Method for generating an image of a route network, use of the method, computer program, and computer-readable storage medium

Abstract

A method for generating an image of a route network that is travelled through by a rail vehicle. The image is generated with the use of activities that are recorded by the rail vehicle as it travels through the route network and sorted in an activity sequence. In order to provide an improved method, patterns in the activity sequence are identified with use of a pattern detection method and the image of the route network is generated with the use of the identified patterns.

Claims

1. A method for generating an image of a route network, the method comprising: recording activities as a rail vehicle is travelling through the route network and sorting the activities into a sequence of activities; the activities of the rail vehicle including holding the rail vehicle and driving a distance of a certain length; recognizing patterns in the sequence of activities by way of a pattern recognition method; and producing the image of the network using the recognized patterns.

2. The method according to claim 1, which comprises sorting the activities in the activity sequence according to a pre-set criterion.

3. The method according to claim 2, which comprises sorting the activities in the activity sequence by time.

4. The method according to claim 1, which comprises: using the pattern recognition method for at least some of the recognized patterns, determining at least one feature of the respective pattern selected from the group consisting of a frequency, a length, and a form of the respective pattern; and using at least one feature of the respective pattern for generating the image of the route network.

5. The method according to claim 1, which comprises, for at least some of the recognized patterns, identifying the respective pattern as a certain section of the route network by at least one feature selected from the group consisting of a frequency, a length, and a form of the pattern.

6. The method according to claim 1, which comprises recognizing a frequently recurring sequence of activities comprising a predetermined minimum number of activities as a pattern and identifying as a main route of the route network.

7. The method according to claim 1, wherein a mirror image pattern that includes a predetermined minimum number of activities with a point of symmetry is recognized as a pattern and the point of symmetry is identified as a turning point within the route network.

8. The method according to claim 1, which comprises, if a part of a sequence of activities is definitely assigned to a previously identified section of the route network using the pattern recognition method, and a second part of the same sequence deviates from the already identified section, determining the second part as a branch from the identified route.

9. The method according to claim 1, which comprises encoding the activities in the sequence of activities by alphanumeric symbols.

10. The method according to claim 1, wherein the pattern recognition method includes a sequence matching algorithm.

11. The method according to claim 10, wherein the pattern recognition method includes a sequence alignment algorithm.

12. The method according to claim 1, wherein the pattern recognition method allows tolerances to a predetermined degree, so that similar sequences of activities can be recognized as the same pattern.

13. The method according to claim 1, wherein the activity of holding the rail vehicle is an activity selected from the group consisting of holding without opening a door, holding with opening a door on the right side of the rail vehicle, holding with opening a door on the left side of the rail vehicle, and holding with opening doors on both sides of the rail vehicle.

14. A method of locating a rail vehicle, the method comprising: implementing the method according to claim 1 for a localization of the rail vehicle at a predetermined time; comparing a further sequence of activities of the rail vehicle that includes a certain number of activities at a specified time with the image of the network previously generated; and assigning the rail vehicle to a specific section of the route network based on the comparison.

15. A computer-readable storage medium having stored thereon commands in non-transitory form which, when run by a computer, cause that computer to perform the method according to claim 1.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

(1) In the figures:

(2) FIG. 1 shows a flowchart of a method for creating an image of a route network and

(3) FIG. 2 shows a flowchart of the use of the method from FIG. 1 for locating the rail vehicle at a given time.

DESCRIPTION OF THE INVENTION

(4) FIG. 1 shows a flowchart 2 that illustrates a method for creating an image of a route network. The route network is being travelled through by a rail vehicle. When travelling through the route network, the activities that are recorded by the rail vehicle are sorted into a sequence of activities 4. The activities are sorted into the sequence of activities 4 by time. In addition, the activities in the sequence of activities 4 will be encoded by alphanumeric symbols.

(5) For example, a sequence of activities 4 might be as follows:

(6) 05-R-03-L-02-B-10-R-10-B-02-L-03-R-05-X-05-R-03-L-02-B-07-L-11-B-01-X

(7) In this example, the activities include driving a route of a certain length, which is encoded here, for example, by a two-digit number. The number corresponds to the rounded length of the route in kilometers. A different alphanumeric encoding would also be possible.

