| |
ZL3AI > APRDIG 21.09.03 09:03l 268 Lines 9725 Bytes #999 (0) @ WW
BID : 1949-ZL3AI
Read: GUEST
Subj: TAPR Digest, Sep 18, 3/5
Path: DB0FHN<DB0RGB<OK0PPL<DB0RES<ON0BEL<SP7MGD<7M3TJZ<LZ3NP<W4JAX<VK4TUB<
ZL2BAU<ZL3VML
Sent: 030920/1351Z @:ZL3VML.#80.NZL.OC #:8729 [Chch-NZ] FBB7.00i $:1949-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: RE: sar/aprs software
From: "Curt Mills, WE7U" <hacker@tc.fluke.com>
Date: Thu, 18 Sep 2003 09:41:35 -0700 (PDT)
X-Message-Number: 16
On Thu, 18 Sep 2003, David VanHorn wrote:
>you know... if you're tracking your searchers, you could draw say a 6' or
>10' or whatever's appropriate for terrain and conditions, circle around
>their tracks, and then the missing party has to be in the unpainted area,
>of course this assumes that he's not moving..
>
>works for crash sites anyway, and lost people with half a clue.
>
>kind of like bob's ht based df..
Yep. You're starting to get into search management theory, and we're
headed in that direction, as soon as I learn more about it!
We're adding probability min/max circles to Xastir this week. They're
useful for concentrating a search within the probable travel distance for
specified types of subjects over different terrain (given collected
statistics from previous searches of similar types, which are in the search
management manuals). It's a good start anyway.
There are more advanced sorts of things in the books, some hint at what you
describe, and of course the science is evolving. It'll be a real challenge
keeping up with the latest techniques in search management.
--
Curt Mills, WE7U hacker_NO_SPAM_@tc.fluke.com
Senior Methods Engineer/SysAdmin
"Lotto: A tax on people who are bad at math!"
"Windows: Microsoft's tax on computer illiterates!" -- WE7U
"The world DOES revolve around me: I picked the coordinate system!"
----------------------------------------------------------------------
Subject: RE: sar/aprs software
From: "Scott Miller" <scott@opentrac.org>
Date: Thu, 18 Sep 2003 10:01:52 -0700
X-Message-Number: 17
Let me know when this is working - I get CVS updates every day, but I rarely
know where to go look for new features.
ArcView's military analyst features have some movment analysis functions...
something like that would be pretty cool to help predict a victim's movement
over given terrain. If they're hiking in the bottom of a canyon, chances
are they're not going straight up the canyon walls.
I've got a SAR presentation coming up later this month, and that'd be a nice
thing to demo.
Scott
N1VG
----------------------------------------------------------------------
Subject: RE: sar/aprs software
From: David VanHorn <dvanhorn@cedar.net>
Date: Thu, 18 Sep 2003 12:03:21 -0500
X-Message-Number: 18
At 09:41 AM 9/18/2003 -0700, Curt Mills, WE7U wrote:
>On Thu, 18 Sep 2003, David VanHorn wrote:
>
>>you know... if you're tracking your searchers, you could draw say a 6' or
>>10' or whatever's appropriate for terrain and conditions, circle around
>>their tracks, and then the missing party has to be in the unpainted area,
>>of course this assumes that he's not moving..
>>
>>works for crash sites anyway, and lost people with half a clue.
>>
>>kind of like bob's ht based df..
>
>Yep. You're starting to get into search management theory, and
>we're headed in that direction, as soon as I learn more about it!
interesting.
one thing i drilled into my kids, is that when you're lost, find a good
spot, and /stay there/. finding a wandering soul is like trying to hit the
lottery with random 'quick-pick' tickets.
>There are more advanced sorts of things in the books, some hint at
>what you describe, and of course the science is evolving. It'll be
>a real challenge keeping up with the latest techniques in search
>management.
i bet a lot of people have been missed, because someone thought an area had
been covered, when it really hadn't, in the pre-gps era.
you could even do this without realtime aprs, dumping people's gps track
log, and parsing that into the map.
realtime is of course a lot better.
----------------------------------------------------------------------
Subject: Re: RELAY,WIDE,WIDE
From: Gonzales Andy <agonzales@arc.nasa.gov>
Date: Thu, 18 Sep 2003 10:12:48 -0700
X-Message-Number: 19
You specify what your destination station is and teh software
determines the most efficient path for you ?
>>I'm personally all for more centralized control, but it still
>>needs to work without control or coordination.
>
>I was thinking more along the lines of a *de*centralized, adaptive pathing
>algorithm of some sort, relying on real-time observations of the network.
>
>Again, the example of listening for my own packet to be digipeated -- if I
>don't hear it, then I conclude that it didn't happen. In which case my
>station might automatically alter my path from WIDE to RELAY,WIDE. Perhaps
>a way to go the other way, too, and thereby truncate unnecessarily long
>paths (e.g., WIDE7-7) -- but based on real-time observations of the network,
>not simply convention.
