• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method of detecting a free area in a car park (20) using at least one environment sensor (30) fixedly installed in the car park (20). The method comprises the steps of: - entering measurement data of at least one segment (32) of the parking (20) using at least one environment sensor (30), comparing the measurement data reference measurement data for recognizing any modification of the segment (32) of the car park (20), and - detecting the segment (32) as a free or occupied zone according to the detection of a modification.
    公开号:FR3069365A1
    申请号:FR1856571
    申请日:2018-07-17
    公开日:2019-01-25
    发明作者:Andreas Lehn；Stefan Nordbruch
    申请人:Robert Bosch GmbH；
    IPC主号:
    专利说明:

    Field of the invention
    The present invention relates to a method for detecting a free area in a parking lot, for example, of a garage building, in particular a free area in a traffic corridor of a parking lot.
    The invention also relates to a system for detecting a free area in a parking lot, for example, in a garage building, in particular the search for a free area in a traffic corridor of a parking lot.
    Finally, the invention relates to a parking lot and to a computer program for implementing the method.
    State of the art
    Document DE 10 2015 201 209 A1 describes an automatic valet parking system for automatically storing a vehicle from a transfer or drop-off area to an allocated location in a predefined parking space. The known system consists of a parking lot monitoring system with at least one sensor unit installed in a fixed position. The parking lot monitoring system tracks vehicles in the predetermined parking space.
    In addition, according to the state of the art, driving assistance systems and highly automated driving systems (autonomous driving) are known. Such systems use, for example, one or more sensors which measure a three-dimensional space and / or analyze this space. The goal is to detect any object or obstacle. In addition, the dimensions (height, width, depth) of the object must be entered with great security or precision. Very precise precision is required to determine, for example, trajectories for passing a vehicle next to an object. The basis or condition for performing the calculations is to have many measurement points on objects at different heights. Another limiting condition is that the systems do not know the environment because the circulation can be done in the whole world upstream. As a result, driver assistance systems, in general, require three-dimensional imagery to make decisions. This is especially necessary if the systems operate in a highly automatic or even autonomous manner.
    Purpose of the invention
    The present invention aims to develop a concept of effective detection of a free area in a parking lot, for example, of a garage building, in particular, a free area in a traffic corridor of a parking lot.
    Presentation and advantages of the invention
    To this end, the subject of the invention is a method of detecting a free area in a car park by using at least one environment sensor fixedly installed in the car park, this method consisting in entering measurement data of at least one segment of the parking lot using at least one environment sensor, compare the measurement data with reference measurement data to recognize a modification of the parking segment and detect the segment as a free or occupied area in function of the detection of a modification.
    The environmental sensor (s) used are, for example, lidar sensors, radar sensors, ultrasonic sensors or other variants of journey time ("flight time") sensors. Environmental sensors in the form of journey time sensors have, for example, one or more measurement planes which are practically parallel to the circulation surface of the parking lot. Alternatively or in addition, image-generating sensors such as cameras (stereoscopic cameras) are used. As a variant or in addition, it is possible, for example, also to use pressure sensors, light barriers and / or magnetic sensors as environmental sensors.
    The environmental sensors are preferably installed on an element of the parking infrastructure such as a wall, a column or the roof, according to a fixed installation.
    The form in which the measurement data are presented generally depends on the type of sensor used. Thus, for example, in the case of journey time sensors, the measurement data are in the form of journey time of signals depending on the angle of measurement or else of distance values deduced from the signal travel time. In the case of an environmental sensor in the form of an image-generating sensor, the measurement data is, for example, in the form of brightness values and / or pixel colors.
    The reference measurement data is preferably available in the form of the measurement data for easy comparison.
    According to a preferred embodiment of the method of the invention, at least two sensors are used in the environment, a segment of the car park is entered with at least two of the environment sensors, which are produced differently one of the other and / or implement different measurement principles from each other.
    In other words, a segment of at least two environmental sensors of different realization and / or operating according to different measurement principles is monitored. As a variant or in addition, at least two environmental sensors capture a segment from different positions, that is to say that they have a different viewing angle, which has the advantage of more monitoring. reliable and more efficient.
    A segment corresponds to a measurement range of an environmental sensor. Alternatively, the measurement range of an environmental sensor can consist of several segments. A segment corresponds, for example, to a certain surface, such as, for example, a surface of 2 x 2 meters.
    A car park within the meaning of the invention is in particular a car park receiving motor vehicles. The parking lot is, for example, a building of garages or a parking area. The segment or area to be entered is, for example, in a traffic corridor in the car park.
    According to a particularly preferred development of the invention, at least two environmental sensors are used and with each of the environmental sensors, the measurement data of the parking segment are entered using the environment sensor, and a comparison is made. the respective measurement data with reference measurement data, associated with each environment sensor and for each environment sensor, any modification of the measurement data with respect to the reference measurement data is detected, and the segment is detected as free zone depending on whether a certain number of environmental sensors does not detect any change. For example, the segment can be detected as a free zone if at least one environmental sensor does not detect any modification or variation.
    This results in the technical advantage of avoiding a single surface detection indicating that an area is free, resulting in a sensor malfunction, upon detection of a non-free area.
