Abstract: The overall field of the invention is that of systems for detecting the posture of a moving object in space. The device according to the invention comprises: a fixed electrooptical device (Ot) comprising a point emission source (S), a telecentric lens comprising a projection . lens (L), a reception lens ( L") and a semireflective optical element (LSR) and a photosensitive matrix-type sensor ( D), the point emission source being arranged at the common focal point of the projection lens and of the reception lens by reflection or transmission through the semireflective optical element, and; an assembly comprising at least three retroreflector devices of the "corner cube" type (Cl, C2, C3) which are arranged on the moving object. The main application of this device is the detection of the orientation of an aircraft pilot helmet.
Optical system for measuring the orientation of a helmet using corner
cubes and a telecentric emission lens
The field of the invention is that of optical devices that can be used
for contactlessly measuring the orientation of an object in space. There are
various possible fields of application but the main application is detecting the
posture of an aircraft pilot helmet so that an image can thus be projected
5 onto his visor so that it is exactly superposed on the external landscape or so
that various systems of the aircraft can be subjugated to his gaze. The
precision sought for such systems is of the order of one milliradian.
There are various optical techniques that can be used for
10 measuring orientation on a helmet. In general , noticeable elements are
installed on the helmet and are pinpointed by a, system of cameras. The
position of the images of these noticeable elements makes it possible
through calculation to determine the orientation of the helmet.
These elements may be passive or active. Passive elements are
15 illuminated by an external source. For this purpose , retroreflective corner
cubes can be used as these make it possible to reduce the problems of
parasitic light caused by solar illumination . All that is required is for the optical
emission and reception members to be arranged on the same axis.
Active elements are generally light-emitting diodes. The cameras
20 have a fixed focus and therefore a depth of field which is of necessity limited.
This technique has a certain number of disadvantages . 1 he quality
of the image of each point imaged on the detector is dependent on the
position of the helmet and on its orientation, thus limiting the precision of the
system if a significant measurement volume or a substantial range of rotation
25 is to be covered.
The system according to the invention overcomes these two
disadvantages. It essentially comprises, mounted on a fixed frame of known
orientation, a single optical device of the telecentric type emitting and
30 receiving beams of parallel light.; The beams emitted come from a point
source, the beams received are the result of the retroreflection of light from
the source by retroreflectors mounted on the moving object the orientation of
which is to be determined.
It may be demonstrated that, with this detection system, the quality
of the measurement is, by construction, independent of the orientation of the
5 helmet. Further, its other advantages are as follows:
- a very simple algorithm for determining the orientation;
- the possibility to adapt the direction of illumination to the position of
the helmet;
- great insensitivity to solar illumination;
10 - use of entirely passive devices mounted on the helmet so that no
connecting cables or electrical power supply cables are required.
More specifically, the subject of, the invention is a system for
detecting the posture of a moving object in space, comprising a fixed
15 electrooptical device of known orientation comprising at least one first point
emission source and a photosensitive matrix-type sensor and an assembly
comprising three retroreflector devices, for example, of the "corner cube"
type, arranged on the moving object,
characterized in that the fixed electrooptical device comprises a
20 telecentric lens essentially comprising a projection lens, a reception lens and
a semireflective optical element which are arranged in such a way that:
the first point emission source is arranged at the focal point of the
projection lens by reflection or transmission through the semireflective optical
element,
25 the image of the first point emission source is arranged at the focal
point of the reception lens by transmission or reflection through the
semireflective optical element.
Advantageously, the first point emission source or the image
thereof is arranged on the optical axis common to the projection lens and to
30 the reception lens.
Advantageously, the system comprises a second point emission
source, the second point emission source or the image thereof being
arranged off the optical axis corr.mon to the projection lens and to the
reception lens.
Advantageously, in a first alternative form, with the first source
emitting in a first spectral band, the second source emits in a second spectral
band which is different from the first spectral band of the first source. In a
second alternative form, with the first source emitting light in a first
5 predetermined state of polarization, the second source emits light in a
second predetermined state of polarization which is different from the first
predetermined state of polarization. In a third alternative form, the first
emission source and the second emission source emit at different moments
in time.
