METHOD AND SYSTEM FOR DETERMINING THE POSITION OF AN ELEVATOR CAR OF AN ELEVATOR INSTALLATION

Abstract

A method and system for determining the position of an elevator car movable in a shaft of an elevator installation capture images of shaft components and/or shaft equipment serving other functions using an image capture unit on the car. A current image is compared with at least one stored comparison image of the components and/or equipment in a travel direction of the car to determine a current position of the car in the travel direction. If there is no information regarding the position of the car from a previous determination step, for example if the elevator installation is restarted, a current image is compared with all the stored comparison images. Based upon this comparison, one possible position or a plurality of possible positions of the car are determined. These possible positions are checked at least once before one of these positions is adopted as the current position of the car.

Claims

1-15. (canceled)

16. A method for determining the position of an elevator car, which elevator car is movable in an elevator shaft of an elevator installation, including capturing images of components and/or shaft equipment serving other functions in the elevator installation using an image capture unit arranged on the elevator car and comparing a current one of the images with at least one stored comparison image of the shaft components and/or shaft equipment in a direction of travel of the elevator car to determine a current position of the elevator car in the direction of travel, the method comprising the steps of: performing a start phase including the steps of taking a start image with the image capture unit when the elevator car is stationary at an unknown start position in the elevator shaft, determining a start comparison characteristic value for every possible position of the elevator car in the direction of travel, which value indicates a measure for a match of the start image with the stored comparison image associated with each of the possible positions, determining a start assumption position of the elevator car based on the start comparison characteristic values and a start evaluation criterion; performing a review phase including the steps of moving the elevator car along a review travel path from the unknown start position to a review position in the elevator shaft, taking a review image with the image capture unit at the review position of the elevator car in the elevator shaft, determining a review assumption position of the elevator car from the previous start assumption position of the elevator car and the review travel path, determining a review comparison characteristic value for the review assumption position of the elevator car, the review comparison characteristic value indicating a measure for a match of the review image and with the stored comparison image associated with the review assumption position, and performing a decision phase including making a decision, on the basis of the review comparison characteristic, whether to determine the review assumption position as the current position of the elevator car, carry out another one of the review phase and another one of the decision phase or, exclude the review assumption position as the current position of the elevator car.

17. The method according to claim 16 including in the decision phase, determining the review assumption position as the current position of the elevator car if the associated review characteristic value fulfills a predetermined decision determination criterion.

18. The method according to claim 17 wherein if the decision determination criterion is checked for more than one of the review assumption position, a review assumption position is only then determined as the current position of the elevator car if only the review comparison characteristic value associated with this review assumption position fulfills the decision determination criterion.

19. The method according to claim 17 wherein the review assumption position is determined as the current position of the elevator car only if at least one additional decision criterion that is independent of the review comparison characteristic value is fulfilled.

20. The method according to claim 19 wherein, as a decision criterion, including checking whether a travel path between the start position and the current review assumption position is greater than a predefined minimum travel path.

21. The method according to claim 16 including terminating the determination of the position of the elevator car if a predetermined termination criterion is fulfilled.

22. The method according to claim 21 including, as a termination criterion, checking whether an entire travel path of the elevator car, starting from the start position, has exceeded a predetermined maximum travel path.

23. The method according to claim 21 including restarting the determination of the position of the elevator car after a termination and moving the elevator car in an opposite direction in the review phase compared with the review phase before the termination.

24. The method according to claim 16 wherein, in the review phase, determining review comparison characteristic values for a region around the review assumption position of the elevator car and using the position belonging to the review comparison characteristic value indicating the greatest match as the review assumption position for the subsequent decision phase.

25. The method according to claim 16 including comparing the current image with the stored comparison image transversely to the direction of travel to determine the current position of the elevator car in the direction of travel.

26. The method according to claim 16 including determining the start comparison characteristic values by also comparing the start image with the comparison image transversely to the direction of travel and using the comparison characteristic value which indicates the greatest match of the start image with the comparison image of a position as the start comparison characteristic value of the particular position.

