 |
Description  |
|
|
BACKGROUND OF THE INVENTION
The present invention concerns a process for localizing a mobile unit on a
network, and especially a process that makes it possible to localize with
high accuracy the position of a mobile unit on a system of railroads, as
well as a system for implementing such a process.
In order to localize a mobile vehicle on a railroad track, a known process
consists of arranging detectors at specific points or sectors on the
network, for instance electrical, electromagnetic or electronic circuits,
placed on the track, and pedals that are probably associated as axle
counters cooperating with the vehicle, etc.
However, such equipment which is linked to the infrastructure is costly,
especially because of the high number of detectors required, and the
investment often seems unproductive on low traffic density lines.
Furthermore, knowledge of the position of the mobile unit is accurate only
when it is in well determined areas on the line. In the areas between the
well determined areas on the line, knowledge of localization is inaccurate
and can become a hamper to the development of specific applications.
Finally, knowledge of localization is locally available on the ground, but
not directly on the mobile unit proper. In some instances, the detector
tells the mobile unit that it is crossing at its level and sends it
instructions pertaining, for instance, to speed limit or stopping
distance. In those cases, for controlling speed or automatic piloting, the
mobile unit can implement a measurement of the distance traveled, but it
is not localization as such. Indeed, aside from the moment when the mobile
unit crosses above the detector, the ground based system does not know
exactly where the mobile unit is; whereas the mobile unit knows the
distance traveled from the detector, but does not know its position in
relation to the network as a whole. More importantly, the mobile unit it
does not know which track it is on, within a network where several tracks
might be contiguous.
Another known process consists of using a triangulation method with
beacons, that can be static or mobile, and for instance loaded on board
satellites placed in orbit.
This second process displays various inconveniences, aside from its high
cost. Indeed, the available accuracy does not make it possible to
distinguish, for certain, on which track of a network the mobile unit is
located. Localization requires the possibility of a link, by radio
usually, with the beacons, and irregularities in infrastructure relief,
especially tunnels, do not make it possible to obtain good coverage of the
network. Finally, at least in the case of radio-spotting, in other words
when the ground based system acquires the knowledge of localization of the
mobile unit, the ltter only obtains periodic knowledge of its position; it
has to resort to estimation for constant monitoring of the speed or
automatic piloting.
The patent FR 2 139 471 describes a control system for a vehicle that moves
on a track divided into consecutive sections, which includes information
on the distance traveled, thus making it possible to identify each section
when the vehicle passes by. The patent FR 2 274 094 describes a device for
localizing vehicles on a road network aimed especially at detecting
changes in direction at crossroads.
SUMMARY OF THE INVENTION
It is an object of the invention is to provide a mobile unit on a railroad
track network having accurate and constant knowledge of its position
without requiring costly infrastructure, and especially to enable the
identification of the track on which the mobile unit can be found, from
among several neighboring tracks.
It is another object of the invention to provide a process which makes it
possible to obtain the above result in a feasible and cost-effective way
by avoiding inconveniences that are inherent to the previously mentioned
known methods.
It is still a further object of the invention to provide also a
localization system for a mobile unit on a network that ensures accurate
and constant localization of the mobile unit, by making use of the current
infrastructure of the network, or inferring only the establishment of
simple and cost-effective markers.
The process according to the invention makes it possible to locate or
localize a mobile unit that operates on a system of railroads which
includes static or fixed markers in the vicinity of the track and is
distinguished in that a sensor is employed that is integral with the
mobile unit and which cooperates with the static markers so as to sense a
characteristic signal emitted or returned by each static marker. Further,
a database is included which contains the description of a graph of which
the knots or nodes are the static markers and the branches the tracks that
connect them. The lengths of the branches of the graph recorded in the
database are compared either with the distances measured been markers, or,
except for one factor, with the time separating the moments of detection
of the characteristic signals, in order to deduce therefrom the position
of the mobile unit.
According to an implementation form of the process of the invention, at
least one radiation or signal source is employed which is integral to the
mobile unit, such that during the motion of the mobile unit, the signal
source scans the zone where the static markers are located. Further, a
data base is employed which contains a graph of which the knots are the
static markers and the branches the tracks that connect them. Also at
least a receiver is employed which is integral to the mobile unit and is
sensitive to the retrotransmitted signal.
In conformance with the process of the invention, the sensor or receiver
cooperates with the static markers to we detect the crossing of the mobile
unit close to the static markers by the characteristic signal which they
emit or that they send back. Alternatively the digital signature of said
characteristic signal is compared with the signatures linked to the knots
of the graph that are recorded in the database, to deduce therefrom the
position of the mobile unit. In a further alternative the distance
traveled from the previous static is measured marker and and the compared
to the length of the branch(es) of the graph starting from the previous
knot, in the database.
