WO2020224862A1

WO2020224862A1 - Method and computer program product for position determination of rail-bound vehicles and position determination system and rail-bound vehicle, in particular railway vehicle

Info

Publication number: WO2020224862A1
Application number: PCT/EP2020/058314
Authority: WO
Inventors: Andreas Schönberger
Original assignee: Siemens Mobility GmbH
Priority date: 2019-05-03
Filing date: 2020-03-25
Publication date: 2020-11-12
Also published as: EP3947101A1; DE102019206347A1

Abstract

In order to be able to determine a vehicle position (FZP) of a rail-bound vehicle (FZ, SFZ) relative to a fixed position obtained by pre-locating and to do so with greater local accuracy, according to the invention the position of the vehicle is determined using a marking (MK, SZ) attached along a route (FST) of the rail-bound vehicle, and the marking position (MKP, SZP) of said marking, wherein - from a region of the route in which the marking is expected, a route image (SB) is captured from the perspective of the vehicle at a fixed vehicle position (FZFP) at a distance from from the marking position, - on the basis of the route image, at regular time intervals at a given vehicle speed, proceeding from the fixed vehicle position, it is iteratively evaluated for corresponding, differing and diminishing distances to the marking position how strongly an image section to be observed, which varies in relation to the marking position relative to the vehicle speed, deviates from a marking (MK', SZ') known according to the reference route images (RSB) with respect to the marking to be detected during the course of a combination of image calculation and image analysis for marking detection, carried out on the basis of the route image, the fixed vehicle position and the saved reference route images and additional data, - it is evaluated which image sections contain the smallest deviations and, according to said image section evaluation, the marking is detected by a consistent marking comparison and the vehicle position is thereby determined.

Description

description

Method and computer program product for position determination of track-bound vehicles and position determination system and track-bound vehicle, in particular rail vehicle

The invention relates to a method for determining the position of track-bound vehicles according to the preamble of claim 1, a computer program product for positioning track-bound vehicles according to the preamble of claim 5, a position determination system according to the preamble of claim 8 and a track-bound vehicle, in particular rail vehicle, according to the preamble of claim 16.

Global position data, e.g. by means of satellite-based position determination systems such as GPS, GALILEO or

GLONASS recorded and in the form of GPS, GALILEO or

GLONASS coordinates are available, play an important role in the information age with the central importance of information as raw material and commodity. So it is e.g. for current and future automatic driving systems with which "automatic driving" is implemented, driver assistance systems such as Navigation systems or exclusive information systems are of essential importance that the position data or coordinates should be as precise or exact as possible. However, current satellite-based receivers (e.g. GPS receivers) only have an accuracy of several meters.

Even with a Differential Global Positioning System (DGPS), in which correction data is broadcast to increase the accuracy of a GNSS navigation (Global Navigation Satellite System), the specified accuracy can usually only be set to 1-3 Improve meters. With the DGPS-based method, a correction of the determined position is achieved through the transmission of the correction data, but due to Due to the small number of stationary correction transmitters, no highly accurate, sufficiently comprehensive correction data can be generated.

However, better accuracy is often desired, ideally of just a few centimeters.

The object of the invention is to provide a method and a computer program product for determining the position of track-bound vehicles as well as a position-determining system and a track-bound vehicle, in particular a rail vehicle, with which the position of a vehicle is relative to a fixed position obtained by pre-positioning tion and locally can be determined more precisely.

This object is achieved on the basis of the method defined in the preamble of Pa tentanspruchs 1 for determining the position by the features specified in the characterizing part of claim 1.

In addition, the object is achieved on the basis of the in the Oberbe handle of claim 5 defined computer program product for position determination by the features specified in the characterizing part of claim 5.

In addition, based on the position determination system defined in the preamble of patent claim 8, this object is achieved by the features specified in the characterizing part of patent claim 8.

Furthermore, starting from the track-bound vehicle defined in the preamble of patent claim 16, the object is achieved by the features specified in the characterizing part of patent claim 16.

