COMPUTER-IMPLEMENTED METHOD FOR CREATING MEASUREMENT DATA DESCRIBING A RAILWAY NETWORK OR A VEHICLE TRAVELLING ON A TRACK

Abstract

A computer-implemented method for creating measurement data describing a railway network or a vehicle travelling on a track. Measurement values are detected by a sensor situated on the vehicle or on the track. A sensor processor in communication with the sensor requests a time value describing a time point and/or a position value describing a position and/or a measurement sample from a processor situated on the vehicle. The time value, the position value, and/or the measurement sample are transmitted from the processor to the sensor processor. The sensor processor assigns a measurement value to the time value, the position value, or the measurement sample, such that the time value, the position value, and/or the measurement sample describe a measurement time point or measurement position, and measurement data comprising the measurement value and the time value and/or the position value and/or the measurement sample are stored in a database.

Claims

1-10. (canceled)

11. A computer-implemented method of generating measurement data describing a railway network or a vehicle travelling on a track, comprising: determining measurement values using a sensor arranged on the vehicle or on the track; requesting, from a computation unit arranged on the vehicle by a sensor computation unit in communication with the sensor, at least one of: a time value describing a time t, a position value describing a position, and a measurement pattern, transmitting, to the sensor computation unit from the computation unit, the at least one of the time value, the position value, and the measurement pattern; assigning, by the sensor computation unit, a measurement value to the at least one of the time value, the position value, and the measurement pattern, the at least one of the time value, the position value, and the measurement pattern describing one of a measurement time and a measurement position; and storing, in a database, measurement data including the measurement value and the at least one of the time value, the position value, and the measurement pattern.

12. The method of claim 11, wherein the time value is generated according to a network protocol.

13. The method of claim 12, wherein the network protocol is one of NTP, PTP, and a specified format.

14. The method of claim 11, wherein the measurement data includes a unified time format.

15. The method of claim 11, wherein the position value is at least one of relative position information, absolute position information, and position information established based on applicable standards in the railway network.

16. The method of claim 11, wherein the measurement pattern includes information on a determination of measurement values per timespan.

17. The method of claim 11, wherein: the time value is matched with a reference time value at the time t; and a time difference between the time value and the reference time value is minimized by a change in time increments of the time value following the time t made within a matching timespan.

18. The method of claim 17, wherein the change in time increments is one of linear and following a mathematical function.

19. The method of claim 11, wherein: a sensor ID is read from the sensor by the sensor computation unit; and the sensor ID is transmitted to the computation unit.

20. The method of claim 11, further comprising at least one of: generating improved measurement values out of first measurement values and second measurement values, the improved measurement values being determined by one of: an average of the first measurement values and the second measurement values; and interpolation of the first measurement values and the second measurement values; and generating improved measurement data out of: first measurement data including first measurement values generated by a first sensor at a time specified by a unified time format, and second measurement data including second measurement values generated by a second sensor at the time specified by the unified time format; wherein the improved measurement data is determined by one of: an average of the first measurement data and the second measurement data; and interpolation of the first measurement data and the second measurement data.

21. The method of claim 11, further comprising matching the measurement data with reference measurement data, the reference measurement data describing a reference object, and the measurement data being assigned to the reference object while determining a similarity measure.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The invention is now additionally explained based on the following embodiments shown in the figures:

[0079] FIG. 1: illustrates an arrangement on a vehicle for performing the inventive method;

[0080] FIG. 2: illustrates another embodiment of the inventive method;

[0081] FIG. 3: illustrates another embodiment of the inventive method;

[0082] FIG. 4 illustrates another embodiment of the inventive method;

[0083] FIG. 5: illustrates another embodiment of the inventive method;

[0084] FIG. 6: illustrates another embodiment of the inventive method;

[0085] FIG. 7: illustrates the technical effect of the inventive method; and

[0086] FIG. 8: illustrates another embodiment of the inventive method.

DETAILED DESCRIPTION

[0087] The embodiments shown in the figures merely show possible embodiments, while it should be noted at this point that the invention is not limited to those specifically shown variant embodiments thereof, but that combinations of the individual variant embodiments among one another and with the above general description are also possible. Those further possible combinations need not be explicitly mentioned since such further possible combinations are within the ability of one of ordinary skill in the relevant art based on technical teaching by the present invention.