    As a variant, the segment can be detected as a free zone if for several environmental sensors capturing a determined part of this segment, in particular more than 50% of the environmental sensors used, no change is detected. As a variant, the segment can be detected as a free zone if for all the environmental sensors used, no variation is detected. This results in the technical advantage of guaranteeing high security and at the same time high availability.
    According to a preferred embodiment variant of the invention, at least two environmental sensors are used and
    - each of the environmental sensors captures the measurement data of the parking segment, and
    - the respective measurement data are compared with the reference measurement data associated with each environment sensor and for each environment sensor the changes are detected, and as a function of the comparison, a value characterizing each environment sensor is associated the modification, in particular a value between 0 and 1, as a function of the characteristic value, a probability is calculated each time. The segment is detected as a free area according to the calculated probabilities. For example, to a first environment sensor which implements a first measurement principle, for the current entry of a segment, the value 0.8 is associated which corresponds to a probability of 80% in that the segment entered is effectively a free zone. With a second environment sensor which captures the same segment as the first environment sensor and which implements, for example, a second measurement principle, a value of 0.5 is associated for this input of the segment which corresponds to a 50% probability that the segment entered is effectively a free zone. The difference in results may, for example, result from different measurement principles. A common probability will only be calculated by multiplying the values. If the result of the calculation exceeds, for example, a determined threshold, the segment is detected as a free area. This principle can be extended to any number of environmental sensors that capture the segment concerned.
    This results in the technical advantage of taking into account the different strengths and weaknesses of the different measurement principles applied by the environmental sensors used and nevertheless making it possible to obtain reliable information indicating whether an area is free or not.
    According to a preferred embodiment variant of the invention, at least two environmental sensors are used and each of the environmental sensors captures the measurement data of the parking segment, the measurement data of the environmental sensors are assembled in one common measurement data set (merge) and this common measurement data set is compared to an associated reference measurement data set to detect a variation in the parking segment. This is done, for example, using a so-called grid model or also called fusion from raw data: first we divide the segment into several sub-segments. For each sub-segment, the measurement data (raw data) are entered using at least two environmental sensors. For each of the sub-segments, we first collect the measurement data (raw data) of all the environmental sensors and compare them with reference data. From the comparison, for each of the sub-segments, we deduce a result (free or non-free state). The decision whether or not the entire segment is detected as free is made from the sum of all the sub-segments. It is then possible to weight the sub-segments differently.
    According to a preferred embodiment of the invention, the measurement data of at least one segment of the car park is entered several times, for example three times in succession, using at least one sensor for environment and comparison of measurement data with reference data to detect a change in the parking segment. The detection of the segment as a free or non-free zone is done according to the frequency of detection of a modification or variation. For example, a segment is detected as a free zone if, for at least one application of the steps of the method, the segment has been detected as being free, that is to say if no modification or variation has been detected. As a variant, the segment will only be detected as free if, for each application, the segment has been detected as being free, that is to say if, for each application, no modification has been detected. Alternatively, the segment can be detected as being free if for at least one determined part or a determined number of executions the segment has been detected as being free, that is to say for a determined part or a determined number of applications no changes were detected. This makes it possible to effectively check the result indicating whether a determined segment has been detected as being free or not.
    A free area is in particular a parking area in which there is no object.
    Preferably, an object is detected if a modification or variation is recognized. Thus, for example, by using the measurement data entered, it is checked whether there has been a modification of the measurement data compared to the reference measurement data by the presence of an object in the area entered.
    An object within the meaning of the invention is an object which is temporarily in the parking lot, in particular in a traffic corridor of a vehicle and constitutes at least a potential risk of collision for the vehicle traveling in the parking lot. An object is, for example, another vehicle, a person, an animal or a concrete object. An object is notably present only temporarily in the parking lot. Within the meaning of the invention, fixed structures (element of infrastructure) of the car park such as, for example, walls, columns, partitions are not considered as objects.
    The object is, for example, on the parking lot floor, for example, on a traffic lane or in a traffic area, that is, for example, in a traffic corridor of the parking lot.
    Preferably, the recognition of a modification in a segment have performed an object recognition for this segment using at least one of the environment sensors. This object recognition preferably consists of entering properties characteristic of the recognized object, in particular of entering the size and / or the speed and / or the type of object.
    Preferably, for each environment sensor, the reference measurement data are captured and stored under defined conditions (reference conditions) and in particular it is guaranteed that for each possible measurement point in at least one segment, there are available 'a reference measurement value. These defined conditions preferably correspond to the situation according to which all the segments of the car park entered by the environmental sensor concerned are free zones. In particular, for environmental sensors measuring distances, one may encounter the difficulty that, for example, due to the limited range of the environmental sensor, for certain measurement points, no measurement value could be measured. reference because there is no static measurement object such as, for example, a parking wall within reach of the environment sensor. In this case, infrastructure elements such as reference objects or reflectors can be installed in the car park which guarantee that for each possible measurement point there is a reference measurement value. The reflectors are used in particular to reflect the signals emitted by an environment sensor (for example laser rays, ultrasound signals, radar waves) with high efficiency back to the environment sensor.