The idea on which the invention according to independent claims 1, 5, 8 and 16 is based consists in the relative, local determination of a position of a lane-bound vehicle stuff, in particular a rail vehicle, with the help of a marker placed along a route of the track-bound vehicle, in particular a signal sign, and its marker position or signal sign position by determining that

- From an area of the route in which the marking is to be expected, a route image is recorded from the perspective of the vehicle at a fixed vehicle position at a distance from the marking position,

- On the basis of the route map at regular time intervals at a given vehicle speed, starting from the fixed vehicle position for corresponding, different and decreasing distances to the marking position, it is iteratively evaluated how much each time in the course of this is based on the route map, the fixed position of the vehicle, of stored reference route images, route image metadata including the marking position and marking data as well as a combination of image calculation and image analysis for marking recognition carried out by stored route and vehicle data, which changes in relation to the marking position relative to the vehicle speed The image section deviates from a marking known from the reference route images with regard to the marking to be recognized,

- It is evaluated which image sections contain the smallest deviations, and according to this image section evaluation, the marking is recognized by a consistent comparison of marking and the vehicle position is thus determined.

The vehicle position is therefore determined as a result of a consistent comparison of markings in the course of image detail evaluation from the image detail position resulting therefrom.

It turns out to be advantageous in the sense of simple, inexpensive and economical as a starting point for the relative, local determination of the position of the track-bound vehicle if the vehicle's fixed position is satellite-supported (claims 2 and 10). Furthermore, given the effort involved, it is useful and also useful for further development for passenger and freight transport with track-bound vehicles, if the vehicle position determined in the course of the combination of image calculation and image analysis to support or implement Autono

men / automated driving is passed on to an autonomous / automated driving system AFS (claims 3, 6 and 11).

In this case, however, it is not only advantageous from a safety point of view, but also advisable to check the vehicle position determined in the course of the combination of image calculation and image analysis in the event of an inconsistent comparison of markings, in order to make a reliable, consistent decision on the vehicle position to be sent to the Autono

me / automated driving system (claims 4, 7 and 12).

Such safety checks are expedient, appropriate and sensible because it could well be that a rear light of a vehicle driving ahead can be confused with a red signal sign that can be recognized as a marker for determining the position. In this case, "No vehicle position" can be output as a result for safety purposes in order to remain on the safe side.

Further advantages of the invention emerge from the following description of an exemplary embodiment of the invention with reference to FIGS. 1 and 2. These show:

FIG. 1 position determination of a track-bound vehicle, in particular a rail vehicle, in ferry operation;

FIG. 2 shows a basic structure of a position determination system for the position determination based on the vehicle and driving operation according to FIG. FIG. 1 shows the position determination of a rail vehicle SFZ as a track-bound vehicle FZ in ferry operation, in which a vehicle position FZP of the vehicle FZ, SFZ relative to a fixed position obtained by pre-localization and locally can be determined more precisely. The rail vehicle SFZ, FZ - shown in FIGURE 1 is a motor vehicle TRW which is automated on a route FST according to a range of driver assistance support, via partial automation, further via conditional automation and high automation up to full automation with a vehicle speed "[v] "Moved - contains a position determination system PBS for position determination, which is located in a driver TFS with an integrated display device AZE for a vehicle driver FZF.

Starting from a pre-localization using satellite-based position determination methods, such as GPS, GALILEO or GLONASS, based on GPS, GALILEO and GLONASS coordinates determined vehicle fixed position FZFP is with an image acquisition unit BAE for capturing images of the position determination system PBS from the vehicle perspective, which are essentially the Perspective of the vehicle driver FZF ent speaks of a marking expected route area - ie a region of the route FST in which a marking MK spaced apart from the vehicle fixed position FZFP is at a marking position MKP, e.g. a signal sign SZ can be at a signal sign position SZP, is expected - a route image SB is acquired.

The image acquisition unit BAE of the position determination system PBS can be any device for acquiring and / or recording single or multiple images (videos), e.g. an image or video camera, a laser sensor, a thermal imaging camera, an infrared camera or a radar device.

FIG. 2 shows a basic structure of the position determination system PBS for the vehicle and according to FIG Driving-based position determination of the vehicle position FZP. For this purpose, the position determination system PBS contains, in addition to the image acquisition unit BAE already mentioned in connection with the description of FIG. 1 for capturing images, a position acquisition unit PAE for acquiring fixed vehicle positions and a control unit STE.

