Celestial Compass with sky polarization

Celestial Compass with sky polarization
US 20150042793A1
(19) United States
(12) Patent Application Publication (10) Pub. No.: US 2015/0042793 A1
(43) Pub. Date:
Belenkii et al.
vs. C].
CPC ............. .. @010 21/02 (2013.01); G01S 3/7864
(2013.01); G01S 3/7861 (2013.01); G01S
3/7867 (2013.01)
(71) ApplicantszMikhail Belenkii, San Diego, CA (US);
Lawrence Sverdrup, Poway, CA (US);
Vladimir Kolinko, San Diego, CA (US)
(72) Inventors: Mikhail Belenkii, San Diego, CA (US);
Lawrence Sverdrup, Poway, CA (US);
Vladimir Kolinko, San Diego, CA (US)
Feb. 12, 2015
........................................................ ..
A celestial compass With a sky polarization feature. The
celestial compass includes an inclinometer, a camera system
(73) Assignee: TreX Enterprises Corporation
(21) Appl. No.: 13/987,604
Aug. 12, 2013
(22) Filed:
Publication Classi?cation
for imaging at least one celestial object and a processor pro
grammed With a celestial catalog providing known positions
at speci?c times of at least one celestial object and algorithms
for automatically calculating target direction information
based on the inclination of the system as measured by the
inclinometer and the known positions of at least one celestial
object as provided by the celestial catalog and as imaged by
Int. Cl.
G01 C 21/02
G01S 3/786
the camera. Preferred embodiments include backup compo
nents to determine direction based on the polarization of the
sky when celestial objects are not visible.
Patent Application Publication
Feb. 12, 2015 Sheet 1 0f9
US 2015/0042793 A1
11? ///J
Patent Application Publication
Feb. 12, 2015 Sheet 2 0f9
US 2015/0042793 A1
M.8a5gn7u5m .
.i g.
Patent Application Publication
[email protected]
Feb. 12, 2015 Sheet 3 0f9
US 2015/0042793 A1
[email protected]
UE%[email protected]$EQ§
Patent Application Publication
US 2015/0042793 A1
Feb. 12, 2015 Sheet 4 0f 9
\[email protected]
Patent Application Publication
Feb. 12, 2015 Sheet 5 0f 9
US 2015/0042793 A1
*///////////////// /
\§$'\ -
I j '/
FIG. 7
Patent Application Publication
Feb. 12, 2015 Sheet 8 0f 9
Fm w
US 2015/0042793 A1
Patent Application Publication
Feb. 12, 2015 Sheet 9 0f 9
Histograms: AOP Difference
Angle (deg)
FIG. 11
US 2015/0042793 A1
Feb. 12, 2015
US 2015/0042793 A1
better than 2 milliradians are available from suppliers such as
Jewell Instruments with of?ces in Manchester, NH. and
Digikey with of?ces in Thief River Falls Minn. The cost of
these inclinometers typically is in the range of about $60.
[0001] This application is a continuation in part of patent
applications Ser. No. 13/373,009 ?led Nov. 1, 2011 which
Digital Magnetic Compasses
eters (such as Analog Devices ADIS 162097) with accuracies
was a CIP of Ser. No. 12/ 283,785, Portable Celestial Compass
?led Sep. 15, 2008, Ser. No. 12/319,651, Angles Only Navi
one degree, and the presence of steel or other local distur
gation System ?led Jan. 8, 2008 and Ser. No. 12/583,776
Miniature Celestial Direction Detector ?led Aug. 25, 2009
and Ser. No. 12 586,813 ?led Sep. 28, 2009, each ofwhich are
bances will often reduce accuracy of the magnetic compasses
to several degrees or render them useless. Therefore, if posi
Magnetic compasses are typically accurate to only
incorporated herein by reference. This application also claims
tioning of a target depends on the use of a magnetic compass,
substantial position errors could likely result. In the case of
the bene?t of Provisional Application Ser. No. 61/742,448,
?led Aug. 10, 2013.
military operations, the accuracy of current and future ?re
support systems strongly depends on the errors in target coor
eral damage and improve target lethality, a target locator error
[0002] The present invention was made in the course of
work under Marine Corps contract number M67854-12-C
6501 and the United States Government had rights in the
on the order, of less than, 10 meters at 5 km range is needed.
Current target location technology does not meet this stan
dard. The main source of error is magnetic compasses. Com
dinates called target location error. In order to reduce collat
[0003] The present invention relates to direction detection
systems, especially to such systems designed for use in deter
mination of precise locations of targets.
Sky Charts
[0004] The position of celestial objects at any time at any
place on earth is known with extremely high accuracy. These
celestial objects include all recognizable stars andplanets, the
monly a ground-based observer determines target coordi
nates using a laser range?nder, GPS receiver, and magnetic
compass. Under ideal magnetic conditions the measurement
error (usually referred to as an “RMS error” of a magnetic
compass is typically 10-17 milliradians. This corresponds to
the locator error of 50-85 meters at a 5 km range. In many
situations knowledge of the true aZimuth to a target with
precision of much better than 1 degree (about 17.45 millira
dians) is needed. Also magnetic compasses are highly sensi
tive to random errors caused by weakly magnetic distur
bances (e.g. vehicles, buildings, power lines etc.) and local
variations in the earth’s geo-magnetic ?eld. These error
sources are random and cannot be accurately calibrated and
modeled to subtract out. A large magnetic disturbance from
sun and the moon. Celestial objects also include visible man
hard or soft iron effects can result in target accuracy errors of
made satellites. Accurate positioning of the celestial objects
depends only on knowledge of the latitude and longitude
up to 30 to 60 degrees.
Attitude Heading and Reference Systems
positions and on the date and the time to within about 1 to 3
seconds of observation. Latitude and longitude generally can
be determined easily with precision of less than one meter
with global positioning equipment. Computer programs with
astronomical algorithms are available that can be used to
calculate the positions of any of these celestial objects at any
Attitude heading reference systems (AHRSs) are
3-axis sensors that provide heading, attitude and yaw infor
mation for aircraft and other systems and components.
AHRSs are designed to replace traditional mechanical gyro
time for any position on or near the surface of the earth. Star
scopic ?ight instruments and provide superior reliability and
pattern recognition computer programs are available in the
prior art. These computer programs are described in several
accuracy. These systems consist of either solid-state or
MEMS gyroscopes, accelerometers and magnetometers on
all three axes. Some of these systems use GPS receivers to
good text books including Astronomical Algorithms by Jean
Meeus, published by Willmann-Bell with of?ces in Rich
mondVa. Techniques for using the programs to determine the
positions of the celestial objects are clearly described in this
reference. Programs such as these are used to provide plan
improve long-term stability of the gyroscopes. A Kalman
?lter is typically used to compute solutions from these mul
tiple sources. AHRSs differ from traditional inertial naviga
etarium programs such as “The Sky” available from Software
tion systems (INSs) by attempting to estimate only attitude
(e.g. pitch, roll) states, rather than attitude, position and
Bisque and “Guide” available from Project Pluto.
velocity as is the case with an INS.