    Preferably, the measurement data entered are distance values. For this, the environment sensor that performs the capture is an environment sensor measuring a distance, for example, a radar sensor, an ultrasonic sensor, a lidar sensor or a scanning laser. The reference measurement data are, for example, the measurement data which were captured by the environment sensor at an earlier time under defined conditions, for example at a time when with certainty there was no object in the monitored segment. The reference measurement data can also include average values for the measurement data entered under defined conditions.
    Preferably, the reference measurement data are selected according to the current situation, in particular the time of the year and / or the time of day and / or the weather conditions and / or the light conditions. For this, reference measurement values are available corresponding to particular situations, for example, recorded in a database.
    Comparing the measurement data with the reference measurement data notably consists in forming a difference.
    In particular, a modification is detected or recognized in particular for a predefined minimum number of measurement points when the comparison of the measurement data and the reference measurement data gives a difference greater than a predefined tolerance value.
    Formulas such as modification or deviation or difference in the sense of the description relate in particular to the case where the measured values, for example, the distance values or the re-emission values differ from a respective reference value or from a respective reference measurement value or a range of reference values according to a predefined tolerance value. These are only differences or modifications greater than the predefined tolerance value which result in the entry of a variation or modification and thus in the detection of an object. This means in particular that small differences for the measured distance values and / or for other measured variables such as brightness or color are not significant to say that the measurement data are equal or identical as long as the differences are less than the preset tolerance value.
    This means in particular that a free zone is detected, for example, only if, for example, the measured distance values differ from the foreseeable distance values by a difference less than a predefined tolerance value. This results in the technical advantage that the environmental sensor can detect a free area in an efficient and reliable way.
    In the case of at least one of the environment sensors produced in the form of a distance measurement sensor, the detection according to the present invention of a free zone and / or the grasping of an object is done in particular by comparing at least one current measured distance value and a reference measurement value, in particular a predictable distance value. A variation will be entered in particular for a predetermined minimum number of measurement points with measurement each time of a distance value which differs from a predictable distance value; in particular the measured distance value is less than the predictable distance value, by a difference less than a predefined tolerance value. Thus, the predictable distance values correspond, for example, to known limits of the parking lot such as walls or columns located within the measurement range of the environmental sensor.
    According to a preferred embodiment of the invention, the measured distance values are used and they are associated with one or more objects, and / or scenarios constructed from several objects and classified according to the operation. Thus, for example, by the use of successive measurements, the direction of movement and / or the speed of the detected object is determined and it is thus evaluated, for example, when the object has again left the segment. from the parking lot and freed up this segment. As a variant or in addition, a classification can be made of the type of object or of the scene. Speed information is thus determined to further verify the classification.
    In the case of at least one environmental sensor in the form of an image-generating sensor such as a video camera, one or more reference images may be used as reference measurement data; a reference image is, for example, that which was taken at an instant earlier with the same image-generating sensor under defined conditions. The reference image can be saved in a memory. We can also provide that, for example, depending on the time of the year and / or the time of day, we save different reference images and select for comparison, the reference image adapted to the current time of year and / or time of day. This notably results in the technical advantage of taking into account, by the adapted reference image, the time of day or the time of year.
    The comparison can be made according to an alternative embodiment of the invention, with a previous image of the same succession of images, for example a comparison with the image taken directly before. We can also consider determining a series of several successive images and using the average image as a reference image.
    According to a possible embodiment of the invention, as reference measurement data, instead of a single reference image, it is also possible to use a succession of reference images or a reference video sequence. According to this embodiment, a single current camera image and / or a current video sequence is compared with the succession of reference images. If the current image or succession of images and the reference image and / or the succession of reference images do not show any modification, i.e. if these images are analogous or identical or have differences which do not exceed a predefined maximum tolerance value, it is assumed that there is no object on a segment of the car park floor and that the segment is free. But, since an object is on the ground segment of the parking lot, the current image will be different from the reference image by a difference greater than the preset tolerance value. If there is a modification of the camera image compared to a reference image or a series of reference images, it will be assumed that there is one or more objects in the image, for example using a known method of digital image processing. The object can be classified more precisely, for example, based on its dimensions and / or its geometric shape. By comparing several camera images taken successively, it will also be possible to determine the speed and / or the direction of movement of the object or objects. From the information collected, we can classify the object (s) more precisely and / or associate a scenario.
    Depending on the classification of the object (s) or the association of the objects with a scenario or a specific obstacle (for example a passenger vehicle, a two-wheeled vehicle, a person, etc.), a recommendation can be deduced. maneuver for vehicles moving in the parking lot. These recommendations are, for example, transmitted to vehicles.
    This results in the technical advantage of being able to assess the objects as to their dangerousness or their residence time in the concerned segment of the car park.
    According to another development, the subject of the invention is a system designed to detect a free location in a car park, comprising at least one environment sensor installed in a fixed manner in the car park, this sensor being designed to enter measurement data of '' at least one segment of the parking lot and a processor for comparing the measurement data taken with reference measurement data to detect a variation in the parking segment and to detect the segment as a free or occupied area, as a function of the recognition of 'a variation.
