Device for controlling an obstacle, the device being in particular intended to co-operate with an access management system

Abstract

The present invention relates to a device (2) for controlling an obstacle in a traffic lane (1), the device comprising: • an obstacle (4) movable between a deployed position and a retracted position, wherein the obstacle (4) is configured to block or allow vehicle access to the traffic lane (1); • a camera (6) configured to capture an image comprising the traffic lane (1); • a control unit (8), wherein the captured image is transmitted to the control unit (8), and wherein the control unit (8) is configured to implement an object-detection algorithm (5). According to the invention, the control unit (8) is configured to control the movement of the obstacle (4) when an object (5) of a predefined type is detected by the control unit (8) in a monitoring zone (3).

Description

OBSTACLE CONTROL DEVICE, IN PARTICULAR INTENDED TO COOPERATE WITH AN ACCESS MANAGEMENT SYSTEM

[0001] The present invention relates to the technical field of deployable obstacles allowing or blocking access to a traffic lane. Such obstacles may take the form of tilting or fan-shaped barriers, retractable bollards, in particular.

[0002] More specifically, the invention relates to the technical field of devices for managing access to such obstacles. STATE OF THE ART

[0003] To manage access to a traffic lane, an obstacle control device is known from the state of the art, which allows or blocks access to a traffic lane, in which the movement of the obstacle is controlled by protective equipment. By "obstacle control device", we mean a device capable of deploying or retracting an obstacle with respect to a traffic lane, for example, but not limited to, opening or closing a barrier, in particular according to instructions from an access management system. By "traffic lane", we mean a lane on which motor vehicles travel and / or a lane reserved for other means of mobility (motorcycle, bicycle, etc.), a pedestrian lane, or an open space such as a parking lot, a square or a hall.Generally speaking, such a "traffic lane" constitutes a space in which objects, in particular vehicles or people, can circulate. The deployable obstacle can in particular be configured to allow or block the passage of motor vehicles.

[0004] It is known from the state of the art to use so-called electrosensitive or electrosensitive pressure equipment, commonly referred to as "EPES or "EPSP" by those skilled in the art, such as infrared cells, ultrasonic detectors, or pressure plates, most often located on or in the direct vicinity of the traffic lane to which access is managed by said device.

[0005] The use of such equipment makes it possible to interrupt the movement of the obstacle when an object is detected. However, such equipment does not make it possible to distinguish the nature of the detected object, which may for example be a pedestrian, a motor vehicle, a truck, or a soft mobility device such as a bicycle or a scooter, and to adapt the control of the obstacle accordingly, in particular to control a resumption of movement which takes place when the equipment no longer detects the object. Therefore, the equipment mentioned above cannot completely prevent the risk of collision between the object crossing the traffic lane and the obstacle. To reduce said risk, it is known to limit the speed of deployment of the obstacle. In addition, the use of such equipment requires a development of the roadway, leading to significant and costly civil engineering works.

[0006] The invention aims to remedy all or part of the drawbacks of the state of the art, by proposing an obstacle control device which in particular better prevents the risk of collision and whose installation is simplified. PRESENTATION OF THE INVENTION

[0007] More specifically, the subject of the invention is an obstacle control device in particular intended to cooperate with an access management system, the device comprising: an obstacle movable between a deployed position in which the obstacle blocks access to the traffic lane and a retracted position in which the obstacle authorizes access to the traffic lane, the movable obstacle being in particular configured to block or authorize access of motor vehicles to the traffic lane; a camera configured to capture in real time an image comprising at least a portion of the traffic lane; a control member, the image captured by the camera being transmitted to the control member, said control member being configured to implement an object detection algorithm on said image.

[0008] The obstacle control device is remarkable in that the control member is further configured to control the movement of the obstacle according to at least one predetermined instruction in the case where at least a part of an object of at least one predefined type is detected by the control member in a vigilance zone formed on the image captured by the camera.

[0009] The term "traffic lane" refers to a roadway on which several types of users can travel. The traffic lane may, for example, lead to a parking lot. In such a configuration, the obstacle control device is configured to cooperate with an access management system, for example of said parking lot.

