EP3227215B1 - Method and system for determining the position of an elevator car - Google Patents

Description

The invention relates to a method and a system for determining the position of an elevator car arranged in a lift cage arranged elevator car of an elevator system according to the preamble of the independent claims.

It is known from the prior art, for example from the EP 1 232 988 A1 To provide elevator systems with a camera, which is attached to the elevator car and is used to take pictures of the elevator shaft and derive information about a position of the elevator car therefrom. In this case, shaft components are set as markings, which are recorded by the camera and processed by a computer connected thereto.

The disadvantage here is that a learning journey is necessary in order to be able to assign the manhole components to an absolute position of the elevator car. In addition, an absolute positioning is associated with such a system with a high computational effort,

JP 2009 220 904 discloses an elevator installation according to the preamble of claim 8.
It is therefore an object of the invention to provide a method and a system of the type mentioned, which avoid the disadvantages of the known and in particular allow a reliable determination of the position of the elevator car. In addition, the inventive system should be inexpensive to produce and operable.

This object is achieved in a method and system according to the invention with the features of the independent claims.

The method according to the invention for determining the position of an elevator car of an elevator system which can be moved in an elevator shaft, wherein the elevator car is equipped with an acceleration sensor, comprises the following steps.

In a first step, the acceleration data from the acceleration sensor are detected by a computing unit. This is followed by a calculation by the arithmetic unit of the current position and / or speed of the elevator car starting from an initial position and the acquired acceleration data. The position or speed of the elevator car is thus determined in accordance with an inertial navigation system. However, it will be appreciated that, due to the characteristics of such a system, delays and errors may occur which affect the reliability of position determination. Thus, for example, vibrations of the elevator car from the acceleration sensor can not be clearly assigned to a movement or a fault, so that in the end result the calculated position will deviate from the actual position. This is referred to as a "drifting" of the calculated position data with respect to the real position of the elevator car.

The acceleration sensor is preferably designed as a 3-axis sensor. Other sensor configurations are conceivable. However, it is important that the accelerations occurring in the direction of travel of the elevator car can be detected.

According to the invention, the elevator system is equipped with an image acquisition unit. The image acquisition unit is attached to the elevator car and arranged to be movable together with the elevator car.

In order to solve the problem, the arithmetic unit compares the recorded images with mapping images of the elevator shaft in order to determine an image-based current position. Further, the arithmetic unit recalibrates the current position using the image-based current position. In this case, a second possibility of position determination and thus redundancy of the method according to the invention is created by comparing the recorded images with the mapping images.

Mapping images are images that, in their entirety, represent an image of the elevator shaft. The mapping images are preferably taken during a learning drive during the commissioning of the elevator and clearly assigned to a position of the elevator car in the elevator shaft, so that the subsequent determination of the image-based position is possible. Here are the mapping images stored with the assigned position values in a database.

The determination of the current position thus takes place initially by means of the calculated current position by the acceleration data obtained by the acceleration sensor until an image-based current position is again determined and the current position is recalibrated. Thus, a so-called "drift" of the calculated current position is counteracted by the image-based current position. It is advantageous in such an embodiment that for recalibration does not need to be approached as in the processes and systems of the prior art, a top and / or bottom floor but the calibration over the entire hoistway at any time, for example during a journey, can take place.

Preferably, in a predetermined or predefinable first time interval, image recordings of the elevator shaft are taken by the image acquisition unit. Two successively recorded images are compared by the arithmetic unit to determine a spatial displacement of both images, wherein the acceleration data are used to determine the position and / or speed of the elevator car only when a spatial displacement of the arithmetic unit based on the recorded images determined has been. The images compared by the arithmetic unit need not necessarily be recorded immediately one after the other.

It can be seen that in order to increase the reliability of the method, it is optically determined with the aid of the image acquisition unit whether the elevator car has moved, i. has traveled a distance in the elevator shaft. Only then will the acceleration data be used to calculate the current position. Thus, disturbances due to vibrations, which occur for example during loading and unloading of an elevator car and are detected by the acceleration sensor, can be excluded.

The images are preferably recorded only when the acceleration sensor measures acceleration data of the elevator cars. This ensures that the arithmetic unit does not constantly have to compare images from the image acquisition unit but a comparison only in case of detection of acceleration (and therefore possible movement) by the acceleration sensor.