    The technical functionalities of the system are analogous to the technical functionalities of the process and vice versa. This means in particular that the characteristics of the system are parallel to the characteristics of the process and vice versa.
    At least one of the environment sensors is especially designed to enter distance values of objects in the area, especially in the scanning plane of the environment sensor. In particular, the environment sensor is a travel time sensor, for example, an ultrasonic sensor or a lidar sensor or a radar sensor or a scanning laser.
    As a variant or in addition, at least one environment sensor is an image generator sensor, in particular a camera.
    The reference values are preferably stored in a memory provided for this purpose.
    The processor preferably groups together the measurement data from several different environmental sensors, in particular of different structure, into a single measurement data set and this set is compared in common with a corresponding reference measurement data set.
    According to one development, the processor is designed to optimally detect an object if at least one current distance value measured using the environment sensor is less than a predicted distance value, in particular by at least one value of predefined tolerance. Predictable distance values can be saved in a memory provided for this purpose.
    According to one development, the processor is optimally designed to detect one or more objects if the current camera image differs by at least a predefined tolerance value compared to a reference image. The processor applies known digital image processing methods for object recognition.
    According to a third development, the invention relates to a car park equipped with a system for detecting a free area inside the car park as described above.
    The development according to the invention advantageously applies to AVP systems (that is to say automatic valet systems, abbreviated as AVP systems). In the case of such an AVP system, it is in particular provided that the motor vehicles automatically circulate in the parking lot to be put away and at the end of the parking time, they are driven automatically from their parking position to the recovery position where the vehicle is picked up by its driver. Advantageously, such systems can use the invention with a centralized infrastructure (for example a server) which provides the moving vehicle in the parking lot with instant information concerning the free zones and / or the objects detected, in particular if an object is found. in a traffic lane provided for the respective vehicle. The affected vehicle can then be stopped, for example until the object is again disappeared. As a variant or in addition, an alternative route and / or an avoidance maneuver can also be developed for the vehicle which no longer passes through the section of the parking lot path where the object has been detected, that is to say say the area that was not detected as free. The development according to the invention has in particular, vis-à-vis the usual driving assistants and usual driver assistance systems, used for object recognition, the advantage that the environmental sensors used are installed fixed way in the parking lot; they are, for example, installed on walls or columns. Thus, the field of vision, that is to say the environment looked at by each sensor, is known, in particular in the case where there is no object. It is then sufficient to determine whether the area into which the vehicle is to enter is free, that is to say if there is no object. As a vehicle in AVP mode does not necessarily have to perform an exhaust maneuver in the traffic zone usually reserved for the AVP system, and in many cases it is sufficient to stop the vehicle, it does not is not necessary to determine the dimensions, speed and other parameter of the detected object. It is enough to detect the presence of an object because for example in a parking lot, usually, the speed of an AVP vehicle is very low (for example less than 10 km / h) compared to the road traffic which can, for example , exceed 200 km / h.
    Preferably, the parking lot according to the invention includes a management system for the automatic valet system (AVP system). The parking management system preferably consists of carrying out an action in the event that a segment on the planned trajectory of a moving vehicle in the parking lot has an area detected as not being free.
    The action to be performed consists, for example, in: the issuance of a stop order by the parking management system towards the vehicle and / or input of a speed adaptation signal in particular of reduction of speed of a vehicle and / or the calculation of an alternative route as a function of the free regions and of the information transmitters for driving on the alternative route, towards the vehicle and / or the calculation of an avoidance maneuver and the transmission of information to perform the avoidance maneuver with the vehicle.
    This results in the technical advantage of reacting effectively to the detection that a specific segment of the parking lot is not free, a segment that a vehicle is about to take, which avoids endangering the vehicle.
    According to another development, the invention relates to a computer program with a program code for implementing the method of detecting a free area in a parking lot, when the program is executed by a computer.
    The invention is based on the consideration that the problem posed above can be resolved with at least one environment sensor installed in the parking lot and which captures a segment of the parking lot, this segment being an area of parking space which is part of the measurement area of the environment sensor. The segment is, for example, a ground segment of the car park which has, for example, a surface of 2 x 2 m 2 . If the environmental sensor captures measurement data at the current time that are not different from the reference measurement data, i.e. if no change or deviation has been found , then the segment will be classified as a free zone. A free area within the meaning of the invention is a parking area in which a vehicle can circulate without risk under normal conditions. The invention is thus based on the idea that a segment of a parking lot can be detected as being free, that is to say free from any obstacle if the capture of the measurement data of the segment using the sensor of the environment does not give any modification compared to reference data. A parking segment is detected as free if no modification has been entered in relation to the reference measurement data. Thus, according to the invention, objects are not captured in a targeted manner using the environment sensor, but first of all, the changes are detected with respect to the reference measurement values so that everything first, the cause of the change is immaterial. In the following optional steps, the cause of the detected change can be determined more precisely. The object of the invention is therefore to provide high security for vehicles traveling in the car park while ensuring very high availability.
    drawings
    The present invention will be described below, in detail with the aid of exemplary embodiments shown in the accompanying drawings in which:
    FIG. 1 shows a flowchart of a method for detecting a free area in a car park, FIG. 2 schematically shows part of a car park with an environment sensor and infrastructure elements for supplying measurement data reference, Figure 3 is a schematic view of a parking lot with a set of environmental sensors for detecting free areas in the parking lot, Figure 4 schematically shows a system for detecting a free area in a parking lot according to a realization of the invention.