(8) Further, in this example, the activities include holding the rail vehicle, which is encoded by a letter in this example. In this example, holding without opening the door is encoded by an “X”, holding with opening a door on the right of the rail vehicle is encoded by an “R”, holding with opening a door on the left of the rail vehicle is encoded by an “L” and holding with opening of doors on both sides of the rail vehicle is encoded by a “B”.

(9) Using a pattern recognition method, patterns 8 in the activity sequence 4 are recognized. The pattern recognition method allows tolerances to a predetermined degree. For example, two sequences of activities that behave like an image and a mirror image relative to each other are recognized as the same pattern. This means that the reading direction in the activity sequence 4 does not matter. For example, the sequence of activities 05-R-03-L-02-B-10 and the sequence of activities 10-B-02-L-03-R-05 are recognized as the same pattern.

(10) In addition, using the pattern recognition method, for at least some of the recognized patterns 8 at least one characteristic 10 of the respective pattern 8 is determined, namely a frequency, a length and/or a form of the respective pattern 8. The frequency is an absolute frequency. A length is a number of activities of the respective pattern.

(11) Detected patterns 8 and identified features 10 thereof may be as follows, for example:

(12) TABLE-US-00001 Pattern Length Frequency Form 05-R-03-L-02-B-10-R- 15 1 Mirror image 10-B-02-L-03-R-05 B-02-L-03-R-05-X- 13 1 Mirror image 05-R-03-L-02-B 05-R-03-L-02-B-10 7 2 05-R-03-L-02 5 3 . . .

(13) The pattern recognition method allows tolerances to a predetermined extent. In this way, similar sequences of activities, also referred to as sequences below, are recognized as the same pattern. For example, two sequences of activities are recognized as the same pattern when the two sequences differ in exactly one activity to a predetermined extent. In particular, a deviation of a length of a travel distance of 1 km may be allowed for recognizing two sequences as the same pattern, for example to take into account rounding errors. In this example, a deviation of a number by ±1 can be allowed in order to recognize two sequences as the same pattern. Furthermore, a deviation of a type of hold may be allowed in order to recognize two sequences as the same pattern. In this example, a change of a letter may be allowed in order to recognize two sequences as the same pattern.

(14) Furthermore, for example, two sequences of activities are recognized as the same pattern when exactly one activity is exchanged for three activities according to a given rule. For example, in order to identify two sequences as the same pattern, a stopover may be allowed, wherein the sum of the distances directly before and after the stopover is the same as the distance without a stopover. In this example, it may be allowed if a first number is exchanged for two numbers and a letter, wherein the sum of the two numbers gives the first number, in order to recognize two sequences as the same pattern. For example, a sequence that includes a 10 and another sequence that includes the sequence 04-X-06 or the sequence 03-R-07 or similar instead of the 10 can be recognized as the same pattern.

(15) For at some of the recognized patterns 8, the respective pattern 8 is identified as a specific section 12 of the route network by the features 10 thereof, in particular by the frequency, the length and/or the form thereof.

(16) A frequently recurring sequence of activities that includes a predetermined minimum number of activities is recognized as a pattern. This frequently recurring sequence of activities that includes a predetermined minimum number of activities can be identified as a main route of the route network.

(17) In this example the mirror-image forms are not considered for identifying the main route. Of the other two longest recognized patterns, the most frequent pattern is identified as a main route of the network. In this way, the pattern 05-R-03-L-02 can be identified as the main route. The main route is shown in italics below.

(18) A mirror-image pattern that contains a predetermined minimum number of activities is also recognized as a pattern. A point of symmetry of the mirror image pattern is determined. The point of symmetry is identified as a turning point within the route network.

(19) For example, in this case the pattern 05-R-03-L-02-B-10-R-10-B-02-L-03-R-05 is recognized as a mirror image pattern. The point of symmetry of the mirror image pattern lies in the middle of the mirror image pattern and is highlighted here in bold and underlined.