AHRSs have proven themselves to be highly reli
Fisheye Lenses
able and are in common use in commercial and business
[0005] Fisheye lenses are lenses with a highly curved pro
truding front that enables it to cover a solid angle of about 180
degrees. The lenses provide a circular image with barrel dis
brought the price of Federal AviationAdministration certi?ed
MEMS Inclinometers
[0006] Vertical at the observation position can easily be
found by using an inclinometer. Tiny MEMS type inclinom
aircraft. Recent advances in MEMS manufacturing have
AHRS’s down to below $15,000.
[0010] Although gyroscopes are used to measure changes
in orientation, without the absolute references from acceler
ometers and magnetometers the system accuracy quickly
degrades. As such, when there are extended periods of inter
ferences or errors introduced into the sensing of gravity or
magnetic ?eld performance of the system can be seriously
Feb. 12, 2015
US 2015/0042793 A1
compromised. As a general reference, gravity is almost per
Standards and published in the Review of Scienti?c Instru
fectiit is a constant force that is not in?uenced dramatically
ments in 1949. The accuracy was estimated to be approxi
by anything. The most dif?cult error introduced in sensing
gravity is the acceleration added during movements. Each
time the system or component is moved, acceleration is
sensed, thus creating a potential for error. This however is
mately 1°, decreasing if the zenith is obscured by clouds.
easily mitigated by applying algorithms to the data that ?lter
out such high frequency accelerations, resulting in a very
accurate means of determining the vector of gravity. Note that
this information is used only for initial setup and system
corrections, and is not needed for real-time tracking of orien
The Need
What is needed is a non-magnetic compass that can
operate day and night, and in most weather conditions, and
does not require an un-obscured line of sight to the sun or
tation. Magnetic ?eld disturbances are much more di?icult to
deal with.
[0018] The present invention provides a celestial compass
including a sky polarization feature. The celestial compass
Sky Polarization
[0011] It is known that in general the sky light is polarized
includes an inclinometer, a camera system for imaging at least
one celestial object and a processor programmed with a celes
tangential to a circle centered in the sun and maximum polar
ization is found at ninety degrees from the circle. Therefore,
with the sun close to the zenith the sky light will be polarized
horizontally along the entire horizon. On the other hand,
when the sun is setting in the West, the sky will be maximally
polarized along the meridian and thus vertically at the due
North and South. Toward the zenith just after sunset (or before
sunrise) the degree of polarization of the sky light can reach
its maximum of about 75 percent on very clear days,
[0012] Numerous creatures utilize the sky polarization
compass for navigation, with new examples being continu
ally discovered. Desert ants cannot leave a pheromone trail
because this biochemical signal is subject to evaporation.
Instead they use a sky polarization compass.i Bees also use a
tial catalog providing known positions at speci?c times of at
least one celestial object and algorithms for automatically
calculating target direction information based on the inclina
tion of the system as measured by the inclinometer and the
known positions of at least one celestial object as provided by
the celestial catalog and as imaged by the camera. Preferred
embodiments include backup components to determine
direction based on the polarization of the sky when celestial
objects are not visible.
[0019] In referred embodiments the camera system
includes a telecentric ?sheye lens that produces an image on
the sensor located at or near the focal plane which remains
spatially constant within sub-micron accuracies despite ther
mally produced changes in the focus of the lens. These
sky polarization compass. Migratory birds utilize the earth’s
embodiments may also include a movable ?lter unit to
magnetic ?eld, the stars and the sun as compasses, but the sky
polarization compass is utilized to calibrate all of the other
compasses. Dung beetles have been shown to use a sky polar
ments the ?lter unit includes an electromagnetic switch. In
ization compass at night where sky illumination is provided
increase greatly the dynamic range of the kit and permit day
and night operation with the single lens. In preferred embodi
other embodiments the switch is a manual switch or a motor
driven switch. The ?lter in preferred embodiments is com
by the moon. The ability to use polarization vision in the
animal kingdom is probably much more widespread than we
[0013] It is known that some animals use green light, many
use blue light, but most use near ultraviolet light for their sky
polarization compass. The reason for this is apparently that in
permits imaging of the moon and sun through light cloud
adverse conditions such as complete overcast, the sky polar
ization signal is largest in the UV. In clear conditions, it is
navigation sensor including a magnetic compass and a
largest in the blue/green spectral region.
[0014] The ?rst known sky polarization compass was built
in the 1940’s as a single pixel device measuring the sky at
zenith. It has been reported that the Scandinavian airlines
SAS used a “single-zenith-pixel” sky polarization compass
during polar ?ights in the 1950’s. In the late 1990s a Swiss
group mimicked desert ant navigation, building a robot that
navigated using a single zenith pixel sky polarization com
[0015] A device known variously as the Pfund compass, the
Kollsman Sky Compass, or simply as the “twilight” compass,
was utilized by the US Navy in 1948. It determined the
prised of a thin Mylar ?lm coated with a special partially
re?ective coating. With the increased dynamic range of the
camera the moon can be imaged during the period after sunset
and before sunrise when stars are not visible. The compass
cover. Other preferred embodiments can include an inertial
memory-based optical navigation system that permits contin
ued operation on cloudy days and even in certain in-door
environments. In some preferred embodiments calibration
components may be provided in a separate module to mini
mize the size and weight of the compass.
[0020] These embodiments use celestial sighting of the
sun, moon or stars to provide ab solute azimuth measurements
relative to ab solute north. In preferred embodiments the incli
nometer is an internal MEMS inclinometer providing mea
surements relative to the local vertical (gravity based). Celes
tial observations are combined with known observer position
and time, which can normally be obtained from a GPS
azimuth of the sun when the sun was not visible by examining
receiver, in order to compute the absolute azimuth pointing of
the polarization of the sky at zenith. This proved to be
extremely valuable in the far north, where magnetic com
passes are minimally useful, and twilight conditions can per
advantages over the similar prior art device discussed in the
sist for long durations, during which both sun and stars are not
visible and therefore useful for navigation. The accuracy was
reported to be about 0.50.
[0016] A version of a sky polarization compass that utilized
sky light at zenith was developed at the National Bureau of
the device.
The present invention has the following principal
background section:
[0022] Nonmagnetic compass
[0023] No performance degradation over time (no drift)
[0024] Compact
[0025] No moving parts (other than the ?lter)
Feb. 12, 2015
US 2015/0042793 A1
[0026] Lightweight
for imaging the sun during daytime and for imaging the moon
Low power
Low cost
RMS azimuth measurement error is about 1 mil
and stars during the nighttime. Since brightness levels during
the day are many orders of magnitude greater during the day
as compared to night, applicants have designed an automatic
[0030] Low production cost
[0031] Allow for operation in urban environments, near
vehicles and power lines, and while wearing body armor
Near zero startup time (azimuth measurement in
about 2 seconds)
[0033] FIG. 1 illustrates a preferred embodiment of the
present invention where the celestial compass is an accessory
of a far target location (FTL) system.
