[NSRCA-discussion] Could this allow automated scoring?
John Gayer
jgghome at comcast.net
Sat May 9 13:02:50 AKDT 2015
Yeah Keith, I read that.
What is missing from the article is the number of satellites required to
achieve that accuracy and what percentage of the time that minimum
configuration exists. Some of the problems they have would be diminished
in our openair environment(multipath, for example). They don't mention
the update rate although the video seems to indicate it is pretty good.
Of course, most of the software could be developed now with the existing
sensor technology. If the sensor technology isn't good enough, just wait
six months for the virtual reality guys to drive it forward since they
need a high update rate as well.
Start with the coaching analysis software which is relatively easy.
For auto-judging I still see the need for one judge to be a zero judge
and score points on stall turns, snaps and spins. Perhaps also to mark
the termination of each maneuver. The software involved is fairly
complex as the maneuver size and centering are undefined until the
maneuver is complete and in absolute terms the downgrade criteria are
totally inadequate.
John
On 5/9/2015 12:39 PM, Keith Hoard wrote:
> If you follow the link, some egg-heads have "broken the code" for
> centimeter accuracy GPS using existing cheap antennas using software
> only.
>
> -Keith Hoard
> -Sent from my Windows Phone
> ------------------------------------------------------------------------
> From: John Gayer via NSRCA-discussion
> <mailto:nsrca-discussion at lists.nsrca.org>
> Sent: 5/9/2015 12:47
> To: Daniel Underkofler <mailto:underdw at gmail.com>; General pattern
> discussion <mailto:nsrca-discussion at lists.nsrca.org>
> Subject: Re: [NSRCA-discussion] Could this allow automated scoring?
>
> Ground based GPS augmentation systems exist and have been tested.
> However, I haven't heard about anything in our price range. Typically
> in an airport environment, the ground based transmitter is
> broadcasting corrections to the satellite signals rather than an
> imitation satellite signal.
> Standard micro GPS probably wouldn't know what to do with this signal.
> Also the ground-based xmitter is not sized/priced for us.
> However , it may not be necessary to go that route. If we could
> improve on the 3 meter accuracy a bit and the 10hz update rate and
> expand the storage capacity....
>
> But rather than auto-judging, setting up a display showing the plan
> view and vertical view graphically, perhaps with color shading to
> indicate deviation from roll angle = 0 in realtime would be an
> incredible advance in practice/coaching. Think about it.
> In a square on corner, you would see directly the effect of the top
> corner roll attitude being off 5 degrees and where that took the third
> leg.
>
> John
>
> On 5/9/2015 7:07 AM, Daniel Underkofler via NSRCA-discussion wrote:
>> It's great to see technology like this approaching for at least
>> partial automated judging.
>>
>> Seems that resolving the altitude component adequately is the largest
>> challenge. To address this, I had envisioned a local set of 3
>> "beacons", each emitting GPS-like signals that are received and
>> processed by a transponder in the model. Yeah - expensive and even
>> more out in the future.
>>
>> In the short term, using high resolution and update GPS to resolve XY
>> location might be useful for partial judging. Human judges would
>> still be responsible for the majority of the scoring, but a
>> "correction factor" for XY errors might be factored in. For
>> instance, it is terribly hard to judge the geometry of the horizontal
>> triangle in F15, but given good XY data, a portion of the score could
>> be addressed. Distance and parallelism to the runway could also be
>> determined and factored in (somehow).
>>
>> Dan
>>
>> On Fri, May 8, 2015 at 9:33 PM, John Gayer via NSRCA-discussion
>> <nsrca-discussion at lists.nsrca.org
>> <mailto:nsrca-discussion at lists.nsrca.org>> wrote:
>>
>> The judges are expected to consider the track of the plane in an
>> earth coordinate system. For example the judges are supposed to
>> consider that a 45 degree climb appears steeper at the end of the
>> box than it does in the center. So the track should be analyzed
>> by the judges in an earth coordinate system and the track of the
>> model rather than attitude.