According to a preferred embodiment, the identification of each static
marker is used for retiming the estimated position of the mobile unit on
the track. In conformance with another advantageous characteristic of the
process, the static markers are executed so that the retrotransmitted
response of the wave emitted by the radiation source is different
according to the marker or the marker category, and the processing of the
response makes it possible then to distinguish one marker from many other
different markers or belonging to a different category.
The echo or reflected signals that are received are instantly processed
directly on the mobile unit, by appropriate loaded means, or remotely, the
link of the processing and comparison unit with the database, being
performed by the usual means, for instance by radio.
For a radiation source, the use of radar makes it possible not only to
detect static markers, but to measure the distance traveled. According to
a variant of the invention, several beams can be emitted, for instance
radar beams, with several antennae each of which is connected to a
distinct radar or all are connected to the same radar. In particular two
radar beams can be emitted that are symmetrical in relation to a plane
which is perpendicular to the axis of the track, one to the front, the
other to the rear, in relation to the direction of the motion of the
mobile unit. If the first antenna emits a radiation which produces an a
angle with the plane of the track, the second antenna is fastened on the
mobile unit so as to form with the first an angle that equals 180.degree.
which is twice reduced from the nominal value of a, and thereupon the mean
of received echo signals is computed.
The device according to the present invention includes:
a sensor that is integral to the mobile unit, which cooperates with the
static markers;
a shaping circuit for the signal that comes from the sensor;
means for measuring the distance traveled or the travel time;
a database which contains the description of a graph of which the knots are
the static markers and the branches the tracks that connect them and,
a circuit that is controlled by the shaped signal, and connected to the
means for measuring distance or time and to the database, which determines
the position of the mobile unit.
According to an embodiment, the circuit that is controlled by the shaped
signal determines the potential branches of the graph based on the last
identified branch, and selects the one with a length that is closest to
the distance measured between the two last detected markers, the position
being defined by the selected branch and by the distance traveled on that
branch.
It is especially advantageous to use a circuit, controlled by the shaped
signal, which determines the possible branches based on the immediately
preceding branch and assigning them a probability as a function of that of
the upward branch of the anticipated branch and the measured distance,
only retaining the n best branches, n being greater than or equal to 2,
the position being defined by the most probable branch and the distance
traveled on that branch.
According to an advantageous embodiment, the sensor is a receiver which is
integral to the mobile unit, sensitive to the signal retrotransmitted by
the static markers and emitted by a radiation source that is integral to
the mobile unit which scans, during motion, the zone where the static
markers are located.
Obviously, the specific sensors, which are the radiation sources, the
particular signal shaping circuits, the particular circuits for assessing
the possible branches of the graph, the particular means for measuring
distance or time, and the particular databases do not figure in the
framework of this invention and are of a type that is known in the art.
For instance, the radiation source can be a radar, a laser, an ultrasound
emitter, etc., mounted on the mobile unit and associated with an
appropriate receiver. Static markers can also be employed that cooperate
with a sensor that is integral with the mobile unit, and for instance
points or magnets placed in the immediate vicinity of a rail of the
railroad track, that cooperate with a sensor wheel or with a magnetic
field sensor which is integral with the mobile unit, respectively.
According to a preferred embodiment, the system includes at least a radar
which is connected at least to an antenna directed at the track and
producing a set angle with it, the direction of the antenna enabling a
scan of the zone where the characteristic static markers are located in
the vicinity of the track, processing means for echo signals to measure
the distance traveled from the departure point and the signal level, as
well as the database and the circuit controlled by the signal, such as
they were described above.
According to a simple embodiment, the static markers can be comprised of
the contact line masts arranged along the track at a distance almost
constant to the latter, in the case of electrified tracks. Generally
speaking, it is also possible to use for markers all the singularities
that arise on the track or in its immediate vicinity, of which the
position will have been noted previously.
In conformance with another embodiment, the markers are comprised of
trirectangular trihedra or assemblies of trihedra, either metallic or
metallized, which send back an echo (for instance a radar echo) that is
easy to distinguish from that of the environment of the track (rail,
ballast, crossties and rail screws, for instance) by the simple measuring
of the level of the retrotransmitted signal. The marker can also be
embodied in the shape of an elbow-shaped pipe, with a diameter that is
greater than the wave length of the radar beam, and arranged in such a way
that the emitted signal penetrates through one end and the signal that
comes out the other end is picked up by the radar antenna. In that case,
the marker is recognizable as the polarizing of the signal, and the
receiving antenna is formed in such a way that it displays the same
polarization. Those markers are set far from one another along the track,
in the field that is scanned by the radar antenna.