[0034] FIG. 2 is a prospective view of a preferred embodi
ment of the present invention.
[0035] FIG. 3 is a cross sectional drawing showing features
of the FIG. 2 embodiment.
[0036] FIG. 4 is an exploded view drawing of the FIG. 2
[0037] FIG. 5 is a breakaway drawing of the electronic ?lter
mechanism of the preferred embodiment.
[0038] FIG. 6 is a drawing showing the lens elements of a
telecentric ?sheye lens specially designed for this preferred
embodiment of the present invention.
[0039] FIG. 7 is a cross sectional drawing ofa portion ofthe
?sheye lens showing detailed features of the lens.
[0040] FIG. 8 is a block diagram showing electronic com
ponents of the above preferred embodiment of the present
[0041] FIG. 9 is a set of speci?cations for the telecentric
?sheye lens system.
FIG. 10 shows an experimental setup for testing sky
polarization components.
FIG. 11 shows test results of the sky polarization
shutter-?lter system permitting the same lens-sensor unit to
be used during the day and at night. The preferred shutter unit
is shown at 20 in the FIG. 3 drawing. The shutter blade is
shown at 22, the ?lter is shown at 24 and the CMOS sensor is
shown at 26. The CMOS sensor is a 5 mega pixel CMOS
sensor Model No. MT9P031 provided by Aptina with of?ces
in San Jose, Calif. FIG. 4, which is an exploded view drawing,
shows additional details of the celestial compass including
lens assembly 14, lens mount 30, shutter unit 20 and shutter
permanent magnetic cover 32. Under the cover (not shown is
an electric magnet in the form of a circularly-shaped coil. The
CMOS sensor is shown at 26. These components are mounted
on circuit board 16.
The shutter-?lter is a modi?ed version of an off-the
shelf shutter available from Uniblitz with of?ces in Osborne,
Wash. The shutter was converted to an “in or out” ?lter. This
shutter-?lter includes a small permanent magnet shown at 32
in FIG. 5 that is positioned within a break in the circularly
shaped coil of the electro magnet. The direction of current
?ow through the coil of the electromagnet determines the
position of ?lter blade 24. A reversal of current in the coil
changes the orientation of the magnet and the shutter blade by
180 degrees. Current ?ow in a ?rst direction orients the ?lter
above CMOS sensor 26 for imaging the sun during daytime
operation of the celestial compass and current ?ow in the
opposite direction orients the ?lter away from the sensor for
nighttime operation for imaging the moon or stars. The ?lter
blade is held in place by friction if no current is ?owing in the
coil. So current is required only when changing the ?lter
position. The ?lter itself is a thin ?lm ?lter on a polyester
(preferably Mylar®) substrate providing 106 blocking.
Telecentric Lens
First Preferred Embodiment
[0048] FIGS. 6 and 7 are drawings of telecentric ?sheye
lens utilized in the preferred embodiment of the present
A ?rst preferred embodiment of the present inven
tion can be described by reference to FIGS. 1 through 9. FIG.
1 shows a celestial compass as a component of a far away
target location system mounted on a tripod. The celestial
compass has imaged the sun and with information from an
invention. The lens unit consists of seven optical elements
shown as elements 1 through 7 in FIG. 6. The mechanical
details of the layout are shown in the cross sectin drawing of
FIG. 7. It consists of a single lens tube with a varying diam
eter. The inner diameter of the tube at each axial position
matches the diameter of the lens elements and spacers that it
inclinometer (not shown), the correct date and time and the
correct geographic position of the laser ?nder, the processor
within the celestial compass has determined the orientation of
a telescope in the far target location system and with the
timing of a return infrared laser pulse from target has deter
mined the exact geographic position of the target.
[0045] A preferred module of the celestial compass of the
present invention is shown in detail in FIGS. 2 through 8. FIG.
threaded retainer ring 48 for holding lens element 2, several
holes 52 in lens mount 42 for permitting injection of adhesive
2 is a prospective view of the celestial compass. Shown in the
to ?x lens elements 3-7 and their associated spacers, spaceri
drawings is celestial compass, with a single ?sheye lens
optical stop 54 hole 46 for adhesive for ?xing lens element 1
and spacer 56 for setting the space between lens elements 6
assembly 14 mounted on circuit board 16. Also shown is
inclinometer unit 18 which is an off-the-shelf unit, Model
ADIS 16209 furnished by Analog Devices with of?ces in
Norwood Mass.
[0046] FIG. 3 is a cross sectional drawing showing some
contains. An integral skirt is part of the lens mounting struc
ture and is used to attach the lens to an outer structure. Shown
in FIG. 7 are lens mount structure 42 to hold the lens ele
ments, a threaded retained ring 44 for holding lens element 1
and to preload in compression all subsequent lens elements, a
and 7. Two sets of cemented doublets are constructed using
lens elements 3&4 and 5&6 as shown in FIGS. 6 and 7. The
additional features of this preferred embodiment. This celes
speci?cations for the optical elements are found in the table in
FIG. 9. Lens element 1 is held in place with retaining ring 44
which compresses the element against a ledge in the lens
tial compass utilizes a single lens and a single CMOS sensor
mount. In order to insure mechanical stability each element of
Feb. 12, 2015
US 2015/0042793 Al
the lens and each spacer is attached to the lens tube by way of
temperature volcanizing (RTV) silicone. For elements 1 and
2 the adhesive is applied in a 360° ring around the lens
element. For elements 3-7 and the spacers around these ele
ments as pictured in FIG. 7. The lens mount structure 42 has
a series of holes 52 in it by which the adhesive may be injected
as described above. The process of delivering the adhesive
should insure that the adhesive contacts the side of the lens
element or spacer that is radially in line with, and ?ll the entire
[0055] 4) Determine azimuth offset by taking difference
between measured azimuth (step 3) and known sun posi
tion (from time and position).
[0056] 5) Mathematically rotate boresight pointing in
inclinometer coordinates to local horizon coordinates
(with unknown azimuth) using inclinometer measure
hole. Four adhesive holes are distributed at 90° increments at
each axial hole position. In order to facilitate applying the
adhesive into the holes in the lens tube corresponding holes
are position radially in the skirt structure. These allow a
hollow adhesive dispensing tube to access the inner holes. To
insure stability over a wide temperature range the housing
structure, retaining rings, and lens spacers are made of tita
Electronic Components
[0049] FIG. 8 is a block diagram showing important fea
tures of the electronic components of the above preferred
embodiment of the present invention. These components
include a set of voltage regulators 60 supplied by and external
5 volt source 62 and an external interface connector 64 in
communication with digital signal processor 66 which is a
3) Mathematically rotate azimuth and zenith from
inclinometer frame to local horizon frame with
unknown azimuth offset.
an adhesive. The preferred adhesive is a non-outgassing room
6) Determine absolute azimuth of boresight by
azimuth offset determined in step (4).
[0058] Calibration procedure: Reverse steps (5) and (6)
above while siting targets with known absolute azimuth. The
calibration procedure and the procedure for absolute target
azimuth and zenith (elevation) angle determination is
described below.
