The doppler effect shows a change in frequency due to a velocity of
observer and vehicle being non-zero, high pitch sound when
approaching, low when receeding etc. Astronomers often use this to
detect velocity of planets, which are moving relative to themselves.
CAN the doppler effect be used if the emitter and the receiver are
co-located in the same moving device? Could we detect velocity in a
black box, with no reference to outside points? Could we alternatively
emit radio waves / sound etc from corners of a box, then measure the
frequency from the centre of the box, to determine variations as
vectors which could be resolved to calculate our true box velocity
vector?
Is it the case that there can and would be no change in the frequency
as the closed moving environment is static, so all corners will report
the same frequency to the centrally mounted sensor, OR does the speed
of light dictate that this does not occur lest some waves then exceed
the speed of light etc. Is there some warping of time involved such
that the frequency is detected as static but the time itself changes
or something complicated?
PART TWO of this question is what method of detecting doppler effects
could be used to detect velocity of a vehicle, which at around 50km/h
is a little slow relative to the speed of light - IE we aim to detect
a very small variation in frequency, so we prefer to use sound waves
over light, or is this the reverse? What medium or frequency range is
best to detect this phenomena, IF it is detectible at all?
PART THREE of this question, is how can we detect velocity in a black
box with no specific reference to static observers, in a way that can
be corrected. An example may be via Doppler effect, but assuming this
did not work, then what else might?
We can't really use just accelerometers, because they drift and will
result in a vehicle appearing to be driving at 1000km/h after long
enough! |
Clarification of Question by
chrispeaks-ga
on
26 Oct 2003 19:04 PST
This question is related to several others I have placed, perhaps you
may wish to consider them if your answere relates to more than one -
IE it may be preferable to answer two or three together by one
researcher, or use any posted answers to one question as a starting
block to others.
Below is the list of my open questions
GIS storage method / clever compression shows angle & radius for
curves?
Detecting velocity via with correction
Accelerometers to detect angular rotation or velocity
Velocity detection via doppler effect with observer situated ON moving
object!
Compass heading determined with tilt correction algorithm (negative
pitch error)
|
Request for Question Clarification by
hedgie-ga
on
27 Oct 2003 10:37 PST
Hi chris
You have posted interesting and sensible questions,
and GA researchers love interesting and sensible questions.
However, few aspects can make your a 'wallflower',
even though she is quite attractive.
1) It is priced low.
While we (well, some of us) know DLVs (doppler laser velocitometers),
even may be used them, we still would have to spend time to
research the issue. How much you can expect $5.00??.
2) You pack in notes, like
'.. We can't really use just accelerometers, because they drift and will
result in a vehicle appearing to be driving at 1000km/h after long
enough! ..
which may require a long clarification dialogs (series of RFCs) and that
combined with point 1)may deter some researchers from even trying.
So,
It may be a good idea to make each question less open-ended and study
the pricing guideline and 'how to ask a question' guides.
A short comment on two of your points:
1)"if the emitter and the receiver are
co-located in the same moving device.."
Yes. You always measure relative velocity. Relativity does apply.
2)" so we prefer to use sound waves.."
Hardly. Light (laser beam) is (almost) always better, for slow and
fast, high or low rate of change....
They use Doppler Radars to map hurricanes.
Could they do that with sound waves?
http://www.arhs.net/News/Weather/Imagery/
3) I myself am leaving for 2 week trip soon, away from any computer,
so please
do not get discouraged if there is no reaction to your eventual
clarification. This question is open to all.
hedgie
|
Clarification of Question by
chrispeaks-ga
on
27 Oct 2003 13:16 PST
Your comment on doppler laser velocitometers was precisely the type of
answer I was seeking, so you have already answered it in part with
your clarification request (including your light frequency
nomination). I was not really seeking huge volumes of information, but
more key points which may be 1) Yes but you need to measure time
changes, 2) Light is best and 3) DLV is the only real answer.