A marker can be elementary or composite. An elementary marker is
characterized by the fact that a particular shape is not assigned to the
curve which expresses the level of the retrotransmitted signal in relation
to the path traveled by the sensor when it passes above the marker. A
composite marker is characterized by the fact that a specific shape is
assigned to that curve, either by the length of the zone above which a
substantial retrotransmitted signal is obtained, or by the fact that this
curve is comprised of the association of several curves corresponding to
an elementary marker, this association being characterized by the number
of elementary responses, by the distance that separates them or by a
combination of those factors. A composite marker can be embodied, for
instance, by a spider of metallic plates, the plates being perpendicular
to one another and to a metal plate used as a bottom to form cells, some
of the cells thus constituted being filled or covered by a substance that
absorbs the signal emitted by the radiation source. Another likely
embodiment is comprised of U-shaped pipes, the information being given by
the number of pipes and their distances to one another.
When the railroad track includes a switch that makes it possible to either
maintain the mobile unit on the direct track, or to deviate on to another
track, it is important to identify precisely the track on which the mobile
unit is located after passing by the switch. This assessment can be
performed in conformance with the invention either by placing a marker on
each branch of the switch at a different distance from the point, or by
using markers which supply different emitted or retrotransmitted signals
on both branches, or by not placing a marker on one of the branches. In
the first case, the difference in distances is preferably greater than
double the tolerated margin of error and the simple measuring of distance
between the identified marker and the last previously recognized marker
makes it possible to determine whether the mobile unit is on the direct
branch or on the deviated branch by way of a comparison with the graph of
the network which is recorded in the database.
BRIEF DESCRIPTION OF THE DRAWINGS
The characteristics and advantages of the invention will be provided in
greater detail in the following description relating to preferred
embodiment that is not restrictive, in reference to the attached drawings,
which depict:
FIG. 1 is a bloc-diagram of an installation in conformance with the
invention;
FIG. 2 is a diagrammatic view of a passive marker that can be used in the
invention, which is placed on the railroad track;
FIG. 3 is a perspective view of a marker variant;
FIG. 4 is a view of another passive marker which can be used in the
invention;
FIG. 5 is a diagram of a marker comprised of a magnet and an appropriate
detection circuit.
DETAILED DESCRIPTION OF THE INVENTION
In FIG. 1, a radar R includes an antenna A, located at a height that equals
from 60 cm above rail r and pointing towards the rail in a direction that
produces a preset angle a of about 30.degree. with the horizontal. The
radar R supplies at its output S1 information on distance traveled
extracted from the Doppler effect between the emitted signal and the
retrotransmitted signal. The radar also supplies at its output S2
information that gives the level of the retrotransmitted signal. The
outputs S1 and S2 are connected respectively to inputs E1 and E2 of a
processing computer OT. This computer OT possesses, in a database BD, a
description of the graph of passive markers RP placed on the track, in the
immediate vicinity ofthe rail and in the zone illustrated by the antenna A
during its passage above them. The computer OT also possesses an
input/output ESS series which allows for dialogue with the outside
(keyboard-monitor, modem, computer . . . ).
The signal shaping circuit stemming from the receiver section of the radar
(not shown) is of the usual kind and can be included indiscriminately in
the radar R or in the computer OT.
The operation is as follows: the processing computer OT receives at its
input/output ESS series the indication of its initial position marked in
relation to the graph that is contained in the database BD. When the
device (i.e., the mobile unit) on which the localization system is mounted
moves, the processing computer OT receives from radar R at its input E1
the signed value of elementary moves; that allows it to calculate the
estimated position.
When the processing computer OT receives from radar R at its input E2
information telling it that the level of the retrotransmitted signal
exceeds a preset threshold, the computer OT notes the estimated position.
If that position corresponds, except for a factor that is included within
the preset tolerance limits, to the position of a passive marker RP such
as it is indicated in the database BD, it is determined that the echo does
correspond to the marker and a correction is made for the estimated
position of the value found for the margin.
FIG. 2 shows an elementary marker (1) which is comprised of a metallic
trihedron while FIG. 3 shows a composite marker which includes an assembly
(2) of metallic trihedra that comprise a unit with several cells of which
some (3) are coated with a substance that absorbs the signl emitted by the
radar while others (4) are not. FIG. 4 shows markers (5) which are
comprised of U-shaped elbow pipes that are placed vertically close to the
track.