27. The method according to claim 16 wherein, in order to determine the review comparison characteristic values, also comparing the review image with the comparison image transversely to the direction of travel and using the comparison characteristic value which indicates the greatest match as the review comparison characteristic value for the subsequent decision phase.

28. The method according to claim 16 including, in the review phase, moving the elevator car at a lower speed in comparison with a normal operation speed of the elevator installation.

29. The method according to claim 16 including evaluating further information that can be acquired in the elevator shaft in order to determine the position of the elevator car.

30. A system for determining the position of an elevator car, which elevator car is movable in an elevator shaft of an elevator installation, comprising: a computing unit; an image capture unit arranged on the elevator car and being adapted to take images, consisting of individual pixels, of shaft components and/or shaft equipment serving other functions in the elevator installation and to transmit the images to the computing unit; and wherein the computing unit compares a current one of the images with at least one stored comparison image of the shaft components and/or shaft equipment in a direction of travel of the elevator car and determines a current position of the elevator car in the direction of travel by performing the following phases directly or indirectly, a start phase having the steps of taking a start image with the image capture unit when the elevator car is stationary at an unknown start position in the elevator shaft, determining a start comparison characteristic value for every possible position of the elevator car in the direction of travel, which value indicates a measure for a match of the start image with a stored comparison image of the particular position, determining a start assumption position of the elevator car on the basis of the start comparison characteristic values and a start evaluation criterion, a review phase having the steps of moving the elevator car along a review travel path to a review position in the elevator shaft, taking a review image with the image capture unit at the review position of the elevator car in the elevator shaft when the elevator car is stationary, determining a review assumption position of the elevator car from the previous start assumption position of the elevator car and the review travel path, determining a review comparison characteristic value for the review assumption position of the elevator car, the review comparison characteristic value indicating a measure for a match of the review image and with the comparison image of the review assumption position, and a decision phase in which a decision is made, on the basis of the review comparison characteristic, whether to determine the review assumption position as the current position of the elevator car, carry out a further review phase and a further decision phase, or exclude the review assumption position as the current position of the elevator car.

Description

DESCRIPTION OF THE DRAWINGS

[0075] In the drawings:

[0076] FIG. 1 schematically shows an elevator installation having a system for determining the position of an elevator car that is arranged so as to be movable in an elevator shaft,

[0077] FIG. 2 shows a comparison image within a current comparison region of a current image of shaft equipment of the elevator shaft,

[0078] FIG. 3 shows correlation coefficients of a comparison image with an image therebelow of a current comparison region of a current image with different shifts transversely to a direction of travel (x direction) and a constant shift in the direction of travel (z direction) of the elevator car,

[0079] FIG. 4a shows start comparison characteristic values of a start image after completion of a start phase of a method for determining a position of an elevator car in an elevator shaft,

[0080] FIG. 4b shows review comparison characteristic values of two review assumption positions after completion of a first review phase following the start phase and

[0081] FIG. 4c shows review comparison characteristic values of a review assumption position remaining after a first decision phase after completion of a second review phase following the decision phase.

DETAILED DESCRIPTION

[0082] According to FIG. 1, an elevator installation 10 has an elevator shaft 12 oriented in the vertical direction. Arranged within the elevator shaft 12 is an elevator car 14, which is connected via a suspension means 16 in the form of a flexible belt or a rope to a counterweight 18 in a known manner. The suspension means 16 extends from the elevator car 14 via a drive pulley 20, which can be driven by a drive machine (not shown). The elevator car 14 can be moved up and down in the elevator shaft 12 by means of the drive machine and the suspension means 16. The elevator car 14 can thus be moved in or counter to a direction of travel 22, which extends upward in the vertical direction, in the elevator shaft 12.

[0083] A guide rail 26 which extends in the direction of travel 22 is secured to a shaft wall 24 of the elevator shaft 12. The shaft wall 24 can be referred to as a shaft component and the guide rail 26 as shaft equipment. When the elevator car 14 is moved, it is guided along the guide rail 26 via guide shoes (not shown).