(20) This point of symmetry is identified as the turning point within the route network.

(21) In this example, the following route is thus identified:

(22) 05-R-03-L-02-B-10-R.

(23) The pattern recognition method further recognizes that the latter route includes the main route. 05-R-03-L-02-B-10-R is thereby recognized as a route section.

(24) Furthermore, for example, the pattern B-02-L-03-R-05-X-05-R-03-L-02-B is recognized as a mirror image pattern. The point of symmetry of the mirror image pattern lies in the middle of the mirror image pattern and is highlighted here in bold and underlined. This point of symmetry is identified as the turning point within the route network. The pattern recognition method further recognizes that the latter route includes the main route. X-05-R-03-L-02-B-10-R is thereby recognized as a route section.

(25) If, using the pattern recognition method, a part of a sequence of activities is positively assigned to an already identified section of the route network and a second part of the same sequence deviates from the already identified section of the route, the second part is detected as a branch from the identified section of the route.

(26) Previously unassignable sequences can then be inserted gradually into the (partially) generated route network using a sequence alignment algorithm, in particular a free shift alignment.

(27) For example, in this example, the sequence 07-L-11-B-01-X could not be assigned. For free shift alignment, a longer sequence is taken that contains the unassigned sequence, and it is determined where the longer sequence matches.

(28) In this example, the sequence R-03-L-02-B-07-L-11-B-01-X is compared with the already identified route section X-05-R-03-L-02-B-10-R.

(29) The comparison of R-03-L-02-B-07-L-11-B-01-X with

(30) X-05-R-03-L-02-B-10-R

(31) provides that the first part, namely R-03-L-02-B, of the (longer) sequence can definitely be assigned to an already identified section of the route network, and the second part 07-L-11-B-01-X of the same sequence deviates from the already identified section of the route. Thus, the second part 07-L-11-B-01-X is recognized as a branch from the identified section of the route.

(32) Using at least one feature 10 of the respective pattern 8 and possibly using a sequence alignment algorithm, the image 14 of the route network can be generated. In this way, the image 14 of the route network can be generated using the recognized patterns.

(33) In this example, the image 14 is as follows:

(34) ##STR00001##

(35) Where appropriate, the generated image 14 of the route network may be compared to a topographical map. Based on this comparison, place names or station names can be assigned to the holding of the rail vehicle in the image, which are encoded here by letters.

(36) FIG. 2 shows a flowchart 16 regarding the use of the method from FIG. 1 to locate 20 the rail vehicle at a given time.

(37) A sequence of activities 18 (other than that referred to in FIG. 1) of the rail vehicle, which comprises a certain number of activities at the specified time, is compared with the image 14 of the route network shown in FIG. 1. On the basis of the comparison, the rail vehicle is assigned to a specific section of the route network (“localization 20” of the rail vehicle).

(38) In this example, the latter sequence of activities 18 is as follows:

(39) B-11-L-07

(40) For example, during the last activity “07”, namely when driving a distance of 7 km, an incident has occurred, and it is to be determined where this incident took place. The individual activity at the time of the incident is not sufficient to locate the rail vehicle at the time of the incident. However, if the latter sequence of activities 18 of the rail vehicle, which comprises a certain number of activities at the specified time, is compared with the image 14 of the route network produced in FIG. 1, the rail vehicle may be located (“localization 20”) in that by using the comparison the rail vehicle is assigned to a specific section of the route network.

(41) When comparing the latter sequence of activities 18:

(42) B-11-L-07

(43) with the generated image 14:

(44) ##STR00002##

(45) it is determined that the rail vehicle was travelling through the branch at the time of the incident and at the time of the incident is travelling on the first section “07” of the branch. The direction of the rail vehicle can also be determined using the comparison.

(46) Although the invention has been illustrated and described in detail by the preferred exemplary embodiments, the invention is not limited by the disclosed examples and other variations can be derived from this by the person skilled in the art without departing from the scope of protection of the invention.