    Description of embodiments of the invention
    Figure 1 shows a method of detecting a segment in a parking lot, to detect a free area using at least one environmental sensor fixedly installed in the parking lot, for example, a lidar sensor.
    The method comprises the following steps: entering 101 the parking segment using the environment sensor by entering measurement values, in particular distance values, comparing 103 measurement values, in particular the measured distance values to reference measurement values, in particular predictable distance values, and detect 105 a free zone if there is no difference between the measurement values and the reference measurement values, in particular if the distance values measurements have been accepted as predictable distance values, and optionally detect an object 106 if a deviation has been detected, and optionally classify the object 107 and / or associate one or more objects with an object or a scenario.
    Information indicating that a free zone or a change has been detected can be transmitted to a forward control system. This regulation system can, in the case where an object has been detected, for example, a remote-controlled vehicle, stop it or send a stop signal to a motor vehicle traveling in autonomous mode if an object has been detected on the parking segment located in the vehicle traffic lane in the parking lot. This ensures that this vehicle will stop in time ahead of the object. The regulation system is, for example, part of the parking management system.
    FIG. 2 shows an environment sensor 30 fixedly installed in the parking lot 20, for example, on a first wall 60 of the parking lot 20; the environment sensor of this exemplary embodiment is a lidar sensor.
    The environment sensor 30 captures or measures a segment 32 of the parking space which corresponds to the measurement range 33 of the environment sensor 30. The segment 32 comprises several sub-segments 32 ′, 32 ”, 32 ″ .
    There is a second wall 61 as an infrastructure element inside the segment 32. This wall 61 or the distance from this wall is captured by the environment sensor 30. This is done, for example, by a laser beam 38 emitted by the environment sensor 30 and which is reflected by the wall 61 in return towards the environment sensor 30. The environment sensor 30 receives, that is to say measures the corresponding reflected signal .
    If a set of measurement points 37 is associated with the wall 61, these points reflect the return travel time measurement signals, towards the environment sensor 30, if no object is between the wall 61 and the sensor environment 30. Figure 2 shows only a selection of measurement points given by way of example. The distances and / or travel times associated with the measurement points 37 can serve as reference measurement data for the method of the invention.
    In the 32 ”sub-segments, the wall 61 is no longer in the measurement range 33 of the environmental sensor 30. To nevertheless have reference measurement data even for the segments
    32 ”, an infrastructure element 80 is provided for the 32” segments so as to enter distances from measurement points 37 ”of this infrastructure element 80 and to have these elements as reference measurement data .
    In the 32 '”sub-segments, the wall 61 is also no longer in the measurement range 33 of the environmental sensor 30. To nevertheless have reference measurement data even for the 32'” segments, we have in front of the wall 61 a reflective infrastructure element 82 at the edge of the measuring range 33 so as to be within range of the environment sensor 30 and to enter the distances from the reflective infrastructure elements 82 and which correspond to the 37 '”measurement points to have them as reference measurement data.
    This ensures that reference measurement data is available for the entire measurement range 33 of the environmental sensor.
    The use of a system according to the invention and / or of a method according to the invention applied to an automatic valet system AVP, uses a set of fixed sensors (environment sensors) to monitor as much as possible the whole parking lot. Thus, FIG. 3 shows by way of example a detail of a parking lot 20 forming part of an AVP system. The car park has different parking spaces 23 for vehicles in AVP mode. The parking lot 20 has a drop-off area 18 where the vehicle 10 is made available to the parking manager 20. For this, the driver drives his vehicle 10 to the drop-off area 18; he leaves vehicle 10 and transmits it to the parking manager 20.
    The monitoring of the vehicle 10 during its movement in the parking lot 20 is done by a monitoring system associated with the parking lot 20 and which comprises a set of environmental sensors 30, for example, in the form of Lidar sensors and each having a zone of measures 33 as well as other environmental sensors 31 in the form of stereo video cameras. The environmental sensors 30, 31 are distributed, preferably in the parking lot 10 so that during its automatic circulation from the drop-off zone 18 to its parking location 23, the vehicle 10 is constantly moving over a segment 32 monitored by at least one environment sensor 30, 31. Each of the environment sensors 30 captures one or more segments 32 of the parking lot 20 and detects them as being free if there is no modification with respect to reference measurement data. Information indicating whether the segment is currently free or not is transmitted by the parking management system
    15.
    After transfer of the vehicle 10 to the parking manager 10, a free parking space as parking space 24 is allocated by the parking management system 15 among the free parking spaces 23 available. The occupied parking spaces are given the reference 22 in FIG. 3.
    After the allocation of the parking space, the vehicle 10 is driven to the allocated free parking space 24. The vehicle 10 is, for example, designed to execute a driving maneuver in autonomous mode using a parking assistance system. The parking management system 15 accesses the parking assistance system so that the vehicle 10 can travel in autonomous mode in the parking lot 20 and arrive by itself in the parking space 24. According to the invention , the appropriate information is provided by the management system 15 of the parking lot to the vehicle 10 so that the vehicle 10 can travel in autonomous mode along a path 40 established according to the information inside the parking lot 20. As a variant, the vehicle 10 can be remote-controlled along the path 40 to its parking position 24.