[0084] A system 28 for determining the position of the elevator car 14 is arranged on the elevator car 14. The system 28 has a computing unit 30 and an image capture unit 32. The image capture unit 32, which is designed as a digital camera, is oriented in such a way that it can capture images of the guide rail 26. It transmits the images of the guide rail 26, which consist of individual pixels, to the computing unit 30, which compares a current image with at least one stored comparison image of the guide rail 26 in order to determine a current position of the elevator car 14 in the direction of travel 22. The computing unit 30 transmits the current position of the elevator car 14 via a signal connection (not shown) to an elevator control 31 arranged in the elevator shaft 12, which control uses the position of the elevator car 14 to control the elevator installation 10.

[0085] The computing unit 30 does not have to be arranged on the elevator car. It can also be arranged so as to be stationary in the elevator shaft and so as to be in signal communication with the image capture unit 32. The image capture unit could also take images of the shaft wall and transmit them to the computing unit.

[0086] To determine the current position of the elevator car 14 in the elevator shaft 12, the computing unit 30 compares a stored comparison image 34 shown in FIG. 2 with a current image 36 from the image capture unit 32.

[0087] Comparison images for a so-called relative and so-called absolute position determination are stored in a memory (not shown) of the computing unit 30. A large number of comparison images 34 are stored for the absolute position determination. When the system 28 is started up, these comparison images 34 are derived from current images from the image capture unit 32 and stored during a so-called training run. During the training run, the elevator car 14 is moved with the system 28 along the entire travel path of the elevator car 14 in the elevator shaft 12. The computing unit 30 derives individual comparison images 34 from the images taken by the image capture unit 32 and associates them with a position in the elevator shaft 12. The computing unit 30 derives the comparison images 34 in such a way that they overlap each other twice. In particular, they overlap in such a way that, in each case, one comparison image abuts the next-but-one comparison image. The stored comparison images 34 thus cover the entire travel path of the elevator car 14. As soon as a comparison image 34 is identified in a current image 36 from the image capture unit 32, the position of the elevator car 14 in the direction of travel 22 can be inferred with the help of the position of the comparison image 34 in the elevator shaft 12 which is likewise stored. The comparison images are selected on the basis of the position of the elevator car 14 at a previous determination time and the speed of the elevator car 14. This severely limits the number of comparison images required for the comparison.

[0088] In order to derive a comparison image 34 from a current image from the image capture unit 32 during the training run, the current image is post-processed by the computing unit 30. For this purpose, the computing unit 30 first selects a section in the center of the current image. The computing unit 30 then calculates the mean value of all the pixel values of the selected section and subtracts the calculated mean value from each pixel value. The result of this post-processing is saved as the comparison image 34. Additional post-processing, such as low- and/or high-pass filtering, can also be carried out.

[0089] In addition, the computing unit 30 determines a structure parameter for each post-processed and stored comparison image 34 and stores this together with the comparison image 34. The computing unit 30 starts from an image post-processed as described above. It squares the pixel values of all pixels and adds them up. The result of this summation or also the root thereof is stored together with the comparison image 34.

[0090] The comparison image for the relative position determination is derived from an image from the image capture unit 32 from the previous position determination. In the relative position determination, the current image is compared with an image captured during a previous position determination, with a shift of the current image relative to the image from the previous position determination in the direction of travel being determined in said comparison of the images. The current position of the elevator car can be determined from said shift and the position determined during the previous position determination. Even if the absolute position is not known in the previous position determination, the travel path covered and the direction can be determined from said shift.

[0091] In order to determine the position of the elevator car 14 in the direction of travel 22 during the normal operation of the elevator installation 10, the computing unit 30 compares a comparison image 34 with a current image 36 from the image capture unit 32 both in and transversely to the direction of travel 22. For this purpose, the computing unit 30 checks whether a comparison image 34 is contained in a current comparison region 38 of the current image 36. If this is the case, the position of the comparison image 34 in the current comparison region 38 is determined at the same time. In the following it is assumed that the comparison image 34 is contained in the current comparison region 38.