The position acquisition unit PAE is the unit of the position determination system PBS, which uses the GPS, GALILEO and GLONASS coordinates to determine the fixed vehicle position FZFP.

The control unit STE in turn contains a computer program product CPP for position determination, which one does not

volatile, readable memory SP, in which processor-readable control program commands of a program module PGM performing the position determination are stored, and a processor PZ connected to the memory SP, which executes the control program commands of the program module PGM for position determination.

For the improved and optimized acquisition of images, the image acquisition unit BAE contains a correction component KOK, which includes weather and brightness data in the evaluation of the captured image material, a focal length change component BVK which, depending on the distance to the marking MK, SZ, is the correct one Selects the recording angle so as to optimally support the multiple evaluation of the marking MK, SZ, and a lighting component BLK, which is preferably designed as a headlight and which works inside or outside the human-visible area.

The image acquisition unit BAE is also designed to be pivotable in an advantageous manner in order to be able to compensate for the angle of the image acquisition unit BAE to the marking MK, SZ. With regard to the safety relevance of the image acquisition unit BAE, it should be redundant in order to at least restrict operation in the event of damage, failure or soiling to enable. In addition, it would be conceivable to have two or more of these image acquisition units BAE work in parallel in order to increase the confidence of the data obtained.

The position acquisition unit PAE, the image acquisition unit BAE and the computer program product CPP Positionsbe ^¬ humor containing control unit STE form a common functional unit to the relative local position determination of the driving generating FZP. This functional unit is det such ausgebil ^¬ that in the image acquisition unit BAE according to the depicting ^¬ lung in FIGURE 1, and as in the description of he already imagines to the to the mark position MKP, SZP the Markie tion MK, SZ spaced vehicle-fixed position FZFP, from the vehicle perspective, the route image SB is acquired from the marking ^¬ expected route area.

Furthermore, the functional unit is designed such that in the control unit STE or in the processor PZ of the computer program product CPP

- on the basis of distance image SB at regular Zeitab ^¬ stands to-ti, ... t _x, such as shown in a time interval to-ti of 200ms (to-ti = 200 ms), as in FIG 1 wherein gegebe ^¬ NEN vehicle speed "[v]" starting from the vehicle fixed position FZFP for corresponding thereto different and decreasing distances so, si, ... s _x to Markierungspositi ^¬ on MKP, SZP - carried out an iterative evaluation (cf. FIG. 1) becomes. In this case, depending ^¬ weils in the course of this in the control unit STE or in the processor PZ of the computer program product CPP

- on the basis of the image acquisition unit BAE receive ^¬ NEN distance image SB, the vehicle-fixed position FZFP obtained from the position acquisition unit PAE, stored in a first database DB1 and from that obtained Refe rence Range images RSB, distance image metadata SBMD including the marking position MKP, SZP and marking data MKD as well as route and vehicle data SFZD stored in and received from a second database DB2 - The combination of image calculation and image analysis for marking recognition carried out evaluates the extent to which an image section of the route image SB that is to be viewed changes in relation to the marking position MKP, SZP relative to the vehicle speed with respect to the marking MK, SZ of one according to the reference route images RSB known marking MK ', SZ' deviates.

The route image metadata SBMD preferably have for the comparison between the reference route images and the route image or the image calculated therefrom of the combination of image calculation and image analysis carried out by an expert for the combination of image calculation and image analysis, with possibly further metadata and Calibration images in order to recognize the marking MK, SZ. The expert-marked image material is used to determine the relevant image section as precisely as possible and to be able to differentiate between relevant and irrelevant markings (e.g. a secondary route).

If necessary, expert markings can also be replaced by standard information, e.g. Standard information on a so-called pre-marking.

In the combination of image calculation and image analysis carried out in the control unit STE or in the processor PZ, any distortions of the route image SB, if the stored reference route images RSB were not recorded at exactly the same point as the route image SB recorded during ferry operation, are detected by positions immediately considered.

For the combination of image calculation and image analysis carried out in the control unit STE or in the processor PZ, the control unit STE or the processor PZ accesses the first database DB1 and / or the second database DB2. The first database DB1 and / or the second database DB2 are Either part of the position determination system PBS (option "A" for DB1 or option "C" for DB2) or assigned to the position determination system PBS for these accesses (option "B" for DB1 or option "D" for DB2).