[0059] A brief description of variable notation is summa
rized in Table 2. The reader should note that all coordinate
rotations are based on small angle approximations. This
seems reasonable since all measurements of the optical axis
offset from the inclinometer z-axis (zenith pointing for zero
readings) show angles less than 10 milliradians. All measure
ments were based on objects with inclinometer pitch and roll
readings less than 5 degrees.
Devices with of?ces in Norwood, Mass. The processor is
[0060] The sun position on the sensor is determined by a
center of mass calculation. A matched ?lter determines the
programmed and de-bugged with JTAG interface 68. The
location of the sun (not necessary simply ?nding the peak is
DSP module (Model Back?n 537) supplied by Analog
output of DSP 66 is an input to an Ethernet PHY chip 70
suf?cient). The background (+camera analog to digital bias)
(Model KS8721BLI) supplied by Micrel Inc. with of?ces in
San Jose, Calif. and a 20 pin connector 72 which provides for
is determined as the average of a 32x32 pixel region centered
on the peak and excluding the center 16x16 pixels. A center of
a connection with a simulator an a image display monitor (not
mass calculation is made including only those pixels in the
shown). The DPS module 66 is also in communication with
16x16 region with signal exceeding 5% of the peak value.
CMOS sensor 26 via an 12C level shifter 73 and a 12 bit Data
Bus as shown in FIG. 8. And the module 66 is also in com
the optical axis on the sensor is a linear function of the zenith
munication with shutter controller 74 and inclinometer 18
through an 8 Bit l/O expander as shown in the drawing. The
inclinometer is a small high accuracy, dual-axis digital incli
The equations assume that the image distance from
angle under the following additional assumptions:
[0062] 1) lnclinometer axes are orthogonal. (Presumably
determined by lithography/etch on MEMS since both
nometer and accelerometer Model ADIS 16209 supplied by
Analog Devices with of?ces in Norwood, Mass.
axes were on a single die).
[0063] 2) Row/ column axes combined with ?sheye bore
sight constitute an orthogonal coordinate system.
Process for Converting Celestial Data Into Target
[0050] To determine the accurate location of a small celes
tial target relative to the camera requires only a centroid
measurement. To determine the accurate celestial location of
the sun or moon requires ?nding the edges of the target and
then calculating the true center based on the size and shape of
the target at the time of the observation. The software as
indicated above must correct for the distortion of the ?sheye
Parameter De?nitions
(x50, yso) = array center in pixels on sensor
Ax = angular pixel size
(ax, [55) pitch and roll of ?sheye optical axis with respect to
inclinometer z—axis (zenith for leveled inclinometer)
(4)17, 61,) = azimuth and zenith angle ofbinocular boresight
in inclinometer reference frame.
lens while also converting image data into astronomical coor
dinates, preferably elevation, bank and azimuth.
[0051] Outline of basic daytime algorithm processing
[0052] 1) Measure sun azimuth and zenith on the ?sheye
where radius to center is proportional to the zenith angle
and azimuth is the angle between column offset and row
offset from the center.
2) Mathematically rotate azimuth and zenith
angle (small angle approximation) from sensor/?sheye
frame to inclinometer frame (i.e. calibrate by determin
ing ?sheye boresight when inclinometer is zeroed).
Measured Quantities
(x5, ys) = sun centroid on sensor
(6%, 6y) = inclinometer measured pitch and roll.
(4):, 65) = measured sun azimuth and zenith angle in
Calculated Quantities
sensor/?sheye frame
(4)0, 60) = measured sun azimuth and zenith angle in
inclinometer frame
((1),, 6,) = measured sun azimuth and zenith angle in
module based local horizon coordinates
A¢SM = yaw of module based local horizon coordinates
relative to true local horizon coordinates (ENU).
Feb. 12, 2015
US 2015/0042793 A1
TABLE 2-continued
cos 0
Parameter De?nitions
sin 0'
¢,' = absolute azimuth of the sun in local horizon
coordinates (ENU) calculated based on solar ephemeris,
time, and geo—location
The slope of a linear least squares ?t provides the axis pitch
¢b,' = absolute azimuth of the target
Detailed equations are set forth below:
(or roll), and the intercept provides the offset in center column
Coordinate system for sun position analysis.
Measure sun centroid (x5, yS)
Azimuth and zenith angles in sensor coordinates
(or row).
Error Analysis
[0066] The following is an error analysis. It is based
directly on the coordinate transformation equations detailed
above, so it cannot be considered an independent check. The
0, = AX
(x. — sz + (Y. — no?
results are based on small value approximations. As a ?rst
approximation two axis values which add in quadrature phase
(a cos x+b sin x) are simply combined in a single “average”
Rotate to optical axis
4%. = ¢. + (6. sin ¢. + (1. cos ¢.> cot 6.
90 = 9. + H5. COS (I). + (1. sin (PS)
term, and systematic errors (such as errors in determining the
Rotate to local horizon using inclinometer measurements, (6%, 6y)
calibration parameters) are treated in the same manner as
4), = (1)0 — (6y sin (1)0 — 6X cos (1)0) cot 60
random errors (centroid measurement error, mechanical drift,
q» = 4%. + (By cos ¢.. + 6.. sin qt.)
inclinometer noise, etc).
M... = M - 4»
Where 4); is the absolute azimuth of the sun.
Rotate boresight to local horizon coordinates
(Pb! = (Pb - (9y Sin (Pb - 9); 005 (Pb) mt 9b
61,, = 61, + (By cos 4), + 6X sin 4);)
(PMy = $171+ A¢SMYI
Where ¢bl' is the absolute azimuth of the target, and 6b, is the
[0067] An attempt is made to maintain consistent notation
with the explanation of the coordinate transformation. For the
simpli?ed case with the inclinometer level, the variance in
determining absolute azimuth is approximately:
absolute zenith angle of the target.
01%, zo'wb +le
+(Sin2 as 10'; +
Calibration Procedures
[0064] Several calibration parameters must be determined
experimentally. They are listed as the ?rst set of items (1)
through (4) in Table 2. Based on small angle approximations
the systematic error in measured azimuth resulting from
errors in the array center point and off zenith ?sheye boresight
A brief summary of the terms is listed in Table 3.
is given by:
Summary of error contributions for leveled operation.
(1) 0% = error in boresight azimuth calibration
cos OS
sin OS
0w = error in calculated sun location in ENU frame.
Time, geo—location, and ephemeris errors are all believed to
be negligible. Error for
(3) as = average of ?sheye boresight angular offset from inclinometer
where Aq) is the error in the azimuth measurement, ((1)6, A66)
0,65 = error in sun position on sensor (centroid accuracy based on
describes the azimuth and zenith angle on the error in center
radiometric SNR, gain variation, and image distortion). SNR
position, and the remaining parameters are described in Table
2. Notice for a ?xed zenith angle, errors in boresight pointing
contribution believed to be small (image ~3 pixels and camera gain,
may be corrected by the errors in center location. The expres
sion may be rewritten in terms of an effective center point and
large zenith angles is under investigation.
divided into sensor row and column,
exposure time set to ~200 counts out of 255, noise measured < 1
bit rms). Gain variation not measured. Image distortion, especially for
A—Xe = fractional error in pixel size (based on linear fisheye
response, more generally( Ax )0; should be re—placed as systematic
error in measuring zenith angle). Response nonlinearity suspected
problem. Correction under investigation
(6) 005 = error in determining ?sheye boresight calibration parameters
plus boresight drift (time/temperature). Fisheye boresight calibration
[0065] The calibration procedure takes advantage of this
property by determining the center location which minimizes
long term repeatability under investigation.