[0088] The scope of the invention is defined by the claims. However, the description and the drawings are to be used to interpret the claims. Individual features or combinations of features of the various embodiments shown and described may as such present independent inventive solutions. The object underlying such independent inventive solutions may be read from the description.

[0089] In the figures, the following elements are each designated by the preceding reference numerals: [0090] 1 computation unit [0091] 2 (not assigned) [0092] 3, 4, 5 sensors on the vehicle [0093] 6 sensor on the track [0094] 7 track [0095] 8, 9 graphs of measurement values of prior art [0096] 10, 11 graphs of measurement values of the inventive method [0097] 12, 13 reference graphs [0098] 14 timespan

[0099] The inventive method of generating measurement data is illustrated by a track processing machine. The measurement values processed in this method describe a railway network or a vehicle travelling on a track. The measurement values can be determined using one of the sensors 3, 4, 5 arranged on the vehicle or the sensors 6 arranged on the track 7.

[0100] In FIG. 1, sensors 3, 4, 5 are arranged on the vehicle by way of example. Sensors 3, 4, 5 on the vehicle may, for example, determine measurement values which are relevant for localizing the vehicle in a network of tracks. While sensors for determining of measurement values relevant for localizing the vehicle are mentioned in the following description, the sensors may also serve to determine measurement values for determining a relative or absolute position of part of the device, which measurement data enters into a control of a (partial) device on the vehicle. The inventive method is not limited to the localization of relevant measurement values or such sensors.

[0101] The inventive method can also be applied to sensors 6 arranged on track 7.

[0102] The inventive method can be applied to sensors of prior art for determining measurement values relevant in the railway industry.

[0103] A time value describing a time t is requested from a computation unit 1 arranged on the vehicle by a sensor computation unit (not shown in FIG. 1) in communication with the sensor. The single time value exclusively describing a time t in the inventive method is transmitted from computation unit 1 to the sensor computation unit. A measurement value generated by sensor 3, 4, 5 is assigned to the time value using the sensor computation unit. Furthermore, measurement data may be determined according to the description above.

[0104] Sensor 3 may be, for example, a sensor for determining the distance travelled by the vehicle. According to common science, the position of the vehicle in a network of tracks can be calculated based on the travelled distance.

[0105] Sensor 4 may be, for example, a sensor for determining gauge. In prior art, the temporal change in gauge may be an input parameter for localizing the vehicle.

[0106] Sensor 5 may be a GPS sensor which allows localizing the vehicle depending on the availability of the GPS signal.

[0107] According to common science, the position of the vehicle at a time t may be determined from the measurement values mentioned above by way of example and/or from the temporal development of said measurement values. This, however, requires that the individual measurement signals are in fact determined at the single time t. The inventive method achieves this by generating a single time value which in the inventive method exclusively describes time t. By generating the measurement values at a time exclusively described by the time value and by linking the measurement values to the single time value, a time difference in generating the measurement values is prevented.

[0108] FIG. 2 shows a possible embodiment of the inventive method. The essential components for performing the inventive method are also included in FIG. 2.

[0109] The arrangement of the components for performing the inventive method includes at least one sensor for determining the measurement values, a sensor computation unit and a computation unit. FIG. 2 merely shows a basic structure and basic arrangement of the components mentioned. The disclosure of the invention does not preclude that the components mentioned can be integrated into one or more electrotechnical members.

[0110] The measurement values describing a condition of the track or of the vehicle are determined with the sensor. The operator may employ sensors of prior art in performing the inventive method.

[0111] For example, the sensor may be arranged on the vehicle and determine measurement values describing the condition of the vehicle or of the railway network. Likewise, the sensor may be arranged on the track and, as a sensor there arranged, determine measurement values describing the condition of the railway network, in particular the railbed and track, or of the vehicle.