[0092] In order to determine the position of the comparison image 34 in the current comparison region 38, the computing unit 30 compares the comparison image 34 and the current comparison region 38 of the current image 36 both in the direction of travel (z direction) and transversely to the direction of travel (x direction). For this purpose, the comparison image 34 is shifted on a pixel-by-pixel basis both in the direction of travel (z direction) and transversely to the direction of travel (x direction) with respect to the current comparison region 38, and a comparison characteristic value is calculated for each position in the form of a correlation coefficient between the comparison image 34 and the image of the comparison region 38 below the comparison image 34. The comparison characteristic value in the form of the correlation coefficient is a measure of the match between the comparison image 34 and the current comparison region 38. The shift of the comparison image 34 is symbolized in FIG. 2 by the arrows 40.

[0093] The correlation coefficient is calculated using the following formula:

[00001]$k\left(r,s\right)=\frac{{\Sigma }_{\left(i,j\right)\in R}\left(I\left(r+i,s+j\right)-\stackrel{\_}{I}\left(r,s\right)\right)*\left(R\left(i,j\right)-\stackrel{\_}{R}\right)}{\sqrt{{{\Sigma }_{\left(i,j\right)\in R}\left(I\left(r+i,s+j\right)-\stackrel{\_}{I}\left(r,s\right)\right)}^2}*\sqrt{{{\Sigma }_{\left(i,j\right)\in R}\left(R\left(i,j\right)-\stackrel{\_}{R}\right)}^2}}$

where [0094] r=shift of the comparison image in the x direction, [0095] s=shift of the comparison image in the z direction, [0096] R(i,j)=pixel values of the comparison image in the x position i and the z position j, [0097] I(r+i,s+j)=pixel values of the current comparison region at the x position r+i and the z position s+j; [0098] R=mean value of all pixel values of the comparison image, [0099] I(r,s)=mean value of all pixel values of the current comparison region below the comparison image shifted by r in the x direction and s in the z direction.

[0100] Since, prior to being stored by the computing unit 30, the comparison image 34 was post-processed such that the mean value of all pixel values of the comparison image 34 was subtracted from each pixel value, the term

(R(i,j)−R)

no longer has to be evaluated during the calculation of the correlation coefficients, but rather it is possible to use the pixel values of the comparison image 34 directly.

[0101] In addition, as described above, a structure parameter of the comparison image 34 is also stored, which can be used directly for the calculation of the correlation coefficient. As described above, the following term is calculated as the structure parameter:

[00002]$\underset{\left(i,j\right)\in R}{\overset{}{.Math.}}{\left(R\left(i,j\right)-\stackrel{\_}{R}\right)}^2$

and either the result or the root thereof is stored. The structural parameter is thus accounted for when comparing the current image 36 with the stored comparison image 34.

[0102] The correlation coefficient is calculated for every possible position of the comparison image 34 in the current comparison region 38, i.e. for every possible shift by r in the x direction and s in the z direction. The correlation coefficients for all possible r and s values result in a three-dimensional surface. The maximum correlation coefficient of the entire surface indicates the position of the comparison image 34 in the current comparison region 38 with the highest match. On the above-mentioned condition that the comparison image 34 is contained in the current comparison region 38, said maximum indicates the position of the comparison image 34 at which there is a match between the comparison image 34 and the image therebelow. As an additional check, it can be checked whether the maximum correlation value is greater than a threshold value. With the information regarding the position of the comparison image 34 in the current comparison region 38 of the current image 36, the position of the elevator car 14 in the elevator shaft 12 in the direction of travel 22 can be determined either via a relative or an absolute position determination.

[0103] FIG. 3 shows, by way of example, the correlation coefficients on a k axis upward above the possible r values on an r axis to the right, i.e. the possible shifts in the x direction and a fixed s value, i.e. a constant shift in the z direction.

[0104] According to FIG. 3, the correlation coefficient reaches the maximum value kMn with an s value of sn and an r value of rMn. This means that the comparison image 34, with a fixed shift of sn in the z direction and with a shift by rMn in the x direction, has the greatest match with the image therebelow of the current comparison section 38 of the current image 36.