>>
>> The location of the pilot is actually irrelevant but the location
>> of the box is clearly of critical importance. Knowing the actual
>> location of the plane with a fairly quick update rate is
>> critical. At 120 feet/sec(approx 80 mph), horizontal position is
>> no problem. At this speed you need an update rate of about 80
>> milliseconds to get a data point about every 10 feet. Not sure
>> how good the GPS update rates are these days but this should be
>> achievable. IF the altitude update rate is too slow then augment
>> with an altimeter. Since we are downloading this data from the
>> model, roll attitude could be transmitted as well for wings level.
>>
>> We would still need a zero judge for snaps, spins and stall turns
>> and to establish when the pilot is lost in a maneuver. It might
>> be necessary to have an "advance to the next maneuver" switch for
>> that case.
>>
>> It would actually be a fascinating project to download
>> accelerations, roll, yaw and pitch rates for analysis while
>> observing snaps visually to see if standards could be established
>> about what is a snap and what isn't.
>>
>> John
>>
>> On 5/8/2015 1:24 PM, Peter Vogel via NSRCA-discussion wrote:
>>> Not by itself. Judging/Scoring is from the perspective of the
>>> pilot + judges, not from the perspective of the plane. (As
>>> opposed to IAC, where it's from the perspective of the pilot in
>>> the plane). For it to be done right you'd need to know the
>>> precise location of the plane and the precise location of the
>>> pilot, then calculate the geometry of the plane's movements
>>> relative to the perspective of the pilot's location.
>>>
>>> The SoloShot 2 automatic cameraman does close to what we want in
>>> tracking the movement of the plane relative to a ground station,
>>> but from that you'd still need to calculate geometries, and
>>> wings level, etc.
>>>
>>> Peter+
>>>
>>> On Fri, May 8, 2015 at 11:44 AM, Keith Hoard via
>>> NSRCA-discussion <nsrca-discussion at lists.nsrca.org
>>> <mailto:nsrca-discussion at lists.nsrca.org>> wrote:
>>>
>>> Cheap, Centimeter accuracy GPS <http://goo.gl/mZg6RG>.
>>>
>>> No more complaints about biased judging?
>>>
>>> -Keith Hoard
>>>
>>> klhoard at hotmail.com <mailto:klhoard at hotmail.com>
>>>
>>>
>>> _______________________________________________
>>> NSRCA-discussion mailing list
>>> NSRCA-discussion at lists.nsrca.org
>>> <mailto:NSRCA-discussion at lists.nsrca.org>
>>> http://lists.nsrca.org/mailman/listinfo/nsrca-discussion
>>>
>>>
>>>
>>>
>>> --
>>> Director, Fixed Wing Flight Training
>>> Santa Clara County Model Aircraft Skypark
>>> Associate Vice President, Academy of Model Aeronautics District X
>>>
>>>
>>>
>>> _______________________________________________
>>> NSRCA-discussion mailing list
>>> NSRCA-discussion at lists.nsrca.org <mailto:NSRCA-discussion at lists.nsrca.org>
>>> http://lists.nsrca.org/mailman/listinfo/nsrca-discussion
>>
>>
>> _______________________________________________
>> NSRCA-discussion mailing list
>> NSRCA-discussion at lists.nsrca.org
>> <mailto:NSRCA-discussion at lists.nsrca.org>
>> http://lists.nsrca.org/mailman/listinfo/nsrca-discussion
>>
>>
>>
>>
>> _______________________________________________
>> NSRCA-discussion mailing list
>> NSRCA-discussion at lists.nsrca.org <mailto:NSRCA-discussion at lists.nsrca.org>
>> http://lists.nsrca.org/mailman/listinfo/nsrca-discussion
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nsrca.org/pipermail/nsrca-discussion/attachments/20150509/97906376/attachment.html>
More information about the NSRCA-discussion
mailing list