Acceleration data with a frequency of 100 Hz are preferably recorded.

Images are preferably recorded at a frequency of 60 Hz.

The image recordings are preferably recorded only if the acceleration data lie above a predefined or predefinable threshold value.

This is to ensure that accelerations generated by the acceleration sensor, e.g. during the loading and unloading of the elevator car, do not trigger the image acquisition unit. It is thus possible to use a relatively inexpensive and simple arithmetic unit, as they do not process continuously image recordings and may need to save.

Acceleration data which are above a predefined or predefinable second threshold value are preferably rejected by the arithmetic unit.

This preferred embodiment is also based on the idea of limiting the computing capacity of the arithmetic unit to a minimum. In addition, thus acceleration data which are above the second threshold, and which experience has caused by disturbances are not taken into account. For example, accelerations greater than 1 g, which occur during emergency braking of the elevator car, can be excluded, since in this case it is ensured by an emergency brake arrangement that the elevator car comes to a standstill.

Particularly preferably, the current position is recalibrated if a deviation between the image-based current position and the calculated current position is above a predefined or predefinable threshold value. In this case, the image-based current position, which has been directly and uniquely determined, is set instead of the calculated current position (which has been determined indirectly via the acceleration data).

Alternatively, the recalibration of the current position with the image-based current position may occur at a second interval. In this alternative, each time the captured images are compared with the mapping images, where an image-based current position is determined, the current position is recalibrated. This recalibration thus takes place continuously at second intervals.

The image-based current position is therefore preferably determined with images recorded in a predefined or predefinable second time interval, the second time interval being greater than or equal to the first time interval. Also in this case, a relief of the arithmetic unit is achieved. Not all images taken by the image acquisition unit are used for the determination of the image-based current position, and thus the computational complexity of the arithmetic unit is reduced. The second time interval is particularly preferably in the range between 500 and 100 ms, which corresponds to a frequency of 2 to 10 Hz.

The mapping images are preferably stored in a database during the learning run of the elevator car. This database is connected to the arithmetic unit. A memory address of a mapping image in the database is defined as a function of the position along the hoistway. The arithmetic unit uses the calculated current position to narrow a search of a mapping image in the database.

In this case, when comparing the recorded images with the mapping images to determine an image-based current position, the mapping image associated with the captured image can be found more quickly in the database. The advantage of this is even twofold, because a mapping image can not only be found faster, but the computing capacity of the arithmetic unit can also be further reduced.

The invention furthermore relates to a system for determining the position of an elevator car of an elevator system which can be moved in an elevator shaft. Such a system may preferably be operated by a method mentioned above. It can therefore be seen that the advantages mentioned above with regard to the method according to the invention also apply correspondingly to the system according to the invention.

The elevator car is equipped with an acceleration sensor. The system further comprises a computing unit, which is designed to detect acceleration data from the acceleration sensor and to calculate a current position and / or speed of the elevator car based on an initial position and the acquired acceleration data.

According to the invention, the system further comprises an image acquisition unit, which is designed to record image recordings of the elevator shaft and to transmit them to the arithmetic unit. Further, the arithmetic unit is configured to compare captured images with mapping images of the elevator shaft to determine an image-based current position and to recalibrate the current position using the image-based current position.

Preferably, the image acquisition unit is further configured to record image recordings of the elevator shaft at a predefined or predefinable first time interval and to transmit them to the arithmetic unit. Furthermore, the arithmetic unit is designed to compare two consecutively recorded images with one another in order to determine a spatial displacement of both images and to use the acceleration data for determining the position and the speed of the elevator car only if a spatial displacement is determined by the arithmetic unit becomes.

Preferably, the arithmetic unit is designed to control and / or regulate the image acquisition unit for image acquisition when acceleration data of the elevator car are detected.

Preferably, the arithmetic unit is designed to detect acceleration data only if they are above a predetermined or specifiable threshold value. More preferably, the arithmetic unit is designed to discard acceleration data which are above a predetermined or predefinable second threshold value.

More preferably, the arithmetic unit is designed to, if a deviation between the current image-based position and the current position is above a predetermined or predeterminable threshold, the current calculated position with to recalibrate the current image-based position. Alternatively, the arithmetic unit is adapted to recalibrate the current position at a second time interval with the image-based current position.