(7) 09X = noise in inclinometer measurement.
the azimuth error (in the least squares since) for a series of
measurements at a constant (or near constant for sun) zenith
angle. The procedure is repeated for several zenith angles, and
an additional error term which is proportional to the magni
If the device is permitted to pitch and bank, there is
the results are plotted as a function of
tude of the pitch and/or bank of:
Feb. 12, 2015
US 2015/0042793 A1
Inertial Navigation
sin 0S cos 0S 2
Ax e
—Sin20S Jahaie ] + I] + o‘as(l + cos 0;) + M )0;)
[0079] One alternative to overcome these limitations Appli
cants have added an inertial navigation component developed
at Innalabs Inc. with o?ices located in Dullas, Virgina and
image-based navigation system for position and weapon atti
tude determination for indoor conditions developed by Evo
lution Robotics with o?ices located in Pasadena, Calif. The
[0070] Where a contribution from the boresight zenith
angle relative to inclinometer zenith has been omitted (as
use of Innalabs component permits the minimization of the
sumed negligible). The reader should note that this corre
sponds to an rrns value instead of the variance shown for
rate on module performance. The use of Evolution Robotics
leveled operation. All of the error terms are the same as
described in Table 3 with the exception of, sex, the inclinom
eter measurement error. For pitched/banked operation, the
inclinometer measurement error now includes not only noise,
but any gain or nonlinearity contributions.
effect of environmental conditions and high angular motion
image based navigation system permits determination of
position and attitude during indoor exercises.
[0080] The memory-based optical navigation system
includes a processor programmed with images of the envi
ronment where the training is to take place. Images of the
In addition to the error sources discussed above, the
environment recorded by a camera mounted on the ri?e are
measurements will have two additional error sources. The
analyzed with special algorithms by a computer processor
?rst is the accuracy of the reference points. The second is
pointing the Vector 21 (~1.2 mr reticule diameter). Current
which determines, from the camera images and the pro
rough estimate is that these error sources are on the order of
[0081] Embodiments of the present invention also include
software permitting users to identify landmarks imaged by
0.5 mr rms.
Test data proving the accuracy of this embodiment
utilized with the Victor 21 binoculars and with a theodolite is
reported in parent patent application Ser. No. 12/283,785
which has been incorporated herein by reference.
grammed images, the pointing direction of the ri?e.
the camera and to determine directions to those landmarks
from speci?c locations during cloudless periods and to use
tho se landmarks and directions as references for determining
ri?e pointing directions when clouds obscure the sun or stars.
Once the target is identi?ed, additional software
determines the orientation of the camera. Astronomical algo
rithms and celestial navigation software suitable for program
ming computer 22 is described and provided in several well
known texts including Astronomical Algorithms by Jean
Meeus that is referred to in the Background Section. Once the
camera orientation is known, the azimuth of the instrument is
easily computed.
Boresighting the Module with Other Instruments
Calibration of the module with other optical instru
Single Camera and Multiple Cameras
[0082] Embodiments of the present invention can be
designed for daytime operation based on the location of the
sun and other embodiments can be designed for operation
based on the position of the moon, the stars and other celestial
objects such as man-made satellites. Or as described above
with respect to FIGS. 1 through 9 the embodiments can be
designed to operate day and night using a single camera.
Alternatively as described in some of the parent application
ments requires a single calibration. A target at a knowrement
more than one camera can be included with at least one
is made. The azimuth reported by the celestial measurements
is then rotated to agree with the other optical instruments.
designed for night-time use.
Calibration Module is Separate
[0075] As indicated in FIG. 8 the calibration module (in
cluding Ethernet PHY chip 70, 20 pin connector 72 and JTAG
connector 68) is a separate module from the DPS Module 66
and circuit board and the optical components in order to
minimize the size and weight of the celestial compass.
Advantages and Limitations of the Celestial
camera designed for day-time use and at least one camera
Applicants’ earlier versions of their celestial com
pass included separate optical sensors optimized for daytime
and nighttime operation along with two small digital cameras
and miniature optical lenses. However, to meet the size,
weight, and power requirements for determining pointing
direction for ri?es, a single-sensor design is preferred. The
challenge is that a very large sensor dynamic range of 101 l to
1013 must be accommodated in order to measure the position
of both the sun and stars. Exposure time and gain control
generally provide for a range of approximately 105 in illumi
nation. To enhance the system’s dynamic range, Applicants
A principal advantage of use of the celestial com
have developed the ?lter described above. The mechanical
pass as compared to a magnetic compass is that it can con
neutral density ?lter described above provides the dynamic
tinuously measure absolute heading relative to the Earth’s
true north with accuracy of 1 mil without the use of pre
emplaced infrastructure and does not rely on the use of mag
netic compass. However the celestial compass shown in
range required for day/night operation. A motor inserts or
removes the ?lter in about 1 second for day/night operation.
The motor is approximately the same size as the ?sheye lens.
Focus maintained by using a very thin ?lter, such as 12 micron
FIGS. 1 through 9 has limitations:
[0077] a) It cannot operate in the presence of heavy
thick aluminized Mylar ?lm, such that the change in focus is
negligible when the ?lter is inserted. An alternative ?lter
clouds, fog, and smoke, and
would be to use a glass ?lter with a transparent piece of glass
or moon is obscured by trees, buildings or other struc
adjacent to the ?lter glass. This second optic would maintain
the optical path length, and would appear in the gap as the
tures, for example, in urban environments.
?lter wheel rotates.
b) It cannot operate when line of sight to the sun
Feb. 12, 2015
US 2015/0042793 A1
Imbedded Micro-Processor
components can function in partly cloudy sky conditions.
[0084] The estimated number of operations required for the
daytime sensor to determine target azimuth by imaging the
for a clear day or night, of 0.1 mil, for a cloudy day of 0.753
sun is 40 million operations per second. As explained above a
preferred micro-processor that meets this requirement is the
BlackFin embedded processor ADSP BF537 available from
Analog Devices. This processor has many several advanta
geous features such as very low power consumption (400
Test results have demonstrated an RMS target azimuth error,
mil, and for cloudy night of 0.75 mil.