I accept your comment about pricing, and made a change, and while it
is not much, it is at least double now. The project I am working on is
a self-funded shoestring budget (which is difficult as a single father
with sole custody of a 3yo girl), and often some issues are personal
interest. Considering several questions were related, I did think they
could be grouped together to increase the overall price. I rather
suspected there was no easy way to detect velocity within a black box,
and could never afford a DLV at any rate. Quantum mechancics and
relativity are not my best subjects either:-)
Still, I have identified two ways that seem possible already,
including measuring the rate of altitude change via altimeter
(assuming device on road with a gradient), then translating that
vector to the road speed by dividing by sin of the gradient angle as
measured from inclinometer / accelerometer
Another method is on curves where I match the radial acceleration to
the angular change detected from a compass or gyro
I figure there may be few real options for a spacecraft, but the
real-world scenario of a vehicle / earth gravity / atmosphere etc
lends itself to other clever means of velocity detection... perhaps as
a broadband carrier wave detection using common FM radio stations to
see a Doppler effect, ala corrected DGPS units... and the list goes
on. I am not seeking a pure physics solution so much as a possible
solution, that may use real world phenomena, when the physics approach
may not be comfortable using FM radio transmitters unless they were
part of the tracking 'system' (IE built and known), instead of
existing infrastructure that can be 'borrowed' etc.
Given I accept you are already half way to the answer with your
comment, and I have doubled the price (I will do so with others too),
I hope this will be a bit of a stimulus for further words of wisdom,
and I am always open to suggestion.
Thanks!
|
Request for Question Clarification by
hedgie-ga
on
28 Oct 2003 08:16 PST
Chris
thanks for the clarification.
It does make a difference, as good answer
to a complex question is often a cooperative project.
As I said, I am in the middle of packing for a trip. We have several
researchers who are qualified to look into this, and I hope one of
them will
pick up this question. If not, I may look at at when I come back.
Within the limits of you feel comfortable to reveal, it may help to
say more about the application
like: are you tracking cars (in line of sight like police radar)?
In wide area (like stolen car trackers)
someting like a gyroscope for airplanes (black box role is not clear,
is there an object in the center of the box which is moving relative
to walls)?
How comfortable are you with design and operation of interferometers,
like:
://www.google.com/search?hl=en&ie=ISO-8859-2&q=Farber+Interferometer&btnG=Google+Search
|
Request for Question Clarification by
hedgie-ga
on
28 Oct 2003 08:20 PST
Correction
Sorry about miss-pelling. I should not work ahen in a hurry.
But the link I need to correct
I mean search term: Fabry Interferometer (not faber), like this:
://www.google.com/search?hl=en&inlang=pl&lr=lang_cs%7Clang_en&ie=ISO-8859-2&safe=off&q=Fabry+Interferometer&btnG=Google+Search
|
Clarification of Question by
chrispeaks-ga
on
28 Oct 2003 15:59 PST
My project aim is to design a non-GPS based, non-triangulation based
terrestrial tracking system. I am aware of the drift and reference
limitations of dead reckonning, and am trying to look at novel
solutions. As such the working space is primarily ON LAND, ON ROADS,
and generally within 'built up areas' which may have GSM or FM radio
broadcasts etc (Which are much stronger than GPS and can be received
internally) though I prefer not to have any form of RF solution at all
due to power consumption and antenna issues.
I have seen a UK based product that provides an accurate speedometer
using nothing more than a 2-axis accelerometer which is hard mounted
to the vehicle but not connected to the wheels for rotation detection
(Does require a cooeficient for the suspension crouch factor when
stopping or starting, which is particular to each vehicle, but then it
claims to provide speed within 5%). I am curious as to how this is
achieved, and am trying to replicate or better this approach with a
non-mounted design, such as a black box that can be orientated in any
direction, yet determine velocity and hence position.
Your simple comments of LDV and Fabry Interferometer are exactly the
type of answers I am seeking. I am capable of some research, but often
the hard part is to know where to start, and what I should look for,
and these two word 'keys' you have supplied as search terms will keep
me busy for some time yet. Given your impressive answer to my last
gravity question, I hope you are able to comment further on your
return. Enjoy your trip! - Chris
|
Clarification of Question by
chrispeaks-ga
on
28 Oct 2003 16:41 PST
To further clarify, I aim to design a remote tracking method, that
could be used to locate stolen cars. In fact I want to see if I can
track any object at all, such as a bag inside the car, which would
normally be shielded from faint GPS signals, or have no GSM signals if
out of coverage area etc. In other words it is not a line of sight
application, but a stand-alone, independant system that operates
exactly like GPS but without the 'reception' and power limitations,
and does not depend solely on third party networks that can be
disrupted in war or fail etc. I know of intertial dead reconning
systems that will track about 15 minutes of driving (destination error
perhaps 200m at this point), but much past this the errors multiply
and velocity drift throws out the position by an exponentially
increasing error - so I am trying to replicate this system myself AND
correct the error problem in a novel way. It's an ambitious project
for an unfunded individual without a science background, but it is an
intriguing challenge none the less, in the spirit of science
discovery!
|