[0112] At least one measurement value and optionally one sensor time value are transmitted to the sensor computation unit. When the sensor is arranged on the vehicle, this data transfer is done preferably, but not limited to, via a wired network, since the sensor computation unit is also arranged on the vehicle. When the sensor is arranged on the track, this data transfer is done preferably via radio, as wired data transfer is not feasible. The forms and ways of data transfers employed, in particular the wired forms of data transfer, such as cable, switch, etc., are preferably standardised to prevent latencies between the data transfer paths. Generally, the skilled person is able to use their expert knowledge to establish a suitable way of data transfer and design it such that the established data transfer is subjected to little interference such as latency.

[0113] The inventive method allows the use of sensors which either deliver or do not deliver a sensor time value in addition to the measurement values. The property of the sensor to be able to deliver sensor time values is irrelevant to the inventive method.

[0114] The inventive method may include the method step of the sensor computation unit requesting the time value from the computation unit. In this request, the characteristic of the requested time value is defined, so that the time value is present in the time format required for further data processing. This is done according to the common teaching on defining the communication protocol, such as Network Time Protocol (NTP) or Precision Time Protocol (PTP), for example.

[0115] The time value is transmitted from the computation unit to the sensor computation unit while observing the required communication protocol. This transmitted time value is the only relevant time value in performing the inventive method. No other time values are processed in a relevant manner in performing the inventive method; such other time values, such as the sensor time values, for example, have no influence on subsequent method steps, in particular they have no relevant influence on the synchronization of the measurement values as achievable by the inventive method.

[0116] The computation unit may be controlled such that the computation unit delivers the time values in the communication protocol requested by the sensor computation unit. The characteristic of the time values may be adapted to the respective request.

[0117] The computation unit may also be controlled such that the computation unit delivers the time values in a rigid format.

[0118] The measurement values are linked to the time value in the sensor computation unit. The measurement data generated therefrom is transmitted to the computation unit and optionally stored in a database. The database is not depicted in FIG. 2.

[0119] The computation unit may include multiple units, which units have individual functions (pattern, database, . . . ) separately.

[0120] With reference to the above description on the occurrence of further time values, it shall be noted that the measurement data may generally include sensor time values. The sensor time values are, however, not of further relevance in the basic performing of the inventive method or subsequent methods. Such methods are executable without the sensor time values as further time values.

[0121] Since the further time values, such as sensor time values, have not relevant significance in performing the inventive methods, the sensor time values can be deleted or the measurement data designed such that the measurement data includes only measurement values and time values.

[0122] FIG. 2 shows an arrangement in which only one sensor 10 is coupled to sensor computation unit 1 and only one sensor 20 is coupled to sensor computation unit 2. This exemplary arrangement by no means excludes arrangements of a plurality of sensors with the respective sensor computation unit. In this sense, a plurality of sensor computation units may also be coupled to the computation unit, which sensor computation units are in turn coupled to one sensor or to multiple sensors. By way of example, FIG. 2 shows the coupling of two sensor computation units to the one computation unit.

[0123] According to common science, a sensor may require a time value in a different format, as is specifically established by the communication protocol. With reference to FIG. 2, processing the measurement values of sensor 10 may require a time value in NTP format, while processing the measurement values of sensor 20 requires a PTP time value. The time protocols referred to herein are to be regarded as merely exemplary. In general, processing the measurement values of a sensor may require a specified time format. The time value is generated in the requested format in the computation unit and transmitted to the sensor computation unit.

[0124] The measurement data may include a unified time format. Linking the measurement values to the single time value, which may be present in different time formats, allows storing the measurement data with a unified time format.

[0125] The method step of querying a time value may be omitted, as set out above in the overall description. It is also feasible in the context of the disclosure of the invention that the time value to which the measurement value is assigned is specified by the central computation unit. This specification may be a timing or measurement pattern, or a plurality of time values, for example.

[0126] A timing is understood to mean determining a number of measurement values within a timespan. In a measurement pattern, the number of measurement values within a timespan is indicated with the aid of a mathematical function.

[0127] FIG. 3 illustrates an embodiment of the inventive method; in which method the sensor computation unit does not query the time value.

[0128] The above description mentions the determination of measurement values per timespan.

