need very accurate electronic compass
Dan Major
means to determine direction on a mobile unit to less than 0.1 degree. ideas?
--
Dan Major ma...@tensor.nssl.noaa.gov
Joint Mobile Research Facility
National Severe Storms Laboratory
Norman, OK
John Fields
>
> Like the header says, I need a very accurate electronic compass, or some other
> means to determine direction on a mobile unit to less than 0.1 degree. ideas?
---
GPS.
---
John Fields, Austin Instruments, Inc.
El Presidente Austin, Republic of Texas
"I speak for my company" http://www.austininstruments.com
John Woodgate
<ma...@tensor.nssl.noaa.gov> inimitably wrote:
>Like the header says, I need a very accurate electronic compass, or some other
>means to determine direction on a mobile unit to less than 0.1 degree. ideas?
Well, a magnetic compass wouldn't be as accurate as that anyway, I
suppose, because the Earth's field is not that well-defined, so we must
be talking gyros. Now, there is a solid-state gyro using a laser and two
coils of fibre-optic cable: it was invented at STL at Harlow, IIRC, not
far from here. But I remember little about it. Web search?
--
Regards, John Woodgate, OOO - Own Opinions Only. Phone +44 (0)1268 747839
Fax +44 (0)1268 777124. http://www.jmwa.demon.co.uk Foxhunters suffer from
tallyhosis. PLEASE do not mail copies of newsgroup posts to me.
Mikael Wahlgren
other
> > means to determine direction on a mobile unit to less than 0.1 degree.
ideas?
> GPS.
You are joking, right? A GPS with cm-accuracy can not show its own
orientation at all when standing still. When moving, it can show which
direction it moved, but not at all as accurate as the most basic magnetic
compass, and 0.1 degrees accuracy a GPS can not even dream about in its most
wild dreams.
An ordinary simple consumer marine fluxgate compass gives 0.5 degrees
accuracy, straight from the box and can probably be even more accurate (the
unit they tested in the test of electronic fluxgate compasses when this tech
became popular on the consumer market didn't have enough digits to show
better accuracy), but if it can match 0.1 degree accuracy I don't know.
A gyro is probably the best, but the cost is quite different...
Mikael
Peter Bennett
<jfi...@austininstruments.com> wrote:
>Dan Major wrote:
>>
>> Like the header says, I need a very accurate electronic compass, or some other
>> means to determine direction on a mobile unit to less than 0.1 degree. ideas?
>
>---
>
>GPS.
>
No, I don't think GPS would be suitable (although we don't know the
full details of the application)
GPS will not indicate the direction the vehicle is currently pointing,
but will give a good measurement of the direction of travel over the
past few seconds.
--
Peter Bennett VE7CEI
GPS and NMEA info and programs: http://vancouver-webpages.com/peter/index.html
Newsgroup new user info: http://vancouver-webpages.com/nnq
Lou Boyd
>
> On Tue, 02 Jan 2001 16:13:45 -0600, John Fields
> <jfi...@austininstruments.com> wrote:
>
> >Dan Major wrote:
> >>
> >> Like the header says, I need a very accurate electronic compass, or some other
> >> means to determine direction on a mobile unit to less than 0.1 degree. ideas?
> >
> >---
> >
> >GPS.
> >
>
> No, I don't think GPS would be suitable (although we don't know the
> full details of the application)
>
> GPS will not indicate the direction the vehicle is currently pointing,
> but will give a good measurement of the direction of travel over the
> past few seconds.
Use differential GPS receivers with their antennas mounted at the front
and rear of the vehicle. Knowing the position of each end of the
vehicle will give it's orientation. Not only azimuth but elevation as
well. With three receivers you get the roll axis too.