More preferably, the arithmetic unit is configured to determine the image-based current position with images recorded in a predefined or predefinable second time interval, wherein the second time interval is greater than or equal to the first time interval.

Preferably, a database is provided, which is designed to store mapping images that were generated during a learning trip of the elevator car. In this case, a memory address of a mapping image in the database is defined as a function of the position along the elevator shaft. Furthermore, the arithmetic unit is designed to limit a search of a mapping image in the database using the calculated current position.

The invention further relates to an elevator installation which is equipped with an abovementioned system for determining the position of the elevator car.

The advantages will be apparent from the description above regarding the method or system.

• Fig. 1. eine schematische Schnittansicht einer exemplarischen Ausführungsform einer Aufzugsanlage mit einem erfindungsgemässen System zur Bestimmung der Position;
• Fig. 2 eine Detailansicht einer exemplarischen Ausgestaltung des Auslegers der Figur 1;
• Fig. 3 einen exemplarischer Bildvergleich von zwei aufeinanderfolgend aufgenommenen Bildern in einem ersten vorgebbaren Zeitabstand.
• Fig. 4 eine graphische Darstellung von exemplarischen Beschleunigungsdaten sowie daraus berechneten Position und Geschwindigkeit der Aufzugskabine;
• Fig. 5 eine graphische Darstellung der berechneten und bildbasierten Position; und
• Fig. 6 einen exemplarischen QR-Code, der zur Anzeige einer Stockwerksposition dient.

The invention will be explained by way of example with reference to an embodiment in conjunction with the figures. Show it:

• Fig. 1 , a schematic sectional view of an exemplary embodiment of an elevator installation with a system according to the invention for determining the position;
• Fig. 2 a detailed view of an exemplary embodiment of the boom of FIG. 1 ;
• Fig. 3 an exemplary image comparison of two successive recorded images in a first predetermined interval.
• Fig. 4 a graphical representation of exemplary acceleration data and from this calculated position and speed of the elevator car;
• Fig. 5 a graphical representation of the calculated and image-based position; and
• Fig. 6 an exemplary QR code that is used to display a floor position.

In the FIG. 1 an elevator system 3 is shown, which is equipped with a system 7 according to the invention for determining the position. The elevator system 3 comprises an elevator car 2, which is arranged to be movable in an elevator shaft 1 along an axis z. Not shown are any carrying and traction means, which are used for carrying and moving the elevator car 2 application.

The elevator car 2 is further provided with an acceleration sensor 4, which is connected to a computing unit 5. The connection between the acceleration sensor 4 and the arithmetic unit 5 is shown schematically with a dashed line. This can be a direct connection via cable, for example with a bus system, or even a wireless connection. In the in the FIG. 1 illustrated embodiment, the computing unit 5 is arranged on the elevator car 2. However, the arithmetic unit 5 does not necessarily have to be arranged in the elevator shaft 1.

The acceleration sensor 4 measures the accelerations Dg occurring in the elevator car 2 and transmits them to the arithmetic unit 5. Particularly important are the accelerations occurring in the Z direction, which can represent a movement of the elevator car 2 and consequently must be detected reliably.

The elevator car is further equipped with a camera 6, here by way of example a CCD camera, which is attached to the elevator car 2 by means of a boom 9. The boom 9 allows adjustment of the orientation of the camera 6 and also allows retrofitting in existing elevator systems.

The camera 6 is also connected to the arithmetic unit 5, as shown schematically by the dashed line. To illuminate the elevator shaft 1 is a Headlight 8, for example, an LED headlight, arranged on the boom 9. The camera 6 can thus record a sufficiently illuminated area of the elevator shaft 1, which improves the quality of the image recordings and consequently increases the reliability of the image comparison.

In the FIG. 2 an exemplary embodiment of the boom 9 is shown. The camera 6 can be pivoted for adjustment about a pivot axis, as indicated by the double arrow 10. In addition, the headlight 8 can be both pivoted about a pivot axis 11 and displaced along the boom 9, as indicated by the double arrows 11 and 12 respectively.