[0129] Likewise, the determination of measurement values per distance can be defined. The determination of measurement values per distance may be defined by an odometer wheel or by another sensor as well as a system for positional determination.

[0130] FIG. 4 illustrates another embodiment of the inventive method. The feasibility of the method illustrated based on FIG. 4 is by no means limited to the linking of measurement values from optionally multiple sensors to a single time value (see FIG. 2) set out above. The method illustrated in FIG. 4 may be regarded as a possible advantageous effect of the method described above of linking measurement values from optionally several sensors to a single time value. The method described in FIG. 4 can also be executed as a stand-alone method.

[0131] The time value is generated by the computation unit. However, the time value issue by the computation unit may be different from a reference time such as atomic time, for example. The operator could solve the present problem of preventing a such time difference by matching the time value and the reference time value. Preferably, said matching is done permanently to thereby prevent the time difference from being created. However, such matching requires a data connection between the computation unit and the unit generating the reference time, the possibility of which is very limited in the railway industry. For example, not such data connection will be available in a tunnel.

[0132] FIG. 4 includes a plot where the time value specified by the computation unit is plotted on the x-axis and the reference time value is plotted on the y-axis. A graph which represents a time difference between the time value and the reference time value crosses zero and has a slope of 1:1 when there is no time difference.

[0133] The method illustrated in FIG. 4 can be divided into the following phases, wherein not all the phases are part of the inventive method.

[0134] In phase 1, reference time value and time value are a match. The graph (or its course) extends through zero and has a slope of 1:1.

[0135] In phase 2, there is a time difference between the reference time value and the time value. The graph (straight line) deviates from the dashed line, which illustrates reference time value and time value being a match. In the method shown in FIG. 4, the time value is increased in smaller time increments than the reference time value. The graph thus runs below the dashed line in phase 2. In phase 2, the graph runs at a smaller slope than the dashed line.

[0136] It is, for example, assumed that no data connection between the computation unit issuing the time value and the reference time issuing the reference time is present in phase 2, so that the time difference cannot be ascertained. In the transition from phase 2 to phase 3, the time difference can be ascertained. In FIG. 4, the time difference is illustrated by the distance of the graph from the dashed line.

[0137] In phase 3, the time value is adapted to the reference time slidingly by applying the inventive method. The sliding adaptation may be done by applying methods of common science. By way of example and not limitation, reference is made to method RFC 5905, which is described, for example, on http://www.ntp.org/. A phase 4 following phase 3 is characterized in that the only time value is a match with the reference time.

[0138] The advantageous effect of the inventive method is also set out as follows. Methods of prior art are based on linking the measurement values determined by individual sensors to the individual sensor time values. Each individual sensor time value may have a time difference to the reference time, so that individual time values of individual sensors have to adapted by an individual stand-alone method similar to the method described above. With a plurality of sensors and the occurrence of a plurality of time differences to different times, a plurality of adaptations of sensor time values to reference time values have to be performed, so that the skilled person can no longer follow the adaptations.

[0139] In contrast to this, a single time value has to be adapted to the reference time when applying the inventive method. In contrast to the prior-art method described above, the adaptation can be followed.

[0140] FIG. 5 and FIG. 6 illustrate methods of generating measurement values by a sensor, which measurement values are linked to a single time value specified by a central computation unit. Communication between the sensor computation unit and the sensor is represented via the time defined by the single time value. Communication between the sensor computation unit and the sensor via the position defined by the single position value would be similar.

[0141] The following is submitted on FIG. 5. As shown in FIG. 2 and described above, the sensor computation unit receives the single time value (or position value, respectively). The sensor computation unit issues the command to the sensor to determine a measurement value to said single time (or position value, respectively) and to transmit said measurement value to the sensor computation unit. In the sensor computation unit, the determined measurement value is linked to the single time value (or position value, respectively), optionally while generating measurement data. This method represented in FIG. 5 is based on the use of sensors with a non-specified timing for the determination of the measurement values.

[0142] FIG. 6 elucidates the use of sensors with a specified timing for the determination of measurement values. The sensor computation unit issues the command to generate the measurement value at a time T1. Since the timing and the time T1 are a match, the sensor can determine a measurement value at a time t1 equaling the time T1 and transmit it to the computation unit.