--
Lou Boyd
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Dan Clingman
I think the easiest is to use three gps antennas. A couple of feet
separation should allow angle determination. One need three because at
the level of accuracy you are asking for what is "level" becomes
important. Movement is going to complicate things but it has been
done. Stanford University did a nice demo with a small plane. IIRC
they were getting 0.1 degrees attitude accuracy
An Inertial Navigation System can also tell which way is north by
measuring earth rotation rate it also does pretty well determining
level. To work one has to update it's position and velocity or it will
drift to never never land.
The best solution for a mobile unit may be a gps and a mid to low grade
INS coupled with a kalmen filter. This would allow the system to work
when a GPS signal was unavalble.
Dan Clingman
Mikael Wahlgren
> and rear of the vehicle. Knowing the position of each end of the
> vehicle will give it's orientation. Not only azimuth but elevation as
Yes, quite clever, but to get the accuracy requested the GPS units have to
have a position accuracy of 3.5 mm if the vehicle is 4 meters long.
Ordinary differential GPS receivers will not do this I believe. You might
be able to achieve this with special reference transmitters, but this makes
the solution semi-immobile...
Mikael
Dan Major
John Woodgate <j...@jmwa.demon.co.uk> wrote:
><92tgs0$ed1$1...@tensor.nssl.noaa.gov>, Dan Major
><ma...@tensor.nssl.noaa.gov> inimitably wrote:
>>Like the header says, I need a very accurate electronic compass, or some other
>>means to determine direction on a mobile unit to less than 0.1 degree. ideas?
>
>Well, a magnetic compass wouldn't be as accurate as that anyway, I
>suppose, because the Earth's field is not that well-defined,
Currently we use fluxgate compasses on some applications. These are combined
with data from a GPS receiver; the heading data is derived from GPS when the
vehicle is in motion, and from the fluxgate when it is stopped. This works
OK for simple meterological data collection and deriving wind speed but not
nearly good enough for what we need now.
> so we must
>be talking gyros. Now, there is a solid-state gyro using a laser and two
>coils of fibre-optic cable: it was invented at STL at Harlow, IIRC, not
>far from here. But I remember little about it. Web search?
Laser ring gyros have been around for almost 30 years now, and work quite
well. They work by having light go on a path through a long fiber which
is wound in a coil. When the coil is rotated about its axis, the light
"thinks" that the path length has increased, and this can be picked up by
an interferometer. What I would like to see is this same design built onto
a single chip. Using currently available MEMS technology, it would not be
difficult to have a folded light path several meters long on the surface
area the size of a postage stamp. Gyors and inertial navigation are one
way to go but I'm looking for a less expensive method.
What this is going to be used for is determining the direction a truck-based
doppler radar system is pointing. Someone had mentioned GPS. "plain" GPS
has a circular error of probability of around 10 meters now that selective
availability has been turned off. The best differential GPS I've found (a
subscription service from trimble or racal/vadic) is only sub-meter at best.
Assuming there is even a COP of one foot, and that there are antennas at each
end of the truck - 30 feet apart. This will still not resolve even a degree
of accuracy (although the accuracy with DGPS gets better over time at a fixed
point). We need this high degree of accuracy in knowing where the truck, and
consequently the radar beam, are pointed is because the resolved scanned volume
at max. distance (30Km) is so small. Additionally, we need to know position
and direction when doing simultanious scans of the same volume from orthoganaly
positioned trucks. One research scenario is to have two trucks at the ends
of the legs of a right triangle with the area of interest at the point of the
90 degree point. Thanks for the ideas though, you guys are on the right
track. I just hope we can afford the solution if I find one!
Keith Wootten
<ma...@tensor.nssl.noaa.gov> writes
>Like the header says, I need a very accurate electronic compass, or some other
>means to determine direction on a mobile unit to less than 0.1 degree. ideas?
Star (including Sun) tracking and time - maybe use a compass for initial
aiming. Possibly.
GPS has been mentioned - maybe you could track GPS (or other?)
satellites with a dish and calculate from their position(s). That could
work whatever the weather.