The camera 6 is operated at a recording rate of 60 Hz. By comparing two consecutively recorded images B1 and B2, it can be determined whether a shift Δz of the images in the z direction has taken place. In FIG. 3 such a shift Δz is shown between two consecutively taken pictures B1 and B2. In particular, the shows FIG. 3 by way of example a displacement Δz based on a fastener 19.1, 19.2. The fastening element 19.1 appears in the lower area of the first image B1. In the second image B2, the fastening element 19.2 appears higher by the displacement Δz. The displacement Δz determined in the images B1 and B2 thus corresponds to a downward travel of the elevator car 2 by Δz. This comparison is preferably carried out on the basis of a gray value comparison of the two images B1 and B2. It can therefore be determined whether the elevator car has been moved in the z direction. These optically determined data are used to supplement the data from the acceleration sensor 4.

Based on the acceleration sensor 4 can be determined whether the elevator car 2 experiences an acceleration Dg. From this, a position zt of the elevator car 2 can be derived. However, a movement at a constant speed is not detected by the acceleration sensor 4, since in this case the measured acceleration of the elevator car is zero. Due to the optical motion detection, however, a distinction can be made between standstill and movement of the elevator car 2. As a result, the (inertia-based) position determination based on the data from the acceleration sensor 4 is used only when a movement of the elevator car 2 is optically detected.

In the FIG. 4 the data acquired by the acceleration sensor 4 are shown. With Dg, a curve of the acceleration of the elevator car 2 measured by the acceleration sensor 4 is shown. When the car is stationary, the acceleration measured by the acceleration sensor 4 is 9.81 m / s 2. By integrating the acceleration Dg, it is thus possible to calculate the velocity vt and the inertia-based position zt, which are also shown in FIG FIG. 4 in m / s or m are shown. In the in the FIG. 4 As shown by the arrows EG, the elevator car 2 was regularly stopped at a stop z = 0 m. It can be seen, however, that the inertia-based position zt calculated from the acceleration data Dg never has the value 0 m after a first drive but steadily diverges from this value. At a time of about 670 s, this divergence, referred to as "drifting", is even about 1 m, as indicated by the arrow 13.

In order to determine the current position of the elevator car, images which have been recorded with a time interval of 100 to 200 ms are also compared with mapping images from a database. The mapping images from the database have been taken during a learning journey, for example during the startup of the elevator system 3, and clearly assigned to a position of the elevator car 2 in the elevator shaft 1. It is thus possible to determine the position eg of the elevator car 2 on the basis of a direct, image-based measurement and not as usual by means of indirect methods.

Particularly advantageously, when determining an image-based current position, eg, in which a captured image is compared with mapping images, the arithmetic unit searches the database for a matching mapping image with the aid of a calculated current position. In this case, the search on the database can be greatly restricted since the memory addresses of the mapping images are formed as a function of the position along the elevator shaft 1.

In particular, due to a thermally induced expansion or shrinkage or due to a gravity-induced setting of a building, the accuracy of indirect methods such as, for example, an incremental disk or a magnetic tape coding decreases. The system 7 is not affected by such a decrease in accuracy because the visually determined, image-based position zBt is independent of the above confounding factors.
The current image-based position, eg, which has been optically determined as described above, is further used to correct the position zt calculated by means of acceleration data from the acceleration sensor 4.

In this case, the optically ascertained, image-based position is compared, for example, with the inertia-based position zt calculated from the acceleration data of the acceleration sensor 4, which is subject to "drifting". If the deviation between the optically determined, image-based position, eg, and the calculated, inertia-based position, is too great, the position is recalibrated. During recalibration, the optically determined, image-based position is set, for example, as the current position. Based on this, the acceleration data from the acceleration sensor 4 are then used as described above in order to further determine the position zt of the elevator car 2. It may thus be based on the use of other positioning systems such as e.g. an incremental disk or a magnetic coding are dispensed with. In addition, such a recalibration is possible at any time and not as usual only at the top or bottom stop of an elevator car 2.

As mentioned above, alternatively, the recalibration of the current position zt at intervals t2 between 100 to 200 ms at each comparison of a recorded image with mapping images, in which an image-based current position is determined, can take place.