The first database DB1 contains the reference route images RSB including any calibration images, the route image metadata SBMD, such as With regard to the route images, the exact position of their recording including information about the route or the track, possibly the angle of the recording and the MKD marking data including metadata such as the type of marking or signal character. These data can preferably be recorded as follows:

Initially statically in test drives or through targeted recordings by recording staff. And then in a dynamic expansion, in which the image material in the first database DB1 is regularly supplemented by the route images SB captured during the journeys.

The second database DB2 contains the route and vehicle data SFZD, which e.g. the exact data about the route FST, e.g. the exact position of the track, the position of the marking MK, MK 'or the signal sign SZ, SZ' the Monta geposition of the position acquisition unit PAE or des

GPS / GALILEO / GLONASS receivers, the mounting positions and resolutions of the image acquisition unit BAE, etc. in the vehicle FZ, SFZ.

In addition, the functional unit is designed in such a way that in the control unit STE or in the processor PZ of the computer program product CPP

it is evaluated which image sections contain the smallest deviations, and according to this image section evaluation the marking MK, SZ is recognized by a consistent comparison of the markings MK, SZ, MK ', SZ', whereby due to a total time "[t]" { [t] = [to-t _x ] in FIG. 1} the time intervals and the vehicle speed "[v]" according to FIG physical "[s] = [v] x [t]" relationship given route "[s]" {[s] = [s ₀ -s _x ] in FIGURE 1} between the known vehicle fixed position FZFP and the position the recognized mark MK, SZ is given the position of the recognized mark MK, SZ and thus the vehicle position FZP is determined.

The vehicle position is therefore determined as a result of a consistent comparison of markings in the course of the image detail evaluation from the image detail position resulting here.

Finally, the functional unit is advantageously designed such that the control unit STE or the processor PZ of the computer program product CPP

the vehicle position FZP determined in the course of the combination of image calculation and image analysis is passed on to an autonomous / automated driving system AFS to support or implement autonomous / automated driving.

The control unit STE or the processor PZ is also designed in such a way that the vehicle position FZP determined in the course of the combination of image calculation and image analysis is checked in the event of an inconsistent marking comparison in order to make a reliable, consistent decision regarding the vehicle position FZP to be determined to be transferred to the Autonomous / Automated Driving System AFS.

The autonomous / automated driving system AFS is when, as shown in FIG. 1, the railcar TRW of the track-bound vehicle FZ, SFZ moves automatically on the route FST, as the position determination system PBS in FIG the driver's cab TFS of the track-bound vehicle FZ, SFZ.

Claims

1.Method for determining the position of a track-bound vehicle (FZ), in particular a rail vehicle (SFZ), in which the track-bound vehicle (FZ, SFZ) is to drive along a route (FZ, SFZ), in particular if it is to drive automatically on the route (FST), a marking (MK) is attached to a marking position (MKP), in particular a signal sign (SZ) at a signal sign position (SZP),

characterized in that

for the relative, local determination of a vehicle position (FZP)

a) at a vehicle fixed position (FZFP) spaced apart from the marking position (MKP, SZP) of the marking (MK, SZ), from the vehicle perspective a route image (SB) of a marking expected route area - i.e. an area of the route (FST) in which the marking (MK, SZ) is to be expected - is acquired,

b) on the basis of the route map (SB) at regular time intervals, in particular of 200ms, at a given vehicle speed starting from the vehicle fixed position (FZFP) for corresponding, different and smaller distances to the marking position (MKP, SZP) iteratively It is assessed how strongly in each case in the course of one on the basis of the route image (SB), the vehicle fixed position (FZFP), stored reference route images (RSB), route image metadata (SBMD) including the marker position (MKP, SZP ) and marking data (MKD) as well as from stored route and vehicle data (SFZD) carried out a combination of image calculation and image analysis for marking recognition an image section to be viewed that changes in relation to the marking position (MKP, SZP) relative to the vehicle speed with regard to the recognizing marking (MK, SZ) from a marking (MK ', S) known according to the reference route images (RSB) Z ') deviates from c) it is assessed which image sections contain the smallest deviations, and according to this image section evaluation the marking (MK, SZ) is recognized by a consistent comparison of the markings (MK, SZ, MK ', SZ') where because of a by a total time "[t]" of the time intervals and a vehicle speed "[v]" according to the physical "[s] = [v] x [t]" relationship given route "[s]" between the known vehicle Fixed position (FZFP) and the position of the recognized marking (MK, SZ) the position of the recognized marking (MK, SZ) is given and thus the vehicle position (FZP) is determined.