    According to the invention, using the environmental sensors 30 and 31, it is monitored whether the segments of the parking lot that the vehicle 10 will use for its journey along the trajectory 40 to the parking space 24 are free. If a segment through which the vehicle 10 must pass is not detected as being free by one or more of the environment sensors, the vehicle 10 is stopped and / or a bypass trajectory is calculated with the management system 15 of the parking to guide the vehicle 10 along this bypass trajectory.
    FIG. 4 schematically shows a system for detecting the free area in the parking lot according to an embodiment of the invention. The system comprises an environment sensor 30 as well as a processor 62 and a memory 64. The memory 64 contains the recording of the reference measurement data, in particular different sets of reference measurement data which correspond to different conditions such as time of year, time of day, weather, light, etc. The processor compares the measurement data entered by the environment sensor 30 with a selected set of reference measurement data, chosen according to current conditions. According to the comparison, a modification is detected between the segment entered from the parking lot and the reference measurement data by the processor 62; this results in the detection of the segment entered which is considered free or not free depending on whether a modification has been detected. This information is provided by an interface 70, for example, the parking management system 15. In the example presented, the processor 62 and the memory 64 are integrated into the control unit 66 of the environment sensor 30. The control unit 66 is provided as an integral component of the surface sensor 30. As a variant, the processor 62 and / or memory 64 can be part of the surface sensor 30 constituting a separate unit. For example, processor 62 and / or memory 64 are part of a parking management system 15. As a variant, processor 62 and / or memory 64 are part of a Cloud computing system. The information indicating that a segment of the parking lot 20 has been detected as free is transmitted by the interface 70, for example to a vehicle which is moving in the parking lot 20 and / or a parking management system 15. The information can be transmitted, for example, over a wireless link.
    NOMENCLATURE OF MAIN ELEMENTS
    999
    80, 82
    101-107
    Vehicle
    Parking management system
    Drop off area
    Car park
    Parking space
    Environmental sensor
    Environmental sensor
    Segment
    Subsegment
    Measuring range
    Measuring point
    Laser beam
    Path
    Wall
    Processor
    Memory
    Control unit
    Interface
    Infrastructure element
    Detection process steps
    权利要求:
    Claims (24)
    [1" id="c-fr-0001]
    1 °) Method for detecting a free zone in a parking lot (20) using at least one environment sensor (30) installed in a fixed manner in the parking lot (20), method comprising at least the steps consisting in:
    enter measurement data from at least one segment (32) of the parking lot (20) using at least one environment sensor (30), compare the measurement data with reference measurement data to recognize a modification of the segment (32) of the parking lot (20), and detecting the segment (32) as a free zone or occupied zone as a function of the detection of a modification.
    [2" id="c-fr-0002]
    2) Method according to claim 1, characterized in that at least two sensors are used in the environment (30, 31), a segment of the car park is captured with at least two of the environment sensors (30, 31) , which are implemented differently from each other and / or implements different measurement principles from each other.
    [3" id="c-fr-0003]
    3 °) Method according to claim 1 or 2, characterized in that at least two environmental sensors (30, 31) are used and for each of the environmental sensors (30, 31), the measurement data are entered of the parking segment using the environment sensor (30, 31), the respective measurement data are compared to the reference measurement data associated with each of the environment sensors (30, 31) and for each sensor d environment (30, 31) a variation is recognized, and the segment is detected as a free area if for a determined number of environment sensors, no variation has been detected.
    [4" id="c-fr-0004]
    4 °) Method according to claim 3, characterized in that the segment is detected as a free area if for at least one of the environmental sensors (30, 31) no variation has been recognized.
    [5" id="c-fr-0005]
    5 °) Method according to claim 3, characterized in that the segment (32) is detected as a free zone if, for more than a determined part of the environmental sensors (30, 31) used, in particular more than 50% of the environmental sensors used (30, 31), no variation was detected.
    [6" id="c-fr-0006]
    6 °) Method according to claim 3, characterized in that the segment is detected as a free area if for all the environmental sensors used (30, 31) no variation has been detected.
    [7" id="c-fr-0007]
    7 °) Method according to claim 1 or 2, characterized in that at least two environmental sensors are used (30, 31) and for each of the environmental sensors, the measurement data of the segment of the parking location using the environment sensor, the respective measurement data is compared to the reference measurement data associated with each environment sensor (30, 31) and a variation is recognized for each of the environment sensors (30, 31), and to each of the environmental sensors (30, 31), as a function of the comparison, a characteristic value of the variation is associated, in particular between 0 and 1, as a function of the characterized values, one calculates the probability and the segment is detected as a free area according to the probability of success calculated.
    [8" id="c-fr-0008]
    8 °) Method according to claim 1 or 2, characterized in that at least two environment sensors (30, 31) are used, and for each of the environment sensors (30, 31) the measurement data are entered of the parking segment (32) using the environment sensor (30, 31), the measurement data of the surrounding field sensors (30, 31) are grouped into a common set of measurement data and this is compared measurement data set common to an associated reference measurement data set to recognize any variation in the parking segment.