[0010] Thanks to such a combination of characteristics, such an obstacle control device makes it possible to prevent, in complete autonomy and with certainty, a risk of collision between the obstacle and an object located in the vigilance zone without limiting the speed of passage from one position to another of the obstacle. In other words, this increases the safety of the obstacle control device. Furthermore, the use of a camera makes it possible to avoid costly civil engineering work upstream and / or downstream of the obstacle. Finally, such a device makes it possible in particular to adapt the instruction according to the type of object detected, so that access rules can be defined for each type of object. In other words, the obstacle control device according to the invention makes it possible to operate a differentiation of instruction according to the type of object.

[0011] According to the invention, the object is associated with one of the following types: pedestrian, motor vehicle, truck, motorcycle, soft mobility device such as bicycle, scooter. Preferably, the association of the object with a given type is carried out by an object detection algorithm whose learning is supervised. More preferably, said association is carried out by a deep neural network.

[0012] Advantageously, the camera is configured to capture an image of the obstacle in real time, the vigilance zone corresponding to a volume in which the obstacle moves between the deployed position and the retracted position. In such a configuration, the risk of collision is further limited, the vigilance zone including the obstacle according to all the positions of its deployment or retraction kinematics.

[0013] Alternatively, the camera is configured to capture in real time an image of the obstacle, the vigilance zone corresponding to a projection on the ground of a volume in which the obstacle moves between the deployed position and the retracted position. In such a configuration, an object extending above the ground also extends above the vigilance zone, and its detection by the control member does not necessarily result in a command of a movement of the obstacle.

[0014] Advantageously, in particular in the case where the object is associated with a pedestrian type, the instruction consists of an interruption of the movement of the obstacle between the deployed position and the retracted position or, conversely, between the retracted position and the deployed position, and the instruction consists of a resumption of the movement in the case where the object leaves the vigilance zone.

[0015] Advantageously, the vigilance zone comprises, at least, a first region and a second region, the control member being configured to control the movement of the obstacle according to a first predetermined instruction in the case where at least a part of an object of at least one predefined type is detected by the control member in the first region, the control member being further configured to control the movement of the obstacle according to a second predetermined instruction in the case where at least a part of an object of at least one predefined type is detected by the control member in the second region. Thus, the location and size of each region in the vigilance zone can be modified, for example to correspond to the topography of the access route or to a given type of obstacle, the retraction and / or deployment speed of which can vary.

[0016] Advantageously, the first region and the second region are located on either side of the obstacle in the image captured by the camera. In such a configuration, different instructions can be assigned, for example depending on a direction of travel when an object crosses the vigilance zone and the obstacle.

[0017] Advantageously, the control member is configured to control the movement of the obstacle according to a third predetermined instruction in the case where at least a part of a first object of a first predefined type is detected by the control member in the vigilance zone, the control member being further configured to control the movement of the obstacle according to a fourth predetermined instruction in the case where at least a part of a second object of a second predefined type is detected by the control member in the vigilance zone. The differentiation of instruction by type of object can naturally be carried out within the vigilance zone or within a given region of the vigilance zone if necessary.

[0018] Advantageously, the obstacle is chosen from: a barrier, a gate, a retractable bollard, a harrow, a roadblocker, a motorized door, a motorized gate, a pedestrian access control product, a turnstile.

[0019] According to another aspect of the invention, it relates to an obstacle control method implemented by an obstacle control device as described above, the device comprising an obstacle movable between a deployed position in which the obstacle blocks access to a traffic lane and a retracted position in which the obstacle allows access to the traffic lane, the method comprising the steps: capturing an image in real time by a camera, the image comprising at least part of the traffic lane; transmitting said image from the camera to a control member; determining at least one vigilance zone on the image transmitted to the control member; implementing an object detection algorithm on the image by the control member;the transmission of a predetermined instruction by the control member to the obstacle, to control a movement of the obstacle, in the case where at least part of an object of at least one predefined type is detected by the control member in said vigilance zone.; PRESENTATION OF FIGURES

[0020] The invention will be better understood on reading the following description, given solely by way of example, and referring to the appended drawings given as non-limiting examples, in which identical references are given to similar objects and in which:

[0021] This is a schematic top view representation of a traffic lane obstacle control device according to a first embodiment, comprising an obstacle in a deployed position in which the obstacle blocks access to the traffic lane;

[0022] Theis a schematic top view representation of the obstacle control device in a retracted position of the obstacle in which the obstacle allows access to the taxiway;

[0023] This is a schematic representation in top view of a device for controlling an obstacle to a traffic lane according to a second embodiment of the invention;

[0024] This is a schematic representation of the management device according to a first use case;

[0025] This is a schematic representation in top view of a device for controlling an obstacle to a traffic lane according to a third embodiment of the invention and according to a second use case;

[0026] This is a schematic representation of the obstacle control device according to a third use case;

[0027] It should be noted that the figures set out the invention in detail to enable the invention to be implemented; although not limiting, said figures serve in particular to better define the invention where appropriate. DETAILED DESCRIPTION OF THE INVENTION

[0028] The invention relates to an obstacle control device 2 for a traffic lane 1 as illustrated in the and in the. The expression "traffic lane" means a roadway on which several types of users can move. The traffic lane 1 may, for example, open onto a parking lot. In such a configuration, the obstacle control device (2) is configured to cooperate with an access management system, for example of said parking lot.

[0029] The management device 2 comprises in particular at least one obstacle 4. The obstacle 4 is movable between a deployed position, as illustrated in the, in which the obstacle 4 blocks access to the traffic lane 1 and a retracted position, as illustrated in the, in which the obstacle 4 allows access to the traffic lane 1. Here, the obstacle 4 is formed by a multi-leaf gate. Naturally, any obstacle blocking the traffic lane 1 can be envisaged, the obstacle 4 being able to alternatively be formed by, but not limited to: a barrier, a gate, a retractable bollard, a portcullis, a roadblocker, a motorized door, a motorized gate, a pedestrian access control product, a turnstile.

[0030] The access management device 2 further comprises a camera 6. The camera 6 is configured to capture in real time an image comprising at least a portion of the traffic lane 1. The camera 6 is preferably a wide-angle camera. Here, the camera 6 is fixed to a static portion of the obstacle 4, but can alternatively be fixed to an independent support, so as to maximize the camera 6's view of the traffic lane 1.

[0031] The access management device 2 further comprises a control member 8. The image from the camera 6 is transmitted in real time to the control member 8. The control member 8 is configured to implement an object detection algorithm on the image received from the camera 6. The object detection algorithm is preferably implemented by a deep neural network of the control member 8. The transmission of the image from the camera 6 to the control member 8 can be carried out by wireless transmission, so that the control member 8 is located at a distance from the traffic lane 1, for example on a dedicated server.

[0032] The control member 8 is configured to control the movement of the obstacle 4 according to at least one predetermined instruction. For example, the control member 8 can control the passage of the obstacle 4 from the retracted position to the deployed position, the passage of the obstacle 4 from the deployed position to the retracted position or the interruption of the movement of the obstacle 4 in an intermediate position. Furthermore, the control member 8 can also modify the speed of the obstacle 4.

[0033] As illustrated in the and in the, the obstacle control device 2 is remarkable in that the control member 8 is further configured to control the movement of the obstacle 4 according to a predetermined instruction in the case where at least a part of an object 5 of at least one predefined type is detected by the control member 8 in a vigilance zone 3, formed on the image captured by the camera 6 by the control member 8.

[0034] Thanks to such a combination of characteristics, such an obstacle control device 2 makes it possible to prevent, completely autonomously and with certainty, a risk of collision between the obstacle 4 and an object 5 located in the vigilance zone 3, without limiting the speed of passage from one position to another of the obstacle 4.

[0035] Furthermore, the use of a camera 6 makes it possible to avoid costly civil engineering work upstream and / or downstream of the obstacle 4. Finally, such an obstacle control device 2 makes it possible to adapt the instruction according to the type of object 5 detected, so that access rules can be defined for each type of object 5.

[0036] Advantageously, the object can be associated with one of the following types: pedestrian, motor vehicle, truck, soft mobility device such as bicycle, scooter. Preferably, the association of the object 5 with a given type is carried out by an object detection algorithm whose learning is supervised. More preferably, said association is carried out by a deep neural network.

[0037] In the embodiment illustrated in and in, the camera 6 is configured to capture an image of the obstacle 4 in real time. The vigilance zone 3 corresponds here to a projection on the ground of a volume in which the obstacle 4 moves between the deployed position and the retracted position. Preferably, the volume is a parallelepiped prism, the dimensions of which are greater than or equal to the maximum space occupied by the obstacle 4 in each position of its kinematics. For this, the obstacle 4 is included in the viewing angle of the camera 6. The vigilance zone 3. In such a configuration, the risk of collision is even further limited, the vigilance zone 3 including the obstacle 4 according to all the positions of its deployment or retraction kinematics.

[0038] In a first use case, the object 5 is for example associated with the pedestrian type by the control member 8. In such a configuration, the instruction may consist of an interruption of the movement of the obstacle 4 between the deployed position and the retracted position. In the case where the object 5 leaves the vigilance zone 3, a new instruction then consists of a resumption of the movement of the obstacle 4.

[0039] In another use case, for example for an object 5 associated with the vehicle type by the control member 8, a different instruction can be sent, for example so as not to interrupt the movement of the obstacle 4, or to maintain the deployed position for an object 5 of the car type. Such an obstacle control device 2 makes it possible to adapt the instruction according to the type of the detected object 5, so that access rules can be defined for each type of object 5 and to operate a differentiation of instruction by type of object 5.

[0040] The obstacle control device according to the invention can thus make it possible to apply differentiated instructions, depending on the type of object 5 detected in the vigilance zone 3. In other words, by way of illustration, in the case of a roadblocker, a vigilance zone 3 can be defined around the roadblocker. When an object 5 is detected in said vigilance zone 3, an instruction can consist of an inhibition or an interruption of the deployment of the roadblocker if the detected object 5 is of the “pedestrian” type and a different instruction can consist of an inhibition or an interruption of the retraction of the roadblocker in the case where the detected object 5 is of the “vehicle” type.

[0041] Laet illustrate an obstacle control device 2, respectively according to a second use case and a third use case. Here, the vigilance zone 3 comprises in particular a first region 7 and a second region 9. The control member 8 is then configured to control the movement of the obstacle 4 according to a first predetermined instruction in the case where at least a part of an object 5 of at least one predefined type is detected by the control member 8 in the first region 7, the control member 8 being further configured to control the movement of the obstacle 4 according to a second predetermined instruction in the case where at least a part of an object 5 of at least one predefined type is detected by the control member 8 in the second region 9.

[0042] The location and size of each region 7,9 in the vigilance zone 3 may be modified, for example to correspond to the topography of the access route 1 or to a given type of obstacle 4, the re-entry and / or deployment speed of which may vary and which requires a larger surface region. Naturally, the vigilance zone 3 may comprise more regions depending on the necessary configuration.

[0043] Here, the first region 7 and the second region 9 are located on either side of the obstacle 4. The first instruction and the second instruction can then be adapted, for example to the direction of travel of the object 5 detected in one of the two regions 7, 9, or to detect two objects 5 in each region 7, 9. Furthermore, a differentiation within each region 7, 9 can be carried out, for example according to a given type of object 5. For example, the first instruction of the first region 7 can depend on an object 5 associated with the pedestrian type, and the second instruction of the second region 9 can depend on an object 5 associated with the vehicle type.

[0044] It will also be noted that the invention is not limited to the embodiments described above. It will indeed appear to those skilled in the art that various modifications can be made to the embodiment described above, in light of the teaching which has just been disclosed to them. In particular, the use cases described above are compatible with each other. In particular, the obstacle control device according to the invention can, according to one embodiment, combine differentiation according to distinct regions of the vigilance zone 3 and according to the type of object detected in these regions.

[0045] In the detailed presentation of the invention given above, the terms used should not be interpreted as limiting the invention to the embodiments set forth in this description, but should be interpreted to include all equivalents the prediction of which is within the reach of those skilled in the art by applying their general knowledge to the implementation of the teaching just disclosed to them.

Claims

Obstacle control device (2), in particular intended to cooperate with an access management system, the device comprising: an obstacle (4) movable between a deployed position in which the obstacle (4) blocks access to a traffic lane (1) and a retracted position in which the obstacle (4) allows access to the traffic lane (1), the movable obstacle (4) being in particular configured to block or allow access of motor vehicles to the traffic lane (1); a camera (6) configured to capture in real time an image comprising at least a portion of the traffic lane (1); a control member (8), the image captured by the camera (6) being transmitted to the control member (8), said control member (8) being configured to implement an object detection algorithm (5) on said image;the obstacle control device (2) being characterized in that the control member (8) is further configured to control the movement of the obstacle (4) according to at least one predetermined instruction in the case where at least a part of an object (5) of at least one predefined type is detected by the control member (8) in a vigilance zone (3) formed on the image captured by the camera (6), the object (5) being associated with one of the following types: pedestrian, motor vehicle, truck, motorcycle, soft mobility device such as bicycle, scooter; the predetermined instruction being adapted according to the type of object (5) detected.; Obstacle control device (2) according to claim 1, wherein the camera (6) is configured to capture in real time an image of the obstacle (4), the vigilance zone (3) corresponding to a volume in which the obstacle (4) moves between the deployed position and the retracted position. Obstacle control device (2) according to claim 1, wherein the camera (6) is configured to capture in real time an image of the obstacle (4), the vigilance zone (3) corresponding to a projection on the ground of a volume in which the obstacle (4) moves between the deployed position and the retracted position. Obstacle control device (2) according to any one of the preceding claims, in which, in particular in the case where the object (5) is associated with a pedestrian type, the instruction consists of an interruption of the movement of the obstacle (4) between the deployed position and the retracted position or, conversely, between the retracted position and the deployed position, and the instruction consists of a resumption of the movement in the case where the object (5) leaves the vigilance zone (3). Obstacle control device (2) according to any one of the preceding claims, wherein the vigilance zone (3) comprises, at least, a first region (7) and a second region (9), the control member (8) being configured to control the movement of the obstacle (4) according to a first predetermined instruction in the case where at least a part of an object (5) of at least one predefined type is detected by the control member (8) in the first region (7), the control member (8) being further configured to control the movement of the obstacle (4) according to a second predetermined instruction in the case where at least a part of an object (5) of at least one predefined type is detected by the control member (8) in the second region (9). Obstacle control device (2) according to claim 5, wherein the first region (7) and the second region (9) are located on either side of the obstacle (4) in the image captured by the camera (6). Obstacle control device (2) according to any one of the preceding claims, wherein the control member (8) is configured to control the movement of the obstacle according to a third predetermined instruction in the case where at least a part of a first object (5) of a first predefined type is detected by the control member (8) in the vigilance zone (3), the control member (8) being further configured to control the movement of the obstacle according to a fourth predetermined instruction in the case where at least a part of a second object (5) of a second predefined type is detected by the control member (8) in the vigilance zone (3). Obstacle control device (2) according to any one of the preceding claims, wherein the obstacle (4) is chosen from: a barrier, a gate, a retractable bollard, a harrow, a roadblocker, a motorized door, a motorized gate, a pedestrian access control product, a turnstile. An obstacle control method implemented by an obstacle control device (2) to a traffic lane (1) according to any one of the preceding claims, the device (2) comprising an obstacle (4) movable between a deployed position in which the obstacle (4) blocks access to the traffic lane (1) and a retracted position in which the obstacle (4) allows access to the traffic lane (1), the method comprising the steps: capturing in real time an image by a camera (6), the image comprising at least a portion of the traffic lane (1); transmitting said image from the camera (6) to a control member (8); determining at least one vigilance zone (3) on the image transmitted to the control member (8); implementing an object detection algorithm (5) on the image by the control member (8);the transmission of a predetermined instruction by the control member (8) to the obstacle (4), to control a movement of the obstacle (4), in the case where at least a part of an object (5) of at least one predefined type is detected by the control member (8) in said vigilance zone (3).;

Images