10 Advantageously, the fixed device comprises a matrix of point
emission sources.
Advantageously, the moving object comprises at least four
retroreflector devices of the "corner cube" type. Each of the retroreflector
devices of the corner cube type arranged on the moving object comprises
15 optical or geometric discrimination means which are different from those of
the other corner cubes. In a first, embodiment, each corner cube comprises a
mask of a shape that is different from that of the two other corner cubes. In a
second embodiment, each corner cube comprises an optical filter the
transmission spectral band of which is different from that of the two other
20 filters of the two other corner cubes.
For preference, the moving object is a pilot helmet.
The invention will be better understood and other advantages will
become apparent from reading the following description given by way of
25 nonlimiting example and by studying the attached figures among which:
Figure 1 depicts a first embodiment of the detection system
according to the invention, comprising a single point source;
Figure 2 depicts a second embodiment of the detection system
according to the invention comprising two distinct point sources;
30 Figure 3 depicts a third embodiment of the detection system
according to the invention comprising a matrix of point sources.
By way of a first example of how the invention is embodied,
Figure 1 depicts a first embodiment of the detection system according to the
35 invention in the simplest case, which means to say in the case comprising a
single point source and in the context of the detection of the orientation of a
pilot helmet. This first configuration can very easily be adapted to suit other
applications.
The system essentially comprises two subassemblies , a fixed
5 electrooptical device and a helmet the orientation of which is to be
determined . It is referenced in a frame of reference (0, x, y, z).
The fixed electrooptical device is situated in an aircraft cockpit and
occupies a known orientation with respect to the frame of reference of the
aircraft.
10 The electrooptical device comprises an almost point source S of
light. This source maybe a light-emitting diode or a laser diode.
It also comprises a telecentric optical system Ot comprising a
projection lens L, a reception lens L' and a semireflective optical element LSR.
The projection lens L, like the reception lens L', may be made up either of
15 single lenses or of groups of individual lenses . The semireflective optical
element may be either a treated simple flat sheet as depicted in the various
figures, or a cube splitter . The assembly comprised of the projection and
reception lenses constitutes an afocal system , which means that their focal
point is common.
20 The image of the source S is arranged at the focal point of the
projection lens L by reflection off the semireflective optical element LSR. As a
result, the image of S is collimated at infinity by the lens L which thus emits a
beam of parallel light in a direction x as indicated in Figure 1. It makes no
difference whether the semireflective optical element LSR is used in reflection
25 on the emission path and in transmission on the reception path, or vice
versa.
The helmet H of Figure 1 is equipped with at least three retro®
reflectors C of the corner cube type. It is known that these optical elements
have the property of reflecting light in its direction of incidence. As a result,
30 each of the corner cubes C will return a pencil beam of light towards the lens
L.
These beams are all mutually parallel. Each of these beams
passes through the optical ass(: rnbly consisting of the projection and
reception lenses L and L' and the semireflective sheet LSR. The beams, on
35 exiting the lens L', are once again mutually parallel and fall onto a matrix-type
detector D which therefore picks up the image of each reflector C. The
detector D is, for example, a matrix of the CCD (Charge Coupled Device)
type. Only two of the three beams have been depicted in Figure 1 for the
sake of keeping the figure clear. These come from the corner cubes C1 and
5 C2. It may be demonstrated that the direction of the central ray of each
reflected beam always passes through the vertex SC of a reflector C
whatever the orientation of the reflector with respect to the illuminating beam.
The centre P of the spot of light projected onto the detector D therefore
always indicates the direction of the vertex SC of the corner cube with
10 respect to the detector.
The respective images of the vertices SC1 and SC2 of the corner
cubes C1 and C2 are therefore situated at P1 and P2 on the detector.
The positions of the points P1 and P2 on the detector D are not
dependent on the abscissa values x of C1 and C2.
15 The major benefit of this optical setup is that the deviation
between the positions of P1 and P2 is not dependent on translational
movements of the helmet H but dependent only on the orientation of said
helmet. Knowing the length d12, which is the distance separating the vertices
SCI and SC2 of the corner cubes C1 and C2, the unknown orientation of the
20 axis C1 C2 is thus completely determined, give or take the sign, by the
relative position of P2 with respect to P1 on the detector D.
More specifically, the helmet H is equipped with three corner
cubes C1, C2 and C3 with vertices SCI, SC2 and SC3. The distances d12
separating the vertices SCI and SC2 and d13 separating the vertices SCI
25 and SC3 are known. It is possible, for example and for the sake of
simplification, to choose for the axis C1 C2 to be perpendicular to C1 C3. The
orientation can still be determined if the axes CIC2 and CIC3 make a
different angle between them, but the calculation is just a little more
complicated.
30 The front face of each reflector is equipped with an optical or
geometric discrimination device which is different from that of the two others.
By way of a first example, the reflectors can be discriminated in
terms of their shape. The outline of:the cross section of each reflector is then
customized using a mask of a particular shape, a circle or a diamond for
35 example, that partially blocks off the incident and reflected beams.
By way of a second example, the reflectors can be discriminated
in terms of their colour. A red, green, blue or yellow coloured filter is placed in
front of each of the reflectors. In such a case it is, of course, necessary to
use a broad-spectrum source and a polychromatic detector or several
5 detectors separated by dichroic filters, each detector being dedicated to a
particular spectral band. It should be noted that the word "colour" is not
necessarily limited to the visible spectrum. It is also possible to use two
different spectral bands situated in the near infrared or in the near ultraviolet.
It is of course possible to combine the two methods of
10 discrimination: shape and colour. Thus, a first reflector would comprise a red
circular mask, a second reflector would comprise a green circular mask and a
third reflector would comprise a red mask or green mask in the shape of a
diamond.
Through this means the points P1, P2 and P3 on the detector
15 can be assigned to the corresponding vertices without the risk of error.
On the detector D, the coordinates measured in the fixed frame of
reference (0, x, y, z) of the points P1, P2 and P3, which are the images of
the vertices SCI, SC2 and SC3, are as follows:
P1(yl, z1), P2(y2, z2) and P3(y3, z3)
20 For an afocal system, for example of unit magnification, the
projection parallel to the axis x onto the vertical plane D connects the
unknown components (x12, y12, z12) of the vector CIC2 of known length
d12 to the coordinates of P1 and P2 by the relationships:
y12 = (y1 -y2)
25 z12 = (z1-z2)
x12 = E [d12 2_(Yl -y2)2_(Zl -z2)2]0,5 with E = +/-1
Likewise, the components (x31, y31, z31) of the vector C I C3 are:
y13 = (y1-y3)
z13 = (z1-z3)
30 x13 = E' [d132-(y1-y3)2-(z1 -z3)2]o. 5 with E' = +/-1
The indeterminacy of the values of E and of E' is partially resolved
by the following relationship:
CI C2 is perpendicular to C1 C3, so:
x12.x13 + y12.y13 + z12.z13 = 0, which also means that:
35 sign of (x12.x13) = E.E' = - sign of (y12.y13 + z12.z13);
So there are now only two solutions which are symmetric about
the vertical plane (y, z).
There are various techniques that can be used to resolve this last
indeterminacy. By way of first example, it is possible to add a fourth corner
5 cube C4 the vertex of which is not coplanar with that of the three others. The
four corner cubes therefore form a tetrahedron.
By way of a second example illustrated in Figure 2, the remaining
indeterminacy can also be resolved by adding a second source S' in the focal
plane of the lens which, by defining a second direction of projection,
10 generates three other images P'1, P'2 and P'3 on the detector D or on a
second detector D'. In Figure 2, for tl 'e sake of clarity, the semireflective
sheet LSR has not been depicted. The rays of light from this source S' are
depicted in dotted line.