Or use differential GPS with one receiver on a remote controlled vehicle
(a model 4WD truck?) connected by telemetry and either tracked with a
theodolite or sighted along the direction of interest.
Or say to the local surveyor, 'If you tell me in which direction I am
pointing, I will give you this fine barometer'.
Cheers
--
Keith Wootten
Ara---
use this method to cancel out the ship movement from the Sonar Bathymetry
data. And the pictures come up very clean.. so they must be getting
resonably precise position data...
Al...
"Mikael Wahlgren" <ma...@ridax.se> wrote in message
news:92tt9m$88867$1...@ID-62111.news.dfncis.de...
Rolie Baldock
What is the capabilities of the Laser-Fibreoptic Gyro in this respect?
Would it meet the spec?
On Wed, 3 Jan 2001 00:06:00 +0100, "Mikael Wahlgren" <ma...@ridax.se>
wrote:
--Rolie Baldock. email: <berd_kalamunda@'nospam'techemail.com>
Lasse Langwadt Christensen
I'm not sure the absulute accuracy needs to be that high at all,
As far as I understand it the concept of DGPS is that
you assume that you calculate the error on gps at a fixed
reference position, and then assume that your other gps
can see the same satellites and thus have the same error.
So if you have two GPS recievers at a fixed distance on
a vehicle and all you want is heading my guess is that
you could just say the one of them is the reference.
assuming the error on the absolute position is an
offset, and the same on both recievers, the heading
calculated fron the position of the two recievers
should still be good, the reference point is just moving
around as the error on the absolute position changes
am I totally wrong ?
--Lasse
(+)--------------------------(+)
| Lasse Langwadt Christensen |
| Aalborg, Denmark |
(+)--------------------------(+)
Steve Sousa
news:92uh2b$g0i$1...@tensor.nssl.noaa.gov...
> Laser ring gyros have been around for almost 30 years now, and work quite
> an interferometer. What I would like to see is this same design built
onto
> a single chip. Using currently available MEMS technology, it would not be
Murata (www.murata.com)sells some solid state gyros, they are used in R/C
models for example, but they have good resolution, maybe coupling them to a
fluxgate compass would do the trick?
From the datasheet:
ENV 05 Series
Maximum angular velocity max. –80 — +80 deg/s
Asymmetry CW & CCW — — 3 deg/s
Temperature coefficient –10 ~ 60°C — — ±5 %FS Scale factor –30 ~ 80°C — —
±10
Drift –30 ~ 80°C — — 9 deg/s
Start up Measure Vo after 5 seconds — — ±1 deg/s/10 min.
Noise level 12kHz noise — — 20 mVrms
Linearity in the maximum angular velocity range — — 0.5 %FS
Response Phase delay: 90deg — 12 — Hz
Operating Temperature Range Topr –30 — 80 °C
Storage Temperature Range Tstg –40 — 85 °C
Weight ——50gr
or
ENC Series
Angular velocity range max. +300 deg/s
Linearity in the maximum angular velocity range –5 — +5 %FS
Response Phase delay: 90deg DC p 50 Hz
Operating Temperature Range Topr –5 — +75 °C
Storage Temperature Range Tstg –30 — +85 °C
Weight — — 1.0 g
--
Steve Sousa
Remove the anti-spam g from the address to reply
Onestone
flux gate magnetometer. I doubt you will achieve the accuracy with a dual axis
type since the error due to tilt can be as high as 2.7x the tilt angle with a 2
axis compass. Thus you would need to be able to guarantee the compass was level
to within 0.03 degrees. Highly unlikely on a mobile unit.
Al
John Woodgate
inimitably wrote:
>It is possible to achieve better than 0.1 degree of accuracy using a three axis
>flux gate magnetometer.
But to do that you need to know the local direction of the Earth's field
also to better than 0.1 degree. I don't think that's possible unless all
iron is a LONG way away.
