housing) or may be a distributed control unit, i.e. the control unit may be distributed over several spots (at least two). For example, sensor data acquired by the first and / or second sensor(s) may processed by a first part of the control unit and may then be sent to a second part, which is located at a different spot. The second part of the control unit may then carry out the combining step and / or the generation of a door operation command. Further, the generation of the door operation command can also be performed at the first or a third part of the control unit. In a particular example, the control unit (or at least some part(s)) thereof are housed in a housing of the first and / or second sensor(s).
[0054] In a particular aspect, the control unit may be configured to control the operation of multiple doors. In this case, e.g. the model may be trained with data acquired by all doors of the automatic door system being controlled by the control unit.
[0055] The control unit may be arranged remotely from the at least one door to be operated. In a particular example, the control unit is part of a server connected to the door(s) to be operated. The connection may be wired and / or wireless.
[0056] The object is further achieved by an automatic door system, comprising at least one door, at least a first sensor and at least a second sensor and a control unit. The control unit being configured to control the operation of the at least one door. Further, the system is configured to carry out the method outlined above. Further, the object is achieved by a computer program, comprising instructions, which when carried out by a processor cause the processor to perform operations of the above method. The processor may be part of the control unit.
[0057] The object is also achieved by a computer program product, embodied on a computer readable storage medium and configured so as when executed on a processor to perform operations of the above method.
[0058] Brief description of the figures
[0059] The present disclosure will be more readily appreciated by reference to the following detailed description when being considered in connection with the accompanying drawings in which:
[0060] Fig. 1A shows a schematic view of an automatic door system (a sliding door being closed);
[0061] Fig. IB shows a schematic view of the automatic door system (the sliding door being opened);
[0062] Fig. 2 A shows a schematic view of an automatic door system, including a sliding door;
[0063] Fig. 2B shows a schematic view of an automatic door system, including a revolving door;
[0064] Fig. 2C shows a schematic view of an automatic door system, including a swing door;
[0065] Fig. 3 shows a schematic illustration of a method for operating an automatic door system, and
[0066] Fig. 4 shows the method steps of acquiring sensor data and combining sensor data in more detail. Detailed description of the figures
[0067] Figs. 1A and IB show a schematic view of an automatic door system 1 including a sliding door 10, having two door leafs 12, 14. The sliding door 10 is provided in a door opening 22 in a wall 20, e.g. of a building. In Fig. 1A, the door 10 is closed and the main closing edges of the door leafs 12, 14 are in contact. In Fig. IB, the door is opened. The opening of the door may be performed via the inventive method.
[0068] The door leafs 12, 14 are supported in a door leaf support 16, here including a guide rail. For actuating the door, the door leafs may be driven e.g. by an electric motor.
[0069] The automatic door system 1 further comprising a set 100 of sensors, including a first sensor and a second sensor. The first sensor has a different sensing principle than the second sensor and may be a camera, while the second sensor may be a radar sensor. A control unit 200 is configured to control an operation of the door (such as opening, closing, stopping the door) based on sensor data acquired from the first and second sensors.
[0070] The first sensor is assigned to a first environmental sensing field 110 and the second sensor is assigned to a second environmental sensing field 120. The first and second environmental sensing fields 110, 120 overlap at least partially. In a particular aspect, first and second environmental sensing fields may be identical.
[0071] The sensor data acquired by the first sensor may include image data. According to a particular aspect, the image data may be processed (such as enhancement of image quality and / or extraction of useful information). The sensor data of the second sensor may include movement data of an object, such as a velocity, an acceleration, a position, a distance to the door and / or a direction of movement.
[0072] The control unit 200 may be configured to perform the following method: acquiring sensor data of the first sensor, acquiring sensor data of the second sensor, combining sensor data of the first and second sensors, thereby providing a more complete understanding of the environmental sensing fields 110, 120 being monitored; based on the combined sensor data, generating a door operation command, and outputting the door operation command for operating the door.
[0073] Accordingly, the door 10 can be operated automatically, using the combined sensor data, wherein the combined sensor data includes an object classification, an object tracking and / or a scene rating. Different scenes and types of doors that can be operated are depicted in Figs. 2A to 2C.
[0074] Fig. 2 A shows a schematic view of an automatic door system 1, including a sliding door 10. Fig. 2B shows a schematic view of an automatic door system 1, including a revolving door having a revolving door leaf 17, and Fig. 2C shows a schematic view of an automatic door system, including a swing door having a swing door door leaf 19. The door leafs 12, 14, 17 and 19 (and / or main and / or secondary closing edges thereof) may be recognized as objects and movement data acquired by a second sensor may be assigned with the respective door leaf and / or the door’s main and / or secondary closing edge(s).
[0075] The sliding door shown in Fig. 2 A may be the sliding door already shown in Figs. 1A and IB.