    [9" id="c-fr-0009]
    9 °) Method according to one of claims 1 to 8, characterized in that upon detection of a variation in a segment, an object recognition is provided for at least one segment using at least one of the respective surrounding field sensors (30, 31).
    [10" id="c-fr-0010]
    10 °) Method according to claim 9, characterized in that the object recognition (30, 31) comprises the entry of characteristic properties of the object, in particular the entry of dimensions and / or speed and / or type d 'object.
    [11" id="c-fr-0011]
    11 °) Method according to one of claims 1 to 10, characterized in that the reference measurement data are entered for each environment sensor (30, 31) under defined conditions and they are stored, * in particular ensures that for each measurement point entered using a surrounding field sensor (30, 31), there is an associated reference measurement value in at least one segment.
    [12" id="c-fr-0012]
    12 °) Method according to one of claims 1 to 11, characterized in that at least the steps of entering measurement data of at least one segment of the parking lot are carried out successively, using at least one environment sensor (30, 31) and the comparator of measurement data to reference measurement data, for detecting a variation of the parking segment, by executing it several times successively over time , the detection of a segment (32) being made for a free space or a parking space as a function of the frequency of the detection of a variation.
    [13" id="c-fr-0013]
    13 °) Method according to one of claims 1 to 12, characterized in that one chooses the reference measurement data according to the current situation and in particular the period of the year and / or the hour of the day and / or the environmental conditions and / or the light ratios.
    [14" id="c-fr-0014]
    14 °) Method according to one of claims 1 to 13, characterized in that the comparison of the measurement data to reference measurement data comprises the formation of a difference.
    [15" id="c-fr-0015]
    15 °) Method according to one of claims 1 to 14, characterized in that a variaLiun is recognized if in particular for a predefined minimum number of measurement points, the comparison of the measurement data with the reference measurement data gives a deviation greater than a predefined value.
    [16" id="c-fr-0016]
    16 °) System made to detect a free location in a parking lot, by implementing the method of any one of claims 1 to 15, comprising at least one environment sensor (30, 31) installed in a fixed manner the parking lot, this sensor being designed to capture measurement data from at least one segment of the parking lot, and a processor, * to compare the measurement data taken with reference measurement data, to detect a variation in the segment of the parking lot parking, and * to detect the segment as a free zone or as an occupied zone, depending on the recognition of a variation.
    [17" id="c-fr-0017]
    17 °) System according to claim 16, characterized in that at least one surrounding field sensor (30) is made to enter values of distance from the object in the area, the environment sensor is in particular an ultrasonic sensor or a Lidar sensor or a radar sensor or a laser scanning sensor.
    [18" id="c-fr-0018]
    18 °) System according to one of claims 16 or 17, characterized in that at least one environment sensor (31) is produced as an image generator sensor, in particular as a camera.
    [19" id="c-fr-0019]
    19 °) System according to one of claims 16 to 18, characterized in that the reference measurement values are stored in a memory unit (64).
    [20" id="c-fr-0020]
    20 °) System according to one of claims 16 to 19, characterized in that the processor can gather several measurement data, in particular two environment sensors (30, 31) of different realization, in a common measurement data set and compare the common measurement data set to a corresponding reference measurement data set.
    [21" id="c-fr-0021]
    21 °) Parking lot (20) comprising a system according to one of claims 16 to 20, for detecting a free area which is inside the parking lot (20).
    [22" id="c-fr-0022]
    22 °) Parking (20) according to claim 21, comprising:
    a management system (15) for managing an automatic valet parking system, the parking management system (15) being able to execute an action, if a segment (32) on a planned trajectory (40) of a vehicle (10 ) who is not moving in the parking lot has been detected as a non-free zone.
    [23" id="c-fr-0023]
    23 °) Parking according to claim 22, characterized in that the management system (15) can execute an action consisting of:
    the emission of a stop order sent by the parking management system (15) to the vehicle (10) and / or the emission of a speed adaptation signal in particular of speed reduction of a vehicle (10) and / or the calculation of an alternative route as a function of the free regions and of the information transmitters for traveling on the alternative route, to the vehicle (10) and / or the calculation of a maneuver d avoidance and the transmission of information to perform the avoidance maneuver with the vehicle (10).