The use of composite markers makes it possible to distinguish each passive
marker RP with a recognizable "signature". We can envision that each
passive marker RP has a signature which makes it possible to distinguish
it from all the others or a signature that makes it possible to
distinguish it only from the passive markers in its vicinity with which
confusion might be likely.
In particular, he use of composite markers makes it possible to distinguish
the direct branch from the deviated branch of a switch without having to
rely on the difference between the distance that separates the previous
marker and the passive marker that is located in the direct branch, on the
one hand, and the distance separatingthe previous marker and the passive
marker which is located on the deviated branch, on the other hand.
Similarly, the signature of a passive marker can also be used to indicate
the traveling direction of the track.
The database BD can contain, aside from information which makes it possible
to know the position of the various passive markers, an indication of the
signature of the passive markers.
As previously stated, the radar R is located at a small height above the
rail. That height can vary and be included for instance between 30 cm and
1.50 m, and preferably between 40 and 60 cm. The angle a of the radar beam
with the plane of the track is preferably less than 45.degree. and for
instance can be close to 30.degree. in order to supply an accurate and
reliable measurement. The position of the radar antenna is preferably
above the rail, but it can vary in relation to the axis of the track,
especially if the possibility of snow covering the track is not a concern.
The radar R might especially be positioned above the middle of the track;
in that case, the passive markers RP should also be in the middle of the
track.
However, if it is preferred to place the radar R above a stretch of rails,
it can be convenient to place passive markers RP in the vicinity of each
stretch of rails. In that case, it can also be advantageous to install the
passive markers on each stretch of rails more or less perpendicular to one
another; it can also be convenient to assign them related signatures, for
instance identical to one another, or else identical, except for traffic
direction.
A concern of availability can lead to the placement of a radar above each
stretch of rails. In a case where the two radars are operating properly,
the comparison of information supplied by each of them at the output S1
provides another source of information that makes it possible, either to
retime the estimated position, or to verify its probability. Indeed,
except for the inaccuracy of the measurement of the path traveled, and for
the error introduced by a probable cant of the track, the difference
between the measurement of the path traveled on each of the two stretches
of rails provides, after dividing by the spacing of the track, the course
variation (in radians) from the source of the measurements. If the
direction of the track is included in the database BD in some locations,
then a possible coherence test is available. A variant consists of
clarifying in the database the beginning and the end of significant curves
and the difference in length of the stretches of rails between those two
points (which eliminates the influence of the cant). The difference in the
estimates of the traveled path from the outputs S1 of the two radars
allows, through a comparison with the information supplied by the database
BD, an interesting control of the validity of the position and of the good
operation of the processing. Practically speaking, it is possible to mark
with certainty only the curves of which the radius is less than the ratio
between the spacing of the two radars and twice their relative accuracy.
As shown in FIG. 5, a magnet (6) is fastened on the rail (r) by way of a
clamp (7) which maintains the magnet about 3 cm from the rail. The
fastening is accomplished with a conventional an usual means, for instance
by gluing, on the outer side of the rail head (8).
An electromagnetic detector (9) is fastened on the axle box (10) of the
wheel (11) so that the distance between the magnet (6), pointed
vertically, and the detector (9) is included between about 1 cm and 5 cm,
during the passage of the detector above the magnet. The detector (9) is
protected by a shield (not shown).
The magnet (6) is a permanent magnet of the usual kind, for instance of the
kind made of an aluminum, nickel and cobalt-based alloy that displays good
stability in the long run. The magnetic field thus created is detected by
the detector (9) which can include for instance a variable reluctance
coil, a Foucault current sensor, or a Hall effect sensor, according to a
known assembly (not shown).
While a single permanent magnet is adequate, it is also possible to mount
several magnets several centimeters from one another and to direct their
polarities in a preset way in order to constitute a signature that can be
identified by the detector.
In the foregoing description, it is presumed that the database BD is loaded
on board the mobile unit that is to be located or localized. It is obvious
that nothing fundamental is altered if that database is far away, for
instance if it is on the ground and there is a radio link between the
processing computer OT on board the mobile unit and the ground system that
manages that database. Neither is anything fundamental altered if the
database is on the ground but the excerpts pertaining to the region where
the mobile unit is located are sent to it, as need be, by a radio link or
localized beacons for instance. Another interesting use of radio signals
is the transmission of changes pertaining to the presence of passive
markers, their position or their signature, which have intervened since
the time to which the distance BD version corresponds and that in which
the mobile unit possesses. The invention obviously covers those different
variants.
* * * * *
|
|
 |
|
 |
Description |
|