[0105] The computing unit 30 determines for each possible s-value s=sn the (local) maximum correlation coefficient kMn and the associated shift rMn in the x direction. The computing unit 30 then determines the maximum correlation value kMax of all determined (local) maximum correlation coefficients kMn, which represents the absolute maximum of the correlation coefficients and thus the three-dimensional surface described. The position of the comparison image 34 in the current comparison region 38 results from the associated s and r values of the absolute maximum of the correlation coefficient. The shift in the z direction and the position in the elevator shaft associated with the comparison image thus result in the position of the elevator car in the elevator shaft. A correlation coefficient can thus be associated with a position of the elevator car.

[0106] It is also possible for the comparison image to be shifted only in the z direction over the current image and for a correlation coefficient to be calculated in each case. In this case, the described determination of the maximum correlation coefficient for different shifts in the x direction, i.e. with different r values, is not required. The rest of the procedure remains otherwise the same.

[0107] After a restart of the system 28 for determining the position of the elevator car 14, the computing unit 30 has no information regarding the current position of the elevator car. The computing unit 30 then carries out a special method for the particularly reliable determination of the position of the elevator car 14, which is described below in connection with FIGS. 4a, 4b and 4c.

[0108] The method begins with a start phase in which the elevator car 14 is at an unknown start position 50. First, when the elevator car 14 is stationary, a start image (analogous to the current image 36 in FIG. 2) is taken using the image capture unit 32. A start comparison characteristic value in the form of an above-described cross-correlation coefficient is then determined for every possible position of the elevator car 14 in the direction of travel 22. For this purpose, a comparison is carried out with all stored comparison images (34 in FIG. 2), the comparison image being pushed over the start image in each case as described above. As described above, the shift can take place only in the z direction or in the z direction and x direction.

[0109] The result of such a determination is shown very schematically in FIG. 4a. The associated cross-correlation coefficient is shown as point 52 (plotted along the k axis) for each of the different positions (plotted along the h axis). The larger the cross-correlation coefficient, the more similar the comparison image of this position is to the current image.

[0110] At the end of the start phase, it is checked which start comparison characteristic values fulfill a start evaluation criterion. As the start evaluation criterion, it is checked whether the start comparison characteristic values are greater than a first threshold value (shown as line 54 in FIG. 4a). In the example shown, this is the case for the start comparison characteristic values 52a and 52b. The positions belonging to these start comparison characteristic values 52a, 52b are referred to as the first start assumption position PS1 and the second start assumption position PS2.

[0111] If no start comparison characteristic value fulfills the start evaluation criterion, the method is terminated.

[0112] After the start phase has been completed, the elevator car 14 is moved downward along a review travel path s1 to a review position 56 in a subsequent review phase. The elevator car 14 is moved at a lower speed in comparison with a normal operation of the elevator installation. The situation after the elevator car is moved is shown in FIG. 4b. To move the elevator car 14, the computing unit 30 sends a corresponding request to the elevator control 31. The elevator control 31 can ensure adherence to the review travel path s1 by appropriately actuating the drive machine and, if necessary, by measuring the speeds of the drive machine. Alternatively or in addition, as described above, the relative position determination can be used to determine the review travel path s1. After the elevator car 14 has been moved, a review image (analogous to the current image 36 in FIG. 2) is taken at the review position 56.

[0113] From the two start assumption positions PS1, PS2, the review travel path s1 and the downward direction of travel, two review assumption positions PA1.1 and PA2.1 are determined, which are each shifted downward along the review travel path s1 relative to the start assumption positions PS1, PS2. For these two review assumption positions PA1.1, PA2.1, review comparison characteristic values are determined in the form of cross-correlation coefficients described above. For this purpose, the review image is compared with the comparison images of the review assumption positions PA1.1 and PA2.1 (comparable with 34 in FIG. 2). As described above, the images can be shifted with respect to one another only in the z direction or in the z direction and x direction. The two review comparison characteristic values 58a, 58b are shown in FIG. 4b.