>
>-R
--
Andy Gonzales
NASA / Ames Research Center
Moffett Field, CA 94035-1000
Bldg. N-244, Rm. 240
MS 244-19
de KG6RWO
agonzales@arc.nasa.gov
650-604-0309
----------------------------------------------------------------------
Subject: Re: RELAY,WIDE,WIDE
From: Rich@Osman.com
Date: Thu, 18 Sep 2003 10:26:39 -0700 (PDT)
X-Message-Number: 20
On Thu, 18 Sep 2003 10:06:16 -0400 (EDT), "Daniel H. Brown" wrote:
>On Thu, 18 Sep 2003, Rochte, Robert wrote:
>
>>Date: Thu, 18 Sep 2003 09:23:46 -0400
>>From: "Rochte, Robert" <rrochte@gpacademy.org>
>kind
>>-tracker, home station, digi) transmits packets and then intelligently
>>*listens* to what happens on-the-air, adapting its own path as needed.
>
>
>I think this would be a great idea -- but rather than having the software
>automatically, by default, changing settings, perhaps it could provide
>the user a "recommended" path, based on data collected off the air,
>over a period of time.
I've suggested this (twice) and been met with a deafaning silence. I'm not
sure why it stirred NO discussion.
mailto:Rich@Osman.com http://www.rich.osman.com
Rich Osman; POB 93167; Southlake, TX 76092 (Near DFW
Airport) ARS: N1OZ
----------------------------------------------------------------------
Subject: RE: sar/aprs software
From: "Scott Miller" <scott@opentrac.org>
Date: Thu, 18 Sep 2003 10:37:42 -0700
X-Message-Number: 21
>one thing i drilled into my kids, is that when you're lost, find a good
>spot, and /stay there/. finding a wandering soul is like trying to hit the
>lottery with random 'quick-pick' tickets.
As opposed to trying to hit the lottery with orderly, systematic number
choices? =]
>i bet a lot of people have been missed, because someone thought an area
>had been covered, when it really hadn't, in the pre-gps era.
Yeah, I can say from personal experience it's not much fun having the
victim turn up in an area you'd already searched and reported clear. Of
course, that guy was moving, so that's not saying much... but still, it
sucked.
Scott
N1VG
----------------------------------------------------------------------
Subject: RE: sar/aprs software
From: David VanHorn <dvanhorn@cedar.net>
Date: Thu, 18 Sep 2003 13:00:24 -0500
X-Message-Number: 22
At 10:37 AM 9/18/2003 -0700, Scott Miller wrote:
>>one thing i drilled into my kids, is that when you're lost, find a good
>>spot, and /stay there/. finding a wandering soul is like trying to hit the
>>lottery with random 'quick-pick' tickets.
>
>As opposed to trying to hit the lottery with orderly, systematic number
>choices? =]
fixed number.
it's like trying to find your wife in the mall, when she's 'trying to find
you' /aka shopping/..
try it with dice, pick a number, and roll one die 30 times, you'll find
about 5 hits. now roll both dice, and record the times that both numbers
are the same.. about 1 in 25. you might have to roll more than 30x to get
a hit.
>>i bet a lot of people have been missed, because someone thought an area
>>had been covered, when it really hadn't, in the pre-gps era.
>
>Yeah, I can say from personal experience it's not much fun having the
>victim turn up in an area you'd already searched and reported clear. Of
>course, that guy was moving, so that's not saying much... but still, it
>sucked.
i know the feeling. i've never had a successful cpr..
in my defence, all my victims were pretty well done before i got to them.
we rolled within three minutes of call, pretty good for an unpaid volunteer
department, but that ends up being about 10 minutes or more between the 911
call, and boots on the ground. 10 minutes without breathing is very likely
to extend to infinity.
----------------------------------------------------------------------
Subject: Re: RELAY,WIDE,WIDE
From: "Bill Vodall - WA7NWP" <wa7nwp@jnos.org>
Date: Thu, 18 Sep 2003 11:08:08 -0700
X-Message-Number: 23
>>>-tracker, home station, digi) transmits packets and then intelligently
>>>*listens* to what happens on-the-air, adapting its own path as needed.
>>I think this would be a great idea -- but rather than having the software
>>automatically, by default, changing settings, perhaps it could provide
>>the user a "recommended" path, based on data collected off the air,
>>over a period of time.
>
>I've suggested this (twice) and been met with a deafaning silence. I'm not
>sure why it stirred NO discussion.
It's a good concept. The problem is that its hundreds of hours of
programming time, per APRS application, to make a small change in an
existing system that's already working. Even worse, the programmers
wouldn't receive any real return on their effort invested. This is one
of the disadvantages to low cost software.
Personally, I'm still in favor of the pseudo destination scheme of using
aliases of "LOCAL", "REGION", "WORLD" and "IGATE".
73,
Bill - WA7NWP
----------------------------------------------------------------------
Read previous mail | Read next mail
| |