2. The method according to claim 1, characterized in that

the vehicle's fixed position (FZFP) is determined with the aid of satellites.

3. The method according to claim 1 or 2, characterized in that

the vehicle position (FZP) determined in the course of the combination of image calculation and image analysis is passed on to an autonomous / automated driving system (AFS) to support or implement autonomous / automated driving.

4. The method according to claim 3, characterized in that

The vehicle position (FZP) determined in the course of the combination of image calculation and image analysis is checked in the event of an inconsistent comparison of markings in order to pass a reliable, consistent decision on the vehicle position (FZP) to be determined to the autonomous / automated driving system (AFS).

5. Computer program product (CPP) for determining the position of a track-bound vehicle (FZ), in particular a rail vehicle (SFZ), with a non-volatile, readable memory (SP) in which processor-readable control program commands of a program module performing the position determination (PGM) are stored, and a processor (PZ) verbun with the memory (SP), which executes the control program commands of the program module (PGM) for position determination, whereby for position determination along a route (FST) of the lane-bound vehicle ( FZ, SFZ), especially if this is to drive automatically on the route (FST), a marker (MK) is attached to a marker position (MKP), in particular a signal sign (SZ) to a signal sign position (SZP),

characterized in that

the processor (PZ) and the program module (PGM) for the relative, local determination of a vehicle position (FZP) are designed and the processor (PZ) executes the control program commands of the program module (PGM) in such a way that a ) on the basis of a vehicle fixed position (FZFP) spaced apart from the marking position (MKP, SZP) of the marking (MK, SZ), from the vehicle perspective of a marking expected route area - ie an area of the route (FST) in which the marking (MK, SZ) is to be expected - acquired and the processor (PZ) supplied route image (SB) at regular time intervals, in particular of 200ms, at a given vehicle speed starting from the vehicle fixed position (FZFP) for corresponding, different and decreasing distances to the marking position (MKP, SZP) are iteratively evaluated, how strong in each case in the course of a related to this on the basis of the route image (SB), the vehicle fixed position ( FZFP), from stored reference route images (RSB), route image metadata (SBMD) including the marking position (MKP, SZP) and marking data (MKD) as well as from stored route and vehicle data (SFZD), a combination of image calculation and image analysis For marking recognition, an image excerpt to be observed that changes in relation to the marking position (MKP, SZP) relative to the vehicle speed with regard to the marking to be recognized (MK, SZ) from a marking (MK ', known according to the reference route images (RSB)) SZ ') deviates, b) it is assessed which image sections contain the smallest deviations, and according to this image section evaluation the marking (MK, SZ) is recognized by a consistent comparison of the markings (MK, SZ, MK ', SZ') where because of a by a total time "[t]" of the time intervals and a vehicle speed "[v]" according to the physical "[s] = [v] x [t]" relationship given route "[s]" between the known vehicle Fixed position (FZFP) and the position of the recognized marking (MK, SZ) the position of the recognized marking (MK, SZ) is given and thus the vehicle position (FZP) is determined.

6. Computer program product (CPP) according to claim 5,

characterized in that

the processor (PZ) and the program module (PGM) are formed in such a way that

7. Computer program product (CPP) according to claim 6,

characterized in that

the processor (PZ) and the program module (PGM) are formed in such a way that

8. Position determination system (PBS) for a track-bound vehicle (FZ), in particular for a rail vehicle (SFZ), in which the track-bound vehicle (FZ, SFZ) runs along a route (FZ, SFZ), especially if it is automated on the route (FST) should drive, a marker (MK) at a marker position (MKP), in particular a signal Sign (SZ) is attached to a signal sign position (SZP),

