I think one of the things you're going to want is a bit of hysterisis. My T2 is set up to switch from profile 1 to profile 2 when I'm going faster than 10mph. It doesn't switch back until I'm under 1mph.
What ends up happening is that drift occurs under around 10mph. But when you start actually moving faster than 10mph (for real) it'll stick in profile 2, until you're completely stopped. In practice I think I'd want to raise the value up to about 15 or 20mph or so if drift were more of a problem. -Dan n7nmd On Nov 2, 2009, at 0:43, "ki6tsf" <[email protected]> wrote: > Hello, > > I posted a pict showing GPS drift at > http://groups.yahoo.com/group/tracker2/photos/album/411553253/pic/812065518/view?picmode=&mode=tn&order=ordinal&start=1&count=20&dir=asc > > That's not very nice for a portable or mobile digipeater. > > A Nuvi 350 is occasionally on port A and the GlobalSat BR-355 puck > GPS (no display) is permanently connected to port B. The GlobalSat > GPS is causing the drift on the picture. > > I tried to set the satellite elevation threshold to a higher value > (12 degrees instead of the default value of 5) to eliminate near > horizon satellites but that did not help at all. It looks like > reflections/multipathing is really bad in some areas and this type > of GPS does not deal with them very well. > > Profile 2 is activated when the GPS is at stopped position, with a > very low transmit rate (it does that partly because the OT2 is used > as a TEMP digi to allow nearby HT's such as VX-8's to get/send APRS > messages through it). Profile 1 is activated when the GPS is moving > > 8 mph. Since profile switching works well, it would be great if > it can be used to decide when to ignore positions because they are > likely caused by GPS drift. > > I was wondering if using the AUTOSAVE followed by POSITION X,Y > commands when entering Profile 2 would let the OT2 save the last > known fix rather than X,Y if X,Y are given special values like 0,0 > or illegal values. There doesn't seem to be any way of getting the > latest position in scripting. If the above works, the POSITION GPS > command should be executed when entering Profile 1 to resume reading > the real position. > > Another option would be to use a 2 axis accelerometer to find out > when the GPS is physically moving and when it stops but the problem > of saving/using the last good fix on the OT2 remains. > > Would it be possible for a script to power on that rogue GPS only > after physical motion is detected, and power it off when no motion > occured for a certain time? I am not sure how to use the extra I/O > pins or the 2-Wire interface from the scripting language of the OT2. > Would that work? > > Bernard KI6TSF > > --- In [email protected], "Tom" <tmcs...@...> wrote: >> >> Hello Bernard, >> Sorry - I replied to one on the subsequent replies to your original >> question and did not see the information below. >> (The earlier messages of the thread were in my spam folder - go >> figure.) >> >> With problems like this, I like to see the signal strength and >> satellite coverage data from the GPS. >> For a displayless GPS (Is that the case here?) you can use a laptop >> running a GPS utility like VisualGPS. >> It's free and you can download it from visualgps.net >> >> Remember too that you can have good coverage and a very accurate >> position at one point in time and then 2 hours later you may see a >> degradation of position as the GPS acquires different (maybe >> fewer) satellites with the rotation of the earth. >> >> best of luck to you, >> Tom, W3TMC >> >> >> >> ----- Original Message ----- >> From: ki6tsf >> To: [email protected] >> Sent: Saturday, October 31, 2009 5:21 PM >> Subject: [tracker2] How to deal with GPS drift with OT2? >> >> >> Hello, >> >> I got a brand new GlobalSat BR-355 SiRF GPS which I use on >> the secondary port of the OT2. Unfortunately, I discovered >> that the position is drifting severely when the car is parked >> for several hours at the same place (covered or uncovered), >> sometimes up to 300ft. This may be due to reflections. >> >> I tried with WAAS enabled and disabled, >> and with Static Navigation enabled and disabled and get the >> same problem. Static Navigation is supposed to fix this by >> correcting the coordinates if the calculated position moves >> slower than 1.6 m/s. It does not work: calculated moves >> are grater than that. >> >> I contacted the manufacturer and they said >> the best option is to have the software deal with it. >> >> Does anybody sees how to eliminate drift in the OT2? >> So far I was unable to find out how scripting could help. >> >> Thanks, >> >> Bernard KI6TSF >> > > > > > ------------------------------------ > > Yahoo! Groups Links > > >