Onestone
their US agent. This makes a good starting point, and gives some insights into
auto-calibration techniques. I combined this with some of Don lancasters notes on
flux gate sensor materials and built my own 3D sensors. The prototypes I built a few
years ago seemed to operate well all over australia and in the UK. It never went
elsewhere though.
As further input to this topic Irvine Sensors have been trying to perfect a 2.4mm
square solid state gyro (coreolis effect) for some time and claim to be near to
release.
James A. Kuzdrall
We have been evaluating some attitude sensing equipment at Ft.
Hood. Trimble, SRI, and Honeywell all make attitude sensing instruments
based on GPS. They measure the phase difference of the carriers
arriving from the GPS satellites to determine orientation. The
advertized accuracy is .3 degrees RMS for a 1 meter base line, but I
found "noise" variations can be as large as 1.5 degrees. See the
Trimble's Tans-Vector equipment for details.
Crossbow makes combined 3-axis flux-gate compass and MEMS
accelerometer with .5 degree accuracy. The model 543 which I am
currently testing does what they say, pretty much. The magnitude of
error from nearby iron, particularly magnetized iron, is a concern.
But, the unit is small and can be battery operated - at least for a few
hours.
If you want to get fancy, get some information on the new POSNAV
system used for the M1A2 tanks. I was told who made it, but I forgot.
There is a commercial version available. It uses sensitive
accelerometers to sense the east-to-west motion of the earth. Any time
the vehicle is still for 15 minutes, it recalibrates. Gyros are used in
between calibrations. The repeatability was better than .1 degrees. We
didn't test the absolute accuracy.
James A. Kuzdrall
John Woodgate
inimitably wrote:
>We have been evaluating some attitude sensing equipment at Ft.
>Hood.
If you applied it to some people on this newsgroup (maybe ANY
newsgroup!) it would turn purple and explode. (;-)
[big snip]
> It uses sensitive
>accelerometers to sense the east-to-west motion of the earth.
But the Earth doesn't accelerate in that (tangential) direction. The
acceleration is radial. So, what you say (maybe to avoid using
'centrifugal'?) is true but potentially somewhat misleading.
Doesn't that mean that you need to know the local value of g
(acceleration due to gravity) pretty accurately? The acceleration due to
rotation at the Equator appears to me to be about (minus) 3.4x10^-3
times that due to gravity.
James A. Kuzdrall
John Woodgate wrote:
> [big snip]
> > It uses sensitive
> >accelerometers to sense the east-to-west motion of the earth.
>
> But the Earth doesn't accelerate in that (tangential) direction. The
> acceleration is radial. So, what you say (maybe to avoid using
> 'centrifugal'?) is true but potentially somewhat misleading.
I was just outlining the general principle, knowing that the astute
readers of this news group would fill in all the details from the
slightest hint.
Quite correctly, only gravitational and centrifugal forces (both
radial) affect a non-moving body on the surface of a rotating sphere.
If the body moves, we have an additional force, the Coriolis force to
consider. But, the vehicle is stationary for the calibration.
If the body we monitor has angular momentum, however, another force
appears. Say the gyroscope is anchored to the vehicle body. It wants
to maintain its plane of rotation in space. The vehicle changes its
orientation in space as the earth rotates. Sense the force needed to
reorient the gyroscope's rotational plane to find the east-west line.
The force is very small, apparently they need to integrate for 15
minutes. Also keep in mind that the tank can be in any orientation, so
the force has to be the vector sum of 3 axis.
>
> Doesn't that mean that you need to know the local value of g
> (acceleration due to gravity) pretty accurately? The acceleration due to
> rotation at the Equator appears to me to be about (minus) 3.4x10^-3
> times that due to gravity.
Gravity is far from constant, as you must know, because the earth is
an ellipsoid. A plum bob can be off as much as .1 degrees. One fellow
recently calculated that the Mississippi River actually flows "up" on
its journey from Minnesota to Louisiana.
James A. Kuzdrall