[0089] When clouds, fog, or smoke interfere with celestial
measurements using the celestial direction components, the
inertial navigation components which includes continuous
input from the magnetometer will serve as a “?y wheel”
mW), a small size in a mini BGA package, a very low cost
carrying the celestial ?x forward and determining the weap
(approx. $45 in small quantities), and a scalable family of pin
and code-compatible parts. The compatible parts allow the
processor to ?t the application without requiring major
on’s orientation. However, even in this case, the input from
the magnetometer will include corrections (based on the last
available azimuth measurement from the celestial direction
changes to either the hardware or the ?rmware.
components) which permit mitigation of the errors caused by
the Earth’s declination angle and by large magnetic distur
Inertial Navigation Component
[0085] The celestial and inertial measurements features of
the present invention complement each other well. The celes
tial measurements are very accurate with essentially no drift
over long intervals, but will only be available intermittently
due to high sensor motion and environmental conditions. The
inertial measurements have very high bandwidth and are
accurate over short time periods, but suffer from drift over
long time periods. The two are integrated in a typical Kalman
?lter architecture. All sensors (i.e. the optical sensor, the
inclinometer, the inertial navigation component and the mag
netic compass if one is used) feed data directly to the main
processor. The main processor will implement a Kalman ?lter
to optimally combine the inputs from all four sensors.
[0086] The Kalman ?lter will include estimates for the
accelerometer gain and bias drift based on the GPS position
updates, gyro gain and bias drift based on the magnetic com
pass and the celestial sensor, and magnetometer bias drift
based on the celestial measurements. Since the celestial mea
surements constitute the most computationally intensive
measurements, they will only be updated once every 10 sec
onds. In the interim, the celestial sensors will be put in
standby mode, and the processor clock will be reduced to
[0090] Finally, the above describe preferred embodiment
has been designed for extremely low power consumption.
Various modes of operation are provided: full sleep mode;
ready, or stand-by, mode; and operational mode. In the stand
by mode, the microprocessor requires less than 1 mW.
Cloudy Weather
[0091] As indicated above in connection with the descrip
tion of preferred embodiments. The primary components of
the present invention cannot function as desired in cloudy
weather or in similar situations when the celestial objects are
not visible to the system’s sensors. For these reasons embodi
ments may be equipped with a backup digital magnetic com
[0092] This magnetic compass can be calibrated periodi
cally using the features of the present invention and can take
over when the heavens are obscured. Alternatively or in addi
tion a miniature attitude and reference system such as the
systems discussed in the background section of this speci?
conserve power.
Power Consumption
In clear sky conditions day and night, the celestial
direction components provides periodic precision azimuth
measurements with respect to Earth’ s true north and provides
periodic (every 10 seconds) updates to the Kalman ?lter. The
module provides a key element to the initial alignment at start
up. Based on celestial azimuth measurements, the Kalman
?lter estimates the magnetometer bias drift, as well as gyro
gain and bias drift. This allows the module of the present
invention to mitigate the errors related to the Earth’ s declina
tion angle occurring over time. The inertial navigation com
ponents correct for ri?e movement over short periods. Addi
tionally, the lO-second updates eliminate errors associated
with local magnetic disturbances. On the other hand, using
inputs from the magnetometer, the effects of highly dynamic
conditions on performance is mitigated. The inertial naviga
tion components continuously measure the weapon’s motion
cation may be added to allow the target information to be
determined in the event that clouds obscure the celestial
objects. Also when systems of the present invention is located
at a particular location the precise location to a local landmark
can be identi?ed by the system and utilized to provide refer
ence directions later in the event of cloudy weather. To utilize
this feature an additional camera may be required to assure
that an appropriate local landmark is in the ?eld of view of
system camera. Another alternative for direction determina
tion when celestial objects are not visible is to include a sky
polarization feature.
Sky Polarization Feature
In order to characterize the polarization of sky light
over any ?eld of view utilizing intensity measurements, a
minimum of three measurements may be required. As
explained in the background section, during daytime the sky
and provide that information to the processor where it is used
to determine the aiming direction of the ri?e.
is polarized in circles around the sun even in cloudy con
Partly Cloudy Skies
measurements of intensity are made after the light has been
made to pass through a linear polarizing ?lter. In order to
make many such measurements over a region of the sky,
several methods have been utilized or proposed:
Best results from the celestial direction components
are achieved on cloudless days and nights. However these
sitions. Applicants have determined that at night the sky is
similarly polarized around the moon. Typically polarization
Feb. 12, 2015
US 2015/0042793 A1
1.A telescope with a single-pixel intensity detec
tor is scanned over the region of interest and a data taken
for at least three orientations of an included polarizing
?lter at each position.
[0095] 2. A telescope is scanned over the region of inter
est. The light from a telescope is split and directed to at
least three single-pixel intensity detectors, each with a
?xed polarizing ?lter oriented in an appropriate fashion.
Reference Images
[0106] The result of polarized light traversing complicated
optics could be extremely difficult to accurately characterize.
(FPA) detector is utilized with a rotating polarizing ?lter.
The optics could include numerous lenses and coatings with
unknown manufacturing variations and defects. The light will
pass through a polarizing ?lter at various angles with respect
to the normal, and the properties of the polarizing ?lter will
not be perfectly uniform. The focal plane array will have
non-uniformities in the pixels. A simple way to circumvent
these dif?culties is the following. Reference images are
At least three exposures, each with a different orienta
recorded of the sun in a clear sky, at various zenith angles,
3. A camera or a telescope with a focal plane array
tion of the polarizing ?lter are required to acquire the
necessary data.
[0097] 4. The light from a camera lens or telescope is
split a delivered to three FPA’s simultaneously. Each
FPA has a ?xed polarizing ?lter oriented in an appropri
ate fashion. Only one set of simultaneous exposures is
required to collect data. Only one exposure is required to
take data.
[0098] 5. A single imaging camera or a telescope with a
single FPA detector is used. Polarizing ?lters are asso
ciated with individual pixels of a focal plane array. At
least three orientations are utilized, and only a fraction of
the pixels (1/3 at most) records information for a speci?c
orientation of the polarizing ?lter. Only one exposure is
required to take data.
There are issues with most of the above schemes.
The sun (and moon) is continually moving, as are clouds. In
order to make accurate measurements, the sun (or moon) and
the atmosphere must be effectively frozen. Schemes l & 2 are
not preferred, as they generally are too slow. Scheme 3 can be
made to work, if the exposures are made at video rates,
although some change in cloud pattern could occur over the
using the ISPC. A reference angle-of-polarization (AOP)
image is computed for each zenith angle and stored in a data
base. The azimuth of the sun with respect to the ISPC is
recorded with each reference image. The location of zenith in
the AOP images is recorded with each reference image.