[0076] According to Figs. 2A to 2C there is provided a first internal environmental sensing field 110a and a second internal environmental sensing field 120a as well as a first external environmental sensing field 110b and a second external environmental sensing field 120b. The internal sensing fields 110a, 120a may be monitored by a first set of first and second sensors and the external sensing fields 110b, 120b may be monitored by a second set of first and second sensors, wherein above method may be performed for the first set and the second set separately, and / or wherein the sensor data of the first set may be combined with sensor data of the second set.
[0077] Further, there may be a first sensor monitoring the first internal and external environmental sensing fields 110a, 110b and / or a second sensor monitoring the second internal and external environmental sensing fields 120a, 120b.
[0078] In a particular example, there may be a radar (second sensor) monitoring the second internal and external environmental sensing fields 120a, 120b, while separate cameras (first sensors) are provided for monitoring the first internal environmental sensing field 110a and the first external environmental sensing filed 110b, respectively.
[0079] For an automated operation of the door 10, objects, such as the door leafs 12, 14, 17 or 19 and / or persons 30, 32 may be recognized and movement data may be assigned to the respective objects.
[0080] For example, image processing may provide information about persons 30, 32 being in proximity to the door 10 (i.e. in a respective environmental sensing field). For example, it may be recognized that object 30 is pedestrian pushing a baby carriage, wherein object 32 is a pedestrian. The sensor data of the second sensor may include movement data of the respective objects such as a velocity, an acceleration, a position, a distance to the door and / or a direction of movement. The combined sensor data may include an object tracking and / or a scene rating. For example, movement data acquired by the second sensor may be assigned with an object being recognized and / or classified by processed image data. Here, it can be concluded that object 30 moves slow, wherein object 32 is moving fast. The door can then be operated (by generating and outputting a respective command) in order to allow both persons to pass the door conveniently.
[0081] Fig. 3 shows a schematic illustration of a method 1000 for operating an automatic door system. The method comprises
[0082] - acquiring 1200 sensor data of a first sensor 102 of an automatic door system 1 (as e.g. shown in Figs. 1A to 2C), wherein the first sensor 102 is assigned to a first environmental sensing field 110, 110a, 110b.
[0083] - acquiring 1200 sensor data of a second sensor 104 of the automatic door system 1, wherein the second sensor 104 is assigned to a second environmental sensing field 120, 120a, 120b, wherein the first and second environmental sensing fields overlap at least partially. The first and second environmental sensing fields represent an environment 1100 of interest.
[0084] - combining 1300 sensor data of the first and second sensors 102, 104, thereby providing a more complete understanding of the environmental sensing fields 110, 120 being monitored;
[0085] - based on the combined sensor data, generating 1400 a door operation command, and
[0086] - outputting 1500 the door operation command for operating the door.
[0087] Fig. 4 shows the method steps of acquiring sensor data 1200 and combining sensor data 1300 in more detail. Particularly, sensor data may be acquired by a first sensor 102 (e.g. a camera), a second sensor (e.g. a radar) and optionally at least one third sensor 106, 108 (e.g. also a camera, having a different viewing angle and / or sensing in a different spectrum than the first sensor).
[0088] Sensor data of the first, second and optionally third sensors 102, 104, 106 is combined (Step 1320), thereby providing a more complete understanding of the environmental sensing fields being monitored. Further, sensor data and / or combined sensor data of any one of sensors 102, 104, 106, 108 may be compared with an environmental model 150. The model 150 may be a selflearning model. Here sensor data of sensor 108 is compared with the model 150 (step 1340). The combined data of steps 1320 and 1340 may then be combined in subsequent combining step 1360. Based on the data of step 1360, a door operation command can then be generated (step 1400).
[0089] Some of the embodiments contemplated herein are described more fully with reference to the accompanying drawings. Other embodiments, however, are contained within the scope of the subject matter disclosed herein. The disclosed subject matter should not be construed as limited to only the embodiments set forth herein; rather, these embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.
[0090] The present invention may, of course, be carried out in other ways than those specifically set forth herein without departing from essential characteristics of the invention. The present embodiments are to be considered in all respects as illustrative and not restrictive, and all changes coming within the meaning and equivalency range of the appended claims are intended to be embraced therein.