    [24" id="c-fr-0024]
    24 °) computer program comprising a program code for the implementation of the method for detecting a free area in a parking lot (20) according to one of claims 1 to 15, when the computer program is executed by a computer.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3069365A1|2019-01-25|METHOD AND SYSTEM FOR DETECTING A FREE AREA IN A PARKING
    EP3447528B1|2021-11-17|Automated driving system that merges heterogenous sensor data
    US10817736B2|2020-10-27|System and methods for identifying unoccupied parking positions
    FR2976246B1|2019-07-05|DRIVING ASSISTANCE SYSTEM WITH DETECTION OF OBJECTS
    KR101891460B1|2018-08-24|Method and apparatus for detecting and assessing road reflections
    EP3491475A1|2019-06-05|System for steering an autonomous vehicle
    FR3001686A1|2014-08-08|METHOD FOR ESTABLISHING A CIRCULATION CORRIDOR OF A VEHICLE AND DRIVING ASSISTANCE SYSTEM
    KR101542564B1|2015-08-06|system for managing traffic based on zone classified architecture
    EP3519267B1|2020-07-08|Driving assistance on a highway with carriageways separated by a guardrail
    CN111047870A|2020-04-21|Traffic violation vehicle recognition system, server, and non-volatile storage medium storing vehicle control program
    JP6392735B2|2018-09-19|Information processing apparatus, information processing method, vehicle control apparatus, and vehicle control method
    FR3059137A1|2018-05-25|METHOD AND SYSTEM FOR DETECTING AN OBJECT IN RELIEF IN A PARKING
    CN108780608B|2021-08-31|Method and device for classifying the validity of a parking space detected by means of a distance-based detection method
    FR3056530B1|2019-07-12|OBSTRUCTION DETECTION BY FUSION OF OBJECTS FOR MOTOR VEHICLE
    FR3059135B1|2019-11-08|METHOD AND SYSTEM FOR DETECTING AN OBJECT IN RELIEF IN A PARKING
    EP3726494A1|2020-10-21|Method for assisting with the driving of vehicles, associated computer program and system
    WO2018041978A1|2018-03-08|Device for determining a speed limit, on-board system comprising such a device, and method for determining a speed limit
    EP3716097A1|2020-09-30|Method for acquiring data captured by a capture module embedded in a mobile device following a predetermined trajectory, corresponding computer program and device
    EP3008664B1|2019-10-23|Method and system for monitoringobjects in motion
    FR3094126A1|2020-09-25|Electronic device for automatic selection of a monitoring configuration of a road traffic area, selection method and associated computer program
    CN111665515A|2020-09-15|Multi-vertical layer light detection and ranging system, automated parking assistance and computer vision lane detection and maintenance
    FR3105961A1|2021-07-09|Method and device for determining a lane change indicator for a vehicle
    CN114084129A|2022-02-25|Fusion-based vehicle automatic driving control method and system
    CN112660128A|2021-04-16|Apparatus for determining lane change path of autonomous vehicle and method thereof
    WO2021170917A1|2021-09-02|Optimisation of the negotiation of a signalling system by an ego vehicle
    同族专利:
    公开号 | 公开日
    US20190027037A1|2019-01-24|
    CN109285381A|2019-01-29|
    US10388164B2|2019-08-20|
    DE102017212513A1|2019-01-24|
    FR3069365B1|2021-12-03|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US6107942A|1999-02-03|2000-08-22|Premier Management Partners, Inc.|Parking guidance and management system|
    WO2003029046A1|2001-10-03|2003-04-10|Maryann Winter|Apparatus and method for sensing the occupancy status of parking spaces in a parking lot|
    DE102007048848A1|2007-10-11|2009-04-16|Robert Bosch Gmbh|Spatial driver assistance system|
    CN103810898A|2012-11-15|2014-05-21|深圳市赛格导航科技股份有限公司|Automatic parking and picking system and method|
    US9666075B2|2013-11-18|2017-05-30|ImageMaker Development Inc.|Automated parking space management system with dynamically updatable display device|
    CN105096649A|2014-05-12|2015-11-25|广州市捷众智能科技有限公司|Multi-mode combined vehicle detection apparatus|
    DE102015201209A1|2015-01-26|2016-07-28|Robert Bosch Gmbh|Valet parking method and valet parking system|
    CN105223952B|2015-09-28|2019-03-29|小米科技有限责任公司|The control method and device of balance car|
    US9718404B2|2015-10-01|2017-08-01|Ford Global Technologies, LLCS|Parking obstruction locator and height estimator|
    CN105702083B|2016-04-13|2018-08-14|重庆邮电大学|Parking lot based on distributed vision-vehicle cooperative intelligent shutdown system and method|
    CN106448187A|2016-11-14|2017-02-22|苏州大学|Vehicle detection system and vehicle detection method based on fusion of magnetic sensor and ultrasonic sensor|DE102016223171A1|2016-11-23|2018-05-24|Robert Bosch Gmbh|Method and system for detecting a raised object located within a parking lot|
    DE102018212998A1|2018-08-03|2020-02-06|Robert Bosch Gmbh|Method for driving a vehicle in a parking environment|
    CN110211420B|2019-06-25|2021-11-30|重庆长安汽车股份有限公司|Method and system for judging whether current scene is parking lot or not and computer readable storage medium|
    法律状态:
    2019-07-24| PLFP| Fee payment|Year of fee payment: 2 |
    2019-08-16| PLSC| Publication of the preliminary search report|Effective date: 20190816 |
    2020-07-27| PLFP| Fee payment|Year of fee payment: 3 |
    2021-07-22| PLFP| Fee payment|Year of fee payment: 4 |
    优先权:
    申请号 | 申请日 | 专利标题
    DE102017212381|2017-07-19|
    DE102017212381.8|2017-07-19|
    DE102017212513.6A|DE102017212513A1|2017-07-19|2017-07-20|Method and system for detecting a free area within a parking lot|
    [返回顶部]