In Use
In use, when a new image is recorded, the AOP
image is computed and is compared to a reference AOP image
with the same zenith angle. If necessary, the comparison
reference AOP image is interpolated between two database
images with zenith angles bracketing the zenith angle for the
current data. Either the reference AOP image or the new AOP
data image is mathematically rotated about zenith and the
degree of correlation with the other used to determine the best
match. The amount of rotation required to obtain the best
match determines the azimuth offset of the current sun posi
tion from that of the sun in the reference image, and deter
mines the azimuth of the ISPC.
required three frames. This can be accomplished utilizing
ferroelectric liquid-crystal modulators. These in general are
[0108] Preferred embodiments of the present invention
includes this hybrid azimuth sensing system will increase the
availability of nonmagnetic highly accurate azimuth solution
up to 85%; enabling operability to persist in cloudy skies,
associated with narrow operating bandwidth and cannot be
made to function in the ultraviolet. There are issues with
completely overcast conditions and conditions when a line
of-sight to the sun is obscured by trees, buildings, or other
reproducibility of the polarization axis versus applied voltage
structures, and even when a forward observer operates in a
at different temperatures. Scheme 4 avoids these issues, but at
hole with only a limited area of the sky available for viewing.
the cost of three cameras and three ?lters instead of one each.
Therefore a system based upon scheme 4 is more expensive,
bulkier and heavier and consumes more power. Scheme 5
Additionally, the hybrid north ?nding system will also pro
avoids all of the previous issues, although the data for each
polarization axis is sparser, and must be interpolated.
Although in cloudy skies the degree-of-polarization pattern
can be quite noisy, the direction-of-polarization pattern is
always determined primarily by single Rayleigh scattering,
and the pattern is smooth and predictable. Hence sparse data
does not present a fundamental problem. Therefore, scheme 5
is the best.
Imaging Sky Polarization Compass
[0100] An imaging sky polarization compass (ISPC) con
sists of ?ve principal components:
[0101] A wide-angle or ?sheye lens, or a simple aperture
[0102] A polarizing ?lter or ?lter array
[0103] A focal plane array sensor
[0104] An inclinometer to determine pointing of the
optical axis of the lens with respect to zenith
[0105] A processor to control the camera, take the appro
priate data and compute an azimuth
vide accurate azimuth in twilight conditions during and after
sunset and prior to and during sunrise, when celestial bodies
are generally not visible. This capability is increasingly
important in higher latitudes (i.e. polar regions) that experi
ence much longer twilight hours.
[0109] Applicant’s sky polarization north ?nding system
mimics a similar solution exploited by nature. As explained in
the background section of this application, many insects and
animals are known to use the sky polarization pattern for
navigation. The hybrid azimuth sensing system will use a
low-cost in-house polarizer-on-pixel technology to enable
two operational modes: i) celestial mode when the sun is
above the horizon and an imaging sensor is able to record the
sun images or when the moon and stars are visible at night and
ii) polarization mode, when the sun cannot be imaged by the
sensor due to adverse weather conditions, or because the
line-of-sight to the sun is obscured by trees, buildings, or
other structures. This technology will increase the availability
of azimuth solution for worldwide weather up to 85%. The
hybrid system achieves a compact and lightweight form fac
tor by cleverly leveraging common hardware architectures,
including a ?sheye lens, processor and electronics board
native to the Applicants’ celestial compasses described in the
Feb. 12, 2015
US 2015/0042793 A1
parent applications listed in the third paragraph of this speci
?cation, for both the celestial and polarization-based north
?nding modules.
FIG. 11 is an example showing histograms of the AOP difer
ence between the polarization images and the reference
images computed from the test data. Current single measure
ment azimuth accuracy in partly cloudy conditions is in the
Applicant’ s Experiments
range from 0.10 to 0.30. Under conditions when the line-of
[0110] Applicants’ experiments hade demonstrated the
sight to the Sun was blocked by clouds or by a nearby object
imaging of bright stars in daytime using an infrared camera
with a 50 mm lens, and azimuth sensing in overcast condi
tions using the novel sky polarization measurement tech
nique. Although longer wavelengths such as short wave infra
red can better penetrate clouds and smoke, the ability to
image the sun and stars is effectively eliminated with signi?
cant levels of cloudiness. The sky polarization pattern, how
ever, typically persists in completely overcast skies at a
detectable level in the near ultraviolet spectral range. The sky
polarization technique, demonstrated by Applicants exploits
this very important phenomenon; enabling a path towards
achieving an accurate all-weather azimuth solution.
[0111] By imaging a signi?cant portion of the sky, the
signal-to-noise ratio and thus the single measurement accu
racy can be improved. In poor sky conditions, the optimal
regions of the sky for polarization measurements are more
likely to be interrogated with a large ?eld-of-view. In cases
with restricted access to the sky such as under canopies or in
urban environments with tall buildings, a portion of the un
obscured sky is likely to be found.
[0112] Applicants experiments have focused on developing
and improving components for sky polarization measure
ments; culminating in a sensor system based upon a rotating
polarizing ?lter with an optical encoder to keep track of the
polarizer angle and trigger the camera at the appropriate
times. FIG. 10 shows an experimental setup.
(building), the system performance is comparable to the clear
sky conditions, 0.1°. System single measurement perfor
mance is typically in the range from 0.3 to 0.5 degrees under
fully overcast sky conditions.
[0123] Applicants anticipate accuracy gains will be
achieved by (ii) increasing the FOV up to 180 degrees, (ii)
increasing system dynamic range from 8 bits to 12 bits, (iii)
increasing camera frame rate up to 120 Hz and averaging of
multiple measurements, and (iv) improving image quality
metric used for “good” pixel selection based on Malus Law.
The use of the Malus Law pixel ?lter quali?es image data
inputs to the azimuth calculation to further increase con?
dence and accuracy.
[0124] Preferably miniature prototypes should incorporate
polarizer-on-pixel technology in order to achieve size weight
and power needs for appropriate to handheld applications, as
well as to increase design robustness by eliminating moving
parts. Ideally, the polarizers would be fabricated on the pixels
at a foundry. However, this process is still in development and
is too expensive to be a viable, near-term solution. For opti
mum extinction coe?icient and transmission, the polarizers
are ideally composed of high conductivity metal strips with a
pitch signi?cantly smaller than the wavelength of interest. In
the near ultraviolet range (350 nm) this means line pitches of
the order of a few hundred nanometers or less. CMOS devices
generated by Mukul Sarkar used a line pitch of 0.48 micron
[0113] The sky polarization compass software developed
and was actually inadequate for use in the near ultraviolet.
by Applicants uses the current sky AOP pattern and a pattern
matching algorithm to ?nd the best reference image of the
Utah, has developed a process for depositing parallel alumi
AOP with known sun azimuth and elevation angle stored in a
digital library. The use of a pattern matching technique elimi
nates the need to take into account the effect of the optical
system on the state of polarization detected at each pixel. The
key steps of determining target azimuth using sky polariza
tion compass are the following:
[0114] Record sky polarization images and create a digi
tal library of wide angleAOP and DOP reference images
under clear sky conditions for a range of solar elevation
Record sky AOP and DOP images for the current
known location and time
[0116] Calculate the solar elevation angle for the current
location and time
[0117] Select reference image data that matches the cur
rent solar elevation
Find the best match between the two images
using the pattern matching algorithm
[0119] Calculate current azimuth position of the Sun
relative to the Sun position at the time of the reference
Calculate true North reference
[0121] Determine target azimuth.