marked by

a position acquisition unit (PAE) for the acquisition of vehicle fixed positions, an image acquisition unit (BAE) for capturing images and a control unit (STE) that uses a computer program product (CPP) for position determination with a non-volatile, Readable memory (SP) in which processor-readable control program commands of a program module (PGM) performing the position determination are stored, and a processor (PZ) connected to the memory (SP), which carries out the control program commands of the program module (PGM) Carries out position determination, the position acquisition unit (PAE), the image acquisition unit (BAE) and the control unit (STE) forming a common functional unit for the relative, local determination of a vehicle position (FZP) such that

a) in the image acquisition unit (BAE) at a vehicle fixed position (FZFP) at a distance from the marking position (MKP, SZP) of the marking (MK, SZ), from the vehicle perspective a route image (SB) of a marking expectation route range - ie an area of the route (FST) in which the marking (MK, SZ) is expected - is acquired,

b) in the control unit (STE) on the basis of the route image (SB) at regular time intervals, in particular of 200ms, at a given vehicle speed starting from the vehicle fixed position (FZFP) for corresponding, different and decreasing distances to the marking position ( MKP, SZP) iteratively evaluates how strongly the vehicle's fixed position is in each case in the course of a corresponding in the control unit (STE) on the basis of the route map (SB)

(FZFP), from reference route images (RSB) stored in a first database (DB1), route image metadata (SBMD) including the marker position (MKP, SZP) and marker data (MKD) as well as from a second database (DB2) stored route and vehicle data (SFZD) carried out combination of image calculation and image analysis for marking ungsrecognition to be considered, in relation to the marking position (MKP, SZP) relative to the vehicle speed changing image section with respect to the marking to be recognized (MK, SZ) from a marking (MK ', SZ) known according to the reference route images (RSB) ') deviates from

c) it is evaluated in the control unit (STE) which image sections contain the smallest deviations, and according to this image section evaluation the marking (MK, SZ) is recognized by a consistent comparison of the markings (MK, SZ, MK ', SZ'), where because of a distance "[s]" given by a total time "[t]" of the time intervals and a vehicle speed "[v]" according to the physical "[s] = [v] x [t]" relationship, between the known vehicle -Fixed position (FZFP) and the position of the recognized marking (MK, SZ) the position of the recognized marking (MK, SZ) is given and thus the vehicle position (FZP) is determined.

9. Position determination system (PBS) according to claim 8,

characterized in that

the first database (DB1) and / or the second database (DB2) are / is either part of the positioning system (PBS) or the positioning system (PBS) for access.

10. Position determination system (PBS) according to claim 8 or 9, characterized in that

the position acquisition unit (PAE) is designed in such a way that the vehicle fixed position (FZFP) is acquired with satellite support.

11. Position determination system (PBS) according to claim 8, 9 or

10, characterized in that

the control unit (STE) is designed in such a way that the vehicle position (FZP) determined in the course of the combination of image calculation and image analysis is passed on to an autonomous / automated driving system (AFS) to support or implement autonomous / automated driving.

12. Position determination system (PBS) according to claim 11, characterized in that

the control unit (STE) is designed in such a way that the vehicle position (FZP) determined in the course of the combination of image calculation and image analysis is checked in the event of an inconsistent marking comparison in order to make a reliable, consistent decision on the vehicle position to be determined (FZP) to the autonomous / Automated Driving System (AFS) to hand over.

13. Position determination system (PBS) according to one of claims 8 to 12, characterized in that

the image acquisition unit (BAE) has a correction component (KOK) which includes weather and brightness data in the evaluation of the image material.

14. Position determination system (PBS) according to one of claims 8 to 13, characterized in that

the image acquisition unit (BAE) has a focal length change component (BVK) that selects the correct recording angle depending on the distance to the marking (MK, SZ) in order to optimally support the multiple evaluation of the marking (MK, SZ).

15. Position determination system (PBS) according to one of claims 8 to 14, characterized in that

the image acquisition unit (BAE) has a lighting component (BLK), in particular a headlight, which works inside or outside the human-visible area.

16. Track-based vehicles (FZ), in particular rail vehicles (SFZ), in particular with an autonomous / automated driving system (AFS) to support or implement autonomous / automated driving,

marked by a position determination system (PBS) according to one of claims 8 to 15 for carrying out the method according to one of claims 1 to 4.