In the FIG. 5 the sequence of such a recalibration is shown, wherein the right diagram represents an enlargement of the framed area of the left diagram. It can be seen that the calculated, inertia-based position deviates zt over time from the optically determined, image-based position z. If the deviation is above a threshold value, the calculated inertia-based position zt is recalibrated zt by the optically determined, image-based position zBt is set as the current position of the inertia-based positioning system, as indicated by the arrow 14. The position is then determined as described above until the deviation between the optically determined, image-based position zBt and the calculated, inertia-based position zt again reaches the threshold and a new recalibration takes place, as indicated by the arrow 14 '.

The FIG. 6 shows a schematic representation of a section of the elevator system 3 at a floor 17, wherein the FIG. 6 shows a situation in which an elevator car 2 in the shaft 1 in vertical travel in the direction z is about to approach the floor 17. The shaft 1 is opposite the floor 17 by a shaft door 16 lockable. At the shaft door 16 side facing the elevator car 2, a car door 15 is provided. The floor 17 is marked with a floor marking 18, here exemplarily designed as a QR code, which lies in the field of view of the camera 6 and can be detected by it. The camera 6 is mounted on the boom 9, which is fastened, for example, to the cabin floor 2.1 of the elevator car 2. The floor marking 18 is preferably characteristic for each floor 17, so that an automatic recognition of the floor positions of all floors 17 along the shaft 1 is possible due to the detectable by the camera 6 floor markings 18.

The floor markings 18 detected imagewise by the camera 6 can also be recorded in a learning run as mapping pictures KB and are stored accordingly in the database. The images taken in the area of the floor markings 18 are particularly easily assignable to a mapping image KB, so that a calibration of the calculated current position zt in the area of the floor markings 18 is particularly robust. In a time-limited failure of the system 7, the floor marking 18 can thus also serve as a catchment point or initial position z0 for the recalculation of the current position zt.

In-house tests have shown that dimensioning the QR code 18 is important for flawless detection of floor positions. Preferably, the QR code 18 has a dimension of at least 3cm x 3cm with an optimum range of dimension between 4cm x 4cm and 6cm x 6cm. In the case of even larger QR codes, recognition is likewise ensured, but only with a correspondingly large field of view of the camera 6.

It can be seen that such a system 7 for determining the position of a Elevator car 2 in existing elevator systems 3 can be easily retrofitted. In this case, only the camera 6 and possibly the headlight 8 must be attached to the elevator car and connected to the arithmetic unit 5. It is advantageous if the computing unit 5 is the already existing control and / or control unit of the elevator system 3, which is upgraded by software update or addition of a hardware module. Optionally, floor markings 18 in the shaft 1 at the floors 17 can be arranged. This is followed by a learning run, in which the mapping images of the elevator shaft 1 are taken and assigned to a position of the elevator car 2.

Such a system 7 allows a very accurate position determination with errors less than 0.5 mm at elevator speeds up to 5 m / s.

Claims (15)

1. A method for determining the position (z_t) of an elevator cab (2) of an elevator system (3), movably arranged within an elevator shaft (1), the elevator cab (2) being equipped with an acceleration sensor (4), comprising the following steps:

   - determination of the acceleration data (D_g) from the acceleration sensor (4) via a computing unit (5),

   - calculation by the computing unit (5) of the current position (z_t) and/or speed (v_t) of the elevator cab (2) based on an initial position (z₀) and on the recorded acceleration data (D_g),

   characterized in that the elevator system (3) is equipped with an image capture unit (6),

   - the image capture unit (6) recording images (B_n) of the elevator shaft (1),

   - the computing unit (5) comparing captured images (B_n) to mapped images (KB) of the elevator shaft (1) in order to derive an image-based current position (z_Bt) and