The source S' is, for example, on the same vertical as the source
15 S, the corresponding oblique axis of projection is parallel to the vertical plane
(x, z), its orientation 0 is given as a function of the focal distance f of L and L'
by the conventional relationship tanO ® S'S/f
For P'1 and P'2, we have the relationship: z12' = (z'1 -z'2) +
x12.tan0
20 For P1 and P2, we have the relationship: z12 ® (z1-z2)
Hence, x12 is given by the unambiguous equality:
x12 = [(z1-z2) - (z'1-z'2)]/tan0
In order to isolate the images of the source S' from those of the
source S, the source, S' may, for example, be centred on another wZ velength,
25 another polarization, or be activated in alternation with S.
In the first instance, each source S and S' radiates in a determined
colour. A coloured filter is therefore positioned in front of each pixel of the
detector D or use is made of two detectors combined using dichroic mirrors.
In such an instance, it is preferable for the reflectors to be discriminated in
30 terms of shape.
In the second instance, the sources are discriminated by
polarization. The two sources of the same colour therefore radiate either in
two crossed directions of linear pn; irization or in two opposed directions of
circular polarization. Use is then made of two detectors which are combined
by a polarization splitter. The corner cubes are metallized in order to
conserve the incident polarization.
In the last instance, the two sources are activated alternately and
images are analysed on the detector separately for two successive images.
5 Once this indeterminacy of sign has been resolved using one of
the two methods described hereinabove, the device makes it possible,
unambiguously, to determine the components of the vectors C1C2 and
C1 C3, which are fixed on the helmet, and therefore the orientation of the
helmet in space.
10 By comparison, a system employing central projection comprising
a fixed focus camera forms the image' P1 of the front face of C1 or of an
equivalent diode on the plane of the detector D for just one single helmet
position. Further, for this helmet position, the image of the front face of C2 is
on the detector only for a particular orientation of C1 C2 and therefore only for
15 particular helmet orientations.
The projection lens L needs to have a sufficient aperture diameter
that it can cover the entire field of movement of the helmet. In order to avoid
the use of excessive diameters, it is possible to use the arrangement
depicted in Figure 3.
20 The source S is replaced by a matrix M of light sources S". The
matrix M is positioned on the focal plane L by reflection off the semireflective
sheet LSR.
In standard operating mode, a single source S1" is illuminated on
the matrix M. When the helmet changes position as depicted in dotted line in
25 Figure 3, the images P1 and P2 reach the edge of the detector. This
configuration can be recognized by simple image processing. When it
happens, the source S1" that was initially active is switched off and another
source S2" is illuminated in order by construction to bring the images P1 and
P2 back towards the centre of the detector D.
30 The axis of projection is oblique, and its orientation 0 is known. For
example, in the vertical plane (x, z), the orientation 0 is given by:
tan0 = S"S"o/f where S"o is that point of the matrix, generally its
centre, that is situated on the opticl axis of the afocal system. The previous
three relationships for C1C2 then become:
y12 = (y1-y2)
z12 ® (z1-z2) + x12.tan0
d122 = x122 + (y1-y2)2 + [(z1-z2) + x12.tan0]2
The latter equality as before gives two values for x12, and
5 therefore two solutions for the vector C1C2. These are no longer symmetric
about the vertical plane (y, z).
One simple way of resolving the ambiguity is to use two sources in
the matrix which are illuminated in succession in order to determine an
orientation of the helmet.,
10
The device according to the, invention makes it possible to achieve
great precision. For example, for an angular field of 45 degrees and a
detector measuring 1000 points by 1000 points, a precision of 0.045 of a
degree, namely 0.7 mrad, is obtained, and this is precise enough for the vast
15 majority of applications.
The helmet can therefore replace the head-up display function
which has high precision in the centre of the angular range and in a wide
range of head positions.
The optical posture-detection device can also be hybridized, in the
20 centre of the angular range, with electromagnetic posture detection which is
not as precise but which does have a very wide measurement range.