[0091] List of reference signs
[0092] 1 automatic door system
[0093] 10 actuated door
[0094] 12 door leaf (sliding door)
[0095] 14 door leaf (sliding door)
[0096] 16 door leaf support
[0097] 17 door leaf (revolving door)
[0098] 19 door leaf (swing door)
[0099] 20 wall
[0100] 22 wall opening 0 object 2 object
[0101] 100 set of sensors (or combined sensor)
[0102] 102 sensor (e.g. camera)
[0103] 104 sensor (e.g. radar)
[0104] 106 sensor
[0105] 108 sensor
[0106] 110 first environmental sensing field
[0107] 110a first environmental sensing field (internal)
[0108] 110b first environmental sensing field (external)
[0109] 120 second environmental sensing field
[0110] 120a second environmental sensing field (internal) 120b second environmental sensing field (external) 150 model 00 control unit
[0111] 1000 method
[0112] 1100 environment
[0113] 1200 acquiring sensor data
[0114] 1300 combining sensor data
[0115] 1320 combining sensor data
[0116] 1340 comparing sensor data and model
[0117] 1360 combining sensor data
[0118] 1400 generating a door operation command
[0119] 1500 outputting the door operation command
Claims
Claims1. A method for operating an automatic door system (1), the automatic door system (1) comprising an actuated door (10); a first sensor (102) and a second sensor (104), the first sensor (102) having a different sensing principle than the second sensor (104), and a control unit (200) configured to control an operation of the door (10) based on sensor data acquired from the first and second sensors (102, 104), wherein the method comprises the following steps: acquiring (1200) sensor data of the first sensor (102), the first sensor (102) being assigned to a first environmental sensing field (110); acquiring (1200) sensor data of the second sensor (104), the second sensor (104) being assigned to a second environmental sensing field (120), wherein the first and second environmental sensing fields (110, 120) overlap at least partially; combining (1300) sensor data of the first and second sensors (102, 104), thereby providing a more complete understanding of the environmental sensing fields (110, 120) being monitored; based on the combined sensor data, generating (1400) a door operation command, and outputting (1500) the door operation command for operating the door.
2. The method according to claim 1, wherein the sensor data of the first sensor (102) includes image data, wherein the image data may be processed prior to the step of combining, and / or wherein the sensor data of the second sensor (104) includes movement data of an object (30, 32), such as a velocity, an acceleration, a position, a distance to the door and / or a direction of movement.
3. The method according to any preceding claim, wherein the combined sensor data includes an object classification, an object tracking and / or a scene rating.
4. The method according to any preceding claim, wherein the method further includes comparing sensor data and / or combined sensor data with an environmental model (150) for generating (1400) a door operation command.
5. The method according to claim 4, wherein the model (150) is a selflearning model.
6. The method according to any preceding claim, wherein the sensor data is combined using at least one of the following:Kalman filtering; a Bayesian network;Dempster-Shafer theory; an artificial neural network, such as a convolutional neural network; a Gaussian process, and / or a Fussy logic.
7. The method according to any preceding claim, further including the step of determining based on the combined sensor data, whether the door (10) can be operated safely and whether the door needs to be operated.
8. The method according to any preceding claim, wherein the door operation command includes at least one of the following: an opening width of the door (10); an opening position of the door (10); an opening speed of the door (10); a keep-open time of the door (10), a not-opening command, aa stop-command, a close-command.
9. The method according to any preceding claim, wherein the first sensor is an optical sensor, particularly a camera, and wherein the second sensor is a radar sensor, particularly a FMCW-radar sensor.
10. The method according to any preceding claim, wherein the automatic door system (1) further includes at least a third sensor (106, 108), the at least third sensor (106, 108) may have a different sensing principle than the first sensor (102) and the second sensor (104), wherein the method further comprises the following steps: acquiring (1200) sensor data of the third sensor (106, 108), the third sensor (102) being assigned to a third environmental sensing field, wherein the third environmental sensing field overlaps at least partially with the first and / or second environmental sensing field (110, 120), and combining (1300) sensor data of the at least third sensor (106, 108) with sensor data of the first and / or second sensors (102, 106).
11. The method according to any preceding claim, wherein the first environmental sensing field (110) and the second environmental sensing field (120) are arranged on an internal area and / or external area, wherein the door (10) separates the internal area from the external area, when being closed.
12. The method according to any preceding claim, wherein a first set of first and second sensors is assigned to an internal area, and wherein a second set of first and second sensors is assigned to an external area, wherein the door (10) separates the internal area from the external area, when being closed, and wherein the method according to any preceding claim is performed for the first set and / or the second set, whereinsensor data of the first set may be combined with sensor data of the second set, thereby providing a more complete understanding of the environmental sensing fields (110a, 110b, 120a, 120b) being monitored.
13. A control unit (200) for an automatic door system (1), wherein the control unit (200) is configured to carry out a method according to any of the preceding claims, wherein the control unit may be a distributed control unit.
14. The control unit (200) according to claim 13, wherein the control unit is configured to control the operation of multiple doors.
15. The control unit (200) according to any one of claims 13 or 14, wherein the control unit is arranged remotely from the at least one door (10) to be operated.
16. An automatic door system (1) comprising at least one door (10), at least a first sensor (102) and at least a second sensor (104), and a control unit (200) configured to control the operation of the at least one door (10) in particular according to any one of claims 13 to 15, wherein the system is configured to carry out a method according to any of the claims 1 to 12.
17. A computer program, comprising instructions, which when carried out by a processor cause the processor to perform operations of the method according to any of the claims 1 to 12.
18. A computer program product, embodied on a computer readable storage medium and configured so as when executed on a processor to perform operations of the method according to any of the claims 1 to 12.