   - the computing unit (5) performing a recalibration of the current position (z_t) using the current, image-based position (z_Bt).
2. The method according to claim 1, characterized in that, within a specified or specifiable first time interval (Δt₁), images (B_n) of the elevator shaft (1) are recorded by image the capture unit (6) and
   two successive captured images (B₁, B₂) are compared to each other by the computing unit (5) to determine a spatial shift (z) of both images (B₁, B₂), the acceleration data (D_g) only being to determine the position (z_t) and/or speed (vt) of the elevator cab (2) if a spatial shift (z) has been detected by the computing unit (5).
3. The method according to claim 2, characterized in that the images (B₁, B₂) are only recorded if the acceleration sensor (4) measures acceleration data (D_g) of the elevator cab (2).
4. The method according to claim 2, characterized in that the images (B₁, B₂) are only recorded if the acceleration data (D_g) are above a specified or specifiable threshold value (D_s) and/or that acceleration data (D_g) that are above a specified or specifiable second threshold value (D_S2) are discarded by the computing unit (5).
5. The method according to claim 1, characterized in that the current position (z_t) is recalibrated using the image-based current position (z_Bt) within a second time interval (Δt₂), or that, if a discrepancy between the image-based current position (z_Bt) and the calculated position (z_t) is above a specified or specifiable threshold value (Z_S), the current position (z_t) is recalibrated using the image-based current position (z_Bt).
6. The method according to claim 1 or 5, characterized in that the image-based current position (z_Bt) is determined using images (B_n) captured within a specified or specifiable second time interval (Δt₂), the second time interval being greater than or equal to the first time interval (Δt₂ ≥ Δt₁).
7. The method according to claim 1, characterized in that the mapped images (KB) are stored during a learning run of the elevator cab (2), a memory address of a mapped image (KB) being defined in the database as a function of the position (z_t) along the elevator shaft (1), and that the calculated current position (z_t) is used by the computing unit (5) in order to narrow a search for a mapped image (KB) in the database.
8. A system (7) for determining the position (z_t) of an elevator cab (2) of an elevator system (3) movably arranged within an elevator shaft (1), in particular having a method according to any of the preceding claims, wherein the elevator cab (2) is equipped with an acceleration sensor (4), including a computing unit (5) that is designed to detect acceleration data (D_g) from the acceleration sensor (4) and to calculate a current position (z_t) and/or speed (v_t) of the elevator cab (2) based on an initial position (z₀) and the detected acceleration data (D_g),
   characterized in that the system (7) also includes an image capture unit (6) that is designed to take images (B_n) of the elevator shaft (1) and to send them to the computing unit (5) and that the computing unit (5) is also designed to compare the captured images (B_n) to the mapped images (KB) of the elevator shaft (1) in order to determine an image-based current position (z_Bt) and to perform a recalibration of the current position (z_t) using the image-based current position (z_Bt).
9. The system (7) according to claim 8, characterized in that the image capture unit (6) is further designed to take images (B_n) of the elevator shaft (1) within a specified or specifiable first time interval (Δt₁) and that the computing unit (5) is further designed to compare two successive captured images (B₁, B₂) to each other in order to determine a spatial shift (z) of both images (B₁, B₂) and only to use the acceleration data (D_g) for determination of the position (z_t) and/or speed (v_t) of the elevator cab (2) if a spatial shift (z) is detected by the computing unit (5).
10. The system (7) according to claim 9, characterized in that the computing unit (5) is designed to control and/or regulate the image capture unit (6) for capturing images (B₁, B₂) only if the acceleration data (D_g) of elevator cab (2) are detected.
11. The system (7) according to claim 10, characterized in that the computing unit (5) is designed to record acceleration data (D_g) only if these are above a specified or specifiable threshold value (D_S) and/or that the computing unit (5) is designed to discard acceleration data (D_g) that are above a specified or specifiable threshold value (D_S2).
12. The system (7) according to claim 8, characterized in that the computing unit (5) is designed to recalibrate the current position (z_t) within a second time interval (Δt₂) using the image-based position (z_Bt), or that the computing unit (5) is designed to recalibrate the current position (z_t) using the image-based current position (z_Bt) if a discrepancy between the image-based current position (_Bt) and the calculated current position (z_t) is above a specified or specifiable threshold value (Z_S).
13. The system (7) according to claim 8 or 12, characterized in that the computing unit (5) is designed to determine the image-based current position (z_Bt) using images (B_n) captured within a second specified or specifiable time interval (Δt₂), the second time interval being greater than or equal to the first time interval (Δt₂ ≥ Δt₁).
14. The system (7) according to claim 8, characterized in that a database is provided that is designed to store mapped images (KB) that were created in a learning run of the elevator cab (2), a memory address of a mapped image (KB) being defined in the database as a function of the position along the elevator shaft (1), and that the computing unit (5) is designed to narrow a search for a mapped image (KB) in the database using the calculated current position (z_t).
15. An elevator system having a system (7) for determining the position of the elevator cab (2) according to any of claims 8 to 14.

Images