[0114] In a subsequent decision phase, it is decided how the method should be continued. First, it is checked whether the two review comparison characteristic values 58a, 58b fulfill a decision determination criterion. For this purpose, it is checked whether they are greater than a second threshold value, which is shown as line 60 in FIG. 4b. In this example, the second threshold value is identical to the first threshold value of the start phase. Both review comparison characteristic values 58a, 58b are smaller than the second threshold value, so that at this point in time neither of the two review assumption positions PA1.1, PA2.1 is determined as the actual, current position of the elevator car 14.

[0115] It is then checked whether the two review comparison characteristic values 58a, 58b fulfill a repetition evaluation criterion. For this purpose, it is checked whether they are greater than a third threshold value, which is shown as line 62 in FIG. 4b. The third threshold value is smaller than the second threshold value. This is only the case for the review comparison characteristic value 58a of the first review assumption position PA1.1. The second review comparison characteristic value 58b of the second review assumption position PA2.1 is smaller than the third threshold value.

[0116] Since the first review comparison characteristic value 58a fulfills the repetition evaluation criterion, a further review phase and a further decision phase are carried out for the associated review assumption position PA1.1. Since the second review comparison characteristic value 58b also does not fulfill the repetition evaluation criterion, the associated review assumption position PA2.1 is excluded as a possible current position of the elevator car 14.

[0117] Finally, it is checked whether a termination criterion is fulfilled. If so, the method is terminated. For this purpose, it is checked whether an entire travel path of the elevator car 14, starting from the start position 50, has exceeded a maximum travel path. Since the elevator car 14 has only been moved along a review travel path s1 since the start of the method, the entire travel path corresponds to a review travel path s1 which is naturally smaller than the maximum travel path. The termination criterion is therefore not fulfilled and the method is continued.

[0118] If the method were to be terminated, it would be restarted, with the elevator car 14 being moved in the opposite direction in the review phase, i.e. upward, in comparison with the review phase before the termination. In particular, at least one review travel path in a review phase, in particular in the first review phase, is selected to be different from the review travel paths before the restart.

[0119] In the example described, the method is continued such that the first decision phase is followed by a further, second review phase. This runs analogously to the first review phase described above, with only one review assumption position PA1.2 resulting from the review assumption position PA1.1 and the review travel path s1. The resulting review comparison characteristic value 64 is shown in FIG. 4c.

[0120] In the following decision phase, it is decided again how the method should be continued. First of all, it is checked whether the review comparison characteristic value 64 fulfills the decision determination criterion. For this purpose, it is checked whether it is greater than the second threshold value, which is also shown as line 60 in FIG. 4c. The review comparison characteristic value 64 is greater than the second threshold value, meaning the decision determination criterion is fulfilled.

[0121] A decision criterion which is independent of the review comparison characteristic value 64 is then checked. For this purpose, it is checked whether a travel path s2 between the start position 50 and the current review assumption position PA1.2 is greater than a definable minimum travel path. This is the case here, meaning the review assumption position PA1.2 is determined as the actual, current position of the elevator car 14. This has confirmed the assumption that the first start assumption position PS1 corresponded to the start position 50 of the elevator car 14 in the start phase.

[0122] Instead of determining the review comparison characteristic values only for a review assumption position in the review phase, this can also be done for a region around the review assumption position of the elevator car. The position which belongs to the review comparison characteristic value which indicates the greatest match is then used as the review assumption position for the subsequent decision phase. The position for which the largest cross-correlation coefficient results is therefore used.

[0123] Further information that can be acquired in the elevator shaft can also be evaluated. For example, an expert (not shown) can be detected who is arranged in the vicinity of a floor and assists with the exact positioning of the elevator car on a floor. If such an expert is identified, for example by means of a special sensor, all review assumption positions that are not in a possible region of such an expert can be excluded.

[0124] Finally, it should be noted that terms such as “having,” “comprising,” etc. do not preclude other elements or steps and terms such as “a” or “an” do not preclude a plurality. Furthermore, it should be noted that features or steps that have been described with reference to one of the above embodiments can also be used in combination with other features or steps of other embodiments described above.

[0125] In accordance with the provisions of the patent statutes, the present invention has been described in what is considered to represent its preferred embodiment. However, it should be noted that the invention can be practiced otherwise than as specifically illustrated and described without departing from its spirit or scope.