10
CLAIMS
1. System for detecting the posture of a moving object (H) in
space, comprising a fixed electrooptical device (®t) of known orientation
5 comprising at least one first point emission source (S) and a photosensitive
matrix-type sensor (D) and, arranged on the moving object, an assembly
comprising three retroreflector devices of the "corner cube" type (C1, C2,
C3),
characterized in that the fixed electrooptical device comprises a
10 telecentric lens essentially comprising a projection lens (L), a reception lens
(L') and a semireflective optical element (LSR) which are arranged in such a
way that:
the first point emission source (S), is arranged at the focal point of
the projection lens (L) by reflection or transmission through the semireflective
15 optical element (LSR),
the image of the first point emission source (S) is arranged at the
focal point of the reception lens (L') by transmission or reflection through the
semireflective optical element (LSR).
20 2. System for detecting the posture of a moving object according
to Claim 1, characterized in that the first point emission source (S) or the
image thereof is arranged on the optical axis common to the projection lens
(L) and to the reception lens (L').
25 3. System for detecting the posture of a moving object according
to Claim 2, characterized in that the system comprises a second point
emission source (S'), the second point emission source or the image thereof
being arranged off the optical axis common to the projection lens (L) and to
the reception lens (L').
30
4. System for detecting the posture of a moving object according
to Claim 3, characterized in that, with the first source (S) emitting in a first
spectral band, the second source () emits in a second spectral band which
is different from the first spectral band of the first source.
35
11
5. System for detecting the posture of a moving object according
to Claim 3, characterized in that, with the first source (S) emitting light in a
first predetermined state of polarization, the second source (S) emits light in
a second predetermined state of polarization which is different from the first
5 predetermined state of polarization.
6. System for detecting the posture of a moving object according
to Claim 3, characterized in that the first emission source (S) and the second
emission source (S) emit at different moments in time.
10
7. System f®r detecting the posture of a moving object according
to Claim 1, characterized in that the fixed device comprises a matrix (M) of
point emission sources.
15 8. System for detecting the posture of a moving object according
to Claim 1, characterized in that the moving object ( H) comprises at least four
retroreflector devices of the "corner cube" type (C1, C2, C3, C4).
9. System for detecting the posture of a moving object according
20 to one of Claims 1 or 8, characterized in that each of the retroreflector
devices of the corner cube type (C1, C2, C3, C4) arranged on the moving
object comprises optical or geometric discrimination means which are
different from those of the other corner cubes.
25 10. System for detecting the posture of a moving object according
to Claim 9, characterized in that each corner cube (C1, C2, C3, C4)
comprises a mask of a shape that is different from that of the two other
corner cubes.
30 11. System for detecting the posture of a moving object according
to Claim 9, characterized in that each corner cube (C1, C2, C3, C4)
comprises an optical filter the transmission spectral band of which is different
from that of the two other filters of the two other corner cubes.
12. System for detecting the posture of a moving object according
to one of the preceding claims, characterized in that the moving object (H) is
a pilot helmet.
| # | Name | Date |
|---|---|---|
| 1 | 2429-del-2012-Correspondence Others-(08-11-2012).pdf | 2012-11-08 |
| 1 | Translation-Search Report.pdf | 2012-08-06 |
| 2 | 2429-del-2012-Form-3-(08-11-2012).pdf | 2012-11-08 |
| 3 | 2429-del-2012-Correspondence Others-(04-09-2012).pdf | 2012-09-04 |
| 4 | 2429-del-2012-GPA-(04-09-2012).pdf | 2012-09-04 |
| 4 | Form-1.pdf | 2012-08-06 |
| 5 | Drawings.pdf | 2012-08-06 |
| 6 | 2429-del-2012-GPA-(04-09-2012).pdf | 2012-09-04 |
| 6 | Form-1.pdf | 2012-08-06 |
| 7 | 2429-del-2012-Correspondence Others-(04-09-2012).pdf | 2012-09-04 |
| 8 | 2429-del-2012-Form-3-(08-11-2012).pdf | 2012-11-08 |
| 9 | 2429-del-2012-Correspondence Others-(08-11-2012).pdf | 2012-11-08 |
| 9 | Translation-Search Report.pdf | 2012-08-06 |