[0143] However, there is also a possibility of the timing and the time T2 not being a match. To still generate a measurement value at a time t2 equaling the time T2, at which time T2 the measurement value is to be determined by command, a measurement value notionally generated at a time t2 is transmitted to the computation unit. The measurement value notionally generated at the time t2 is a measurement value averaged or interpolated from multiple measurement values, which multiple measurement values are determined within a timespan including the time T2 equaling t2. The measurement values generated by way of example at a time t2.1 and t2.2, respectively, are included in FIG. 6, with the times t2.1 and t2.2 falling within the timespan.

[0144] Furthermore, measurement data including a measurement value and a time value may be generated from said measurement value and said time value.

[0145] FIG. 7 illustrates, among other things, the technical effect of the inventive method in the context of matching determined measurement values with reference measurement values.

[0146] FIG. 7 includes three plots. Plot 3 illustrates the temporal development of measurement values which are determined using methods of prior art. The time is plotted on the x-axis and the measurement values are plotted on the y-axis-a principle applied in plots 1 and 2 as well. Plot 3 includes two graphs 8, 9, which represent measurement values determined using methods of prior art as a function of time.

[0147] As set out above in detail, with methods of prior art, there is the technical problem of a real single time being described by different time values. This problem is illustrated in the plot by different temporal positions of times t1 of graphs 8, 9 on the x-axis as the time axis.

[0148] Plot 2 of FIG. 7 illustrates the temporal development of measurement values which are generated by applying the inventive method, in particular by assigning a single time value describing a time t1. The inventive method has the technical effect of a time t1 being described by a single time value and only this single time value being assigned to the measurement values. This is illustrated in plot 2 of FIG. 7 such that the times t1 of graphs 10, 11 are assigned at the same positions on the x-axis as time axis.

[0149] Plot 1 includes graphs 12, 13 as reference measurement values.

[0150] A possible technical effect of the description of a time t1 by multiple time values in prior art may be that graphs 8, 9 of a timespan 14 cannot be assigned to the reference graphs 12, 13 of said timespan. In the case that graphs 8, 9 relate to the temporal development of measurement values for describing a position, such as GPS signal, gauge, etc., for example, the vehicle would hardly or not at all be able to be localized, or only inaccurately. No single time period is found in the course of reference graphs 12, 13 in which graphs 8, 9 resemble both reference graphs 12, 13 with a high similarity measure.

[0151] Experiments have shown that graphs 10, 11 of the measurement values determined by applying the inventive methods can be matched with reference graphs 12, 13 at tolerable expense. In the case mentioned earlier that graphs 10, 11 describe measurement values for localization, the vehicle could be localized with sufficient accuracy.

[0152] Localization while matching graphs 8, 9 with reference graphs 12, 13 may be effectively done only if the time difference between the time t1 of graph 8 and the time t1 of graph 9 is minimized. The process of minimizing this time difference includes, for example, detecting this time difference. Said detection alone is subject to inaccuracy, which inaccuracy affects subsequent matchings.

[0153] The above description of FIG. 7 is targeted to the representation of measurement values or reference measurement values, respectively, as a function of time. A person of skill in the art is able to apply this description to a representation of measurement values as a function of position.

[0154] In addition to FIG. 2 and FIG. 3, reference is made to the embodiment of the inventive method shown in FIG. 8, which embodiment is basically similar to the embodiments shown in FIGS. 2 and 3.

[0155] Sensor 10, sensor computation unit 1 and computation unit may be formed as a single component in the sense of the disclosure of the invention. Sensor 10 transmits the measurement value and the sensor time value to the sensor computation unit 1. Sensor computation unit 1 transmits the measurement value in the form of measurement data and the sensor time value as a time value to the computation unit.

[0156] The central computation unit transmits the time value determined from the sensor time value to the sensor computation unit 2. This method step may be performed as the result of a query of a time value; said time value query is not necessarily required, as is explained above in sufficient detail.

[0157] The method shown in FIG. 8 differs from the methods described above in that the time value is generated based on the sensor time value.