[0122] The rotating polarizer sky compass was demon
strated under various sky conditions. Sky images were taken
for a ?xed (standard) orientation of the system using a spot
ting scope pointing at a reference marker located about one
half mile across a canyon from the sky compass equipment.
Polarization images and reference images were compared.
[0125] Altemately, Moxtek, lnc., with of?ces in Orem,
num nanowires onto glass substrates in intricate patterns and
with pitch adequate for use to 300 nm wavelength. They can
produce micro-polarizer arrays matching the pixel pitch of
any focal plane array. These micro-polarizer arrays can be
“glued” onto commercial off-the-shelf focal plane arrays
(FPAs) to cost-effectively convert them to polarizer-on-pixel
sensors. This is the preferred approach Applicants proposes
for near term, low-cost polarizer-on-pixel sensors.
[0126] The military uses compasses to determine the azi
muth of surrounding locations and targets. However, conven
tional magnetic or digital-magnetic compasses are sensitive
to the nearby presence of metals and alloys such as iron.
However, much of military equipment, including vehicles,
armament and weapons include such materials. Hence a non
magnetic compass is highly desired. The sky polarization
compass is insensitive to the presence of magnetically active
materials such as iron. By combining a sky polarization com
pass with GPS in a cell phone or other device containing a
GPS receiver, the device becomes capable of pointing to
known objects, or equally to displaying the azimuth of objects
at which the device is pointed at. This could permit, for
instance, a cell phone to point to the door of the emergency
room, or any other known landmark. One could be guided
more accurately and ef?ciently to a known destination, when
the device can point. A backpacker in the Sierra mountain
range could use a GPS receiver augmented with a sky polar
ization compass to determine which of the jagged points on a
Feb. 12, 2015
US 2015/0042793 A1
ridge, was actually Mount Whitney, and which gully is the
Mountaineer’s Route. All of this is possible, because the
device can now accurately point to items of interest.
The imaging sky polarization compass can be used
Test Results
Actual test results of prototype units con?rm that the
accuracy of Applicants compasses are about an order of mag
nitude better than magnetic compasses. As indicated in the
to determine the direction of zenith. The angle-of-polariza
Background section magnetic compasses under ideal mag
tion pattern in the sky is symmetric about the solar meridian,
the plane containing the observer, the sun and zenith. Within
this plane are two neutral points, the Arago and Babinet
points, which are readily identi?able in the processed images.
netic conditions operate with a measurement error typically
in the range of about 10 to 17 milliradians which results in a
The positions of the two neutral points and the sun from
zenith are all known, so if any of the three are visible, then the
location of zenith is also determined. Hence the requirement
of a separate inclinometer device to determine vertical is
unnecessary. This might be particularly useful on moving
platforms such unmanned aircraft, airplanes, boats, ground
vehicles, missiles, etc, on which an inertial sensor for deter
mination of vertical is dif?cult if not impossible.
locator error of about 50 to 85 meters at a 5 km range.
Applicants’ celestial compasses (with the sun, moon or vis
ible stars at least 45 degrees off zenith (vertical)) operate with
an a measurement error in the range of about 1 to 2 millira
dians which corresponds to a locator error of about 5 to 10
meters at the 5 km range.
[0132] There are many variations to the above speci?c
embodiments of the present invention. Many of these will be
obvious to those skilled in the art. For example in many
embodiments focal plane arrays with only about 350,000
pixels will be adequate. Preferably time should be accurate to
The sky polarization compass uses knowledge of
at least three seconds. For a less expensive system, the inertial
time and approximate position, along with sky polarization
navigation system and the memory-based navigation could be
data to determine a very accurate value for the absolute azi
omitted. In this case the system would in general not be
muth of the device, and thus the absolute azimuth of sur
rounding objects and landmarks. It is possible to use the
mechanism backwards to determine location. This could be
advantageous, for instance, in a GPS denied environment.
The direction of vertical can be determined either from the
sky polarization pattern and the position of the sun or neutral
point, or from an included inclinometer. The time could be
determined using a clock of suf?cient accuracy. The azimuth
of the sun could be determined through the use of the sky
polarization pattern in combination with the use of a conven
operative in cloudy weather. However, local landmarks that
tional magnetic or digital-magnetic compass. With this data,
geographic location can be determined.
It is possible to use the mechanism backwards to
are visible to the camera could be substituted for celestial
objects if the system is properly calibrated using celestial
information to determine the position of the landmarks.
Operators could also install a substitute landmark to use in
this situation. These landmarks could also be used in the full
system with the inertial navigation for re-calibration in the
event of cloudy weather. So the scope of the present invention
should be determined by the appended claims and their legal
What is claimed is:
1. A celestial compass comprising:
A) a camera system adapted for viewing at least portions of
a GPS denied environment. The direction of vertical can be
the sky and comprising:
1) a telecentric ?sheye lens,
determined either from the sky polarization pattern and the
2) a sensor having a focal plane array of at least 350,000
determine time. This could be advantageous, for instance, in
position of the sun or neutral point, or from an included
inclinometer. The azimuth of the sun could be determined
through the use of the sky polarization pattern in combination
with the use of a conventional magnetic or digital-magnetic
compass. If the geographic location is also known from
topography or landmarks, then the time is determined.
Embodiments of the present invention include in
many applications where high accuracy directional equip
ment is needed such as for use in surveying, on cruise ships,
?shing boats and private and commercial aircraft. The inven
tion may also be utilized on robotic vehicles including
unmanned aerial vehicles, unmanned marine vehicles and
unmanned surface vehicles. A particular important use of the
invention will be as a guidance and control feature for robotic
vehicles designed for use in dangerous situations where accu
rate directional information is required. For example, in addi
tion to the telescopic equipment the celestial camera and the
MEMS mirror of the present invention, the robotic surveil
lance vehicle could be equipped with a GPS unit, and a
backup digital magnetic compass and a camera for monitor
pixels, and
B) an inclinometer
C) a processor programmed with a celestial catalog pro
viding known positions at speci?c times of at least one
celestial object and algorithms for automatically calcu
lating target direction information based on the inclina
tion of the system as measured by the inclinometer and
the known positions of at least two celestial objects as
provided by the celestial catalog and as imaged by the
D) a polarization ?lter orpolarization ?lter array adapted to
permit the camera system to measure polarization of
light from a plurality of regions of the sky so as to
determine location of at least one celestial object.
2. The compass as in claim 1 wherein the at least one
celestial object comprises the sun and the moon.
3. The compass as in claim 1 wherein the at least one
celestial object comprises the sun.
4. The compass as in claim 1 wherein the camera system
also comprises a movable ?lter unit comprising an optical
ing the ?eld of view of the telescopic equipment. Communi
?lter and adapted to block portions of sunlight to permit day
cation equipment would be needed for remote control of the
robotic vehicle. Utilizing features described in the embodi
ments described above dangerous targets could be identi?ed
and neutralized. Embodiments could include weapons for
defense or even offence which could be operated remotely.
time and night time operation of the kit with the single camera
5. The celestial compass as in claim 1 wherein the at least
one celestial object is the sun, the moon and a plurality of
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF