If it works as hoped, anyway. Anyone else reading the updates on the Triumph LS test on Javad's website?
The last couple of responses elude to a $10,000 challenge if anyone can defeat the new auto-verify feature by storing a bad fix!
Pretty cool move, I think!
How far off does it have to be, to be considered "bad"?
N
I have caught a half dozen or so bad inits in the last 2 months. Not very impressive for a 2 crew operation. If this works I want 4 of them...
> Anyone else reading the updates on the Triumph LS test on Javad's website?
Which update? I check the page fairly often, but am not seeing anything on bad inits.
I don't know, I'm sure more details will follow. From Javad's comment, though, it seems as if any "bad initialization" counts.
Aren't all bad initializations the result of incorrect integer ambiguity solution from multipath?
I wont be impressed, though, if the "Auto-Verification" just chokes the unit down to where it won't work with slight obstructions.
I can "feel" my way around under canopy with a GPS if it will fix. You just gotta know the tricks: Dump the unit, tape over 10', shoot it twice, and so on...
A Javad:
In our next advertisement we are challenging with a $10,000 reward if one can show one bad fix with TRIUMPH-LS, provided they use the auto-verify feature that we are just releasing. Please see the attached document.
We will work on all your suggestions.
If it's important it's gotta be shot twice, with TWO DIFFERENT INITIALIZATIONS. I actually have this printed out and taped up where the crews can see.
If you can only initialize once I won't use the point. You've just got to get the gun out sometimes.
Out of curiosity what kind of units are they using?
I've used Leica for fourteen years, both 530 and 1230, usually very heavy usage, and in that time I've had maybe five bad fixes on individual points. In other words, actual bad fixes are kind of rare if you are careful and pay attention to the screen. Out in the open they just don't happen. Note that the 1230 re-initializes every ten seconds, which does help to avoid problems. On critical points I take five shots somewhat slowly, which means under several initializations.
Agreed. On boundary we also require significant constellation change. We run Leicester 1202gg and Topcon hyper V.
I had a bad init with a Leica in an open parking lot a few months ago. It happens...
They keep this up....
Looks like they're just getting better to work with....
They're going to make me take another look at them at this rate!
I wonder how they'll stack up to the R10's ... Hmmmmm
They keep this up....
Our first units were the original Hipers. It would get a bad intilization and you wouldn't always know it. You just had to be carefull. We now have Hiper II's and they are much better. I've got them to obtain a bad fix but its in such a bad area with such heavy tree cover you no its a problem one. I don't doubt Javad has figured out a way to all but eliminate a bad fix.
They keep this up....
When the Verify setting is turned on in LS it lets you specify how many times it verifies the initialization. It does this by automatically resetting the RTK engines (there are 6 simultaneous engines that run in the LS) and then re-initializing and comparing the results of each initialization. You can specify how many seconds of data to record when fixed, the number of times to verify and the minimum number of RTK engines that must have a fixed solution.
"Javad’s recommended settings: Timer set to 10 seconds;
Verify 3 times; minimum number of RTK engines 1 or 2."
It will then show the horizontal and vertical spread of the verification set after the process is complete. I am eager to try this when the software update is released later today.
They keep this up....
I will be asking for a unit to demo once they are "out of beta". I just hope this feature doesn't eliminate usability on the fringe.
I have gotten some shots in bad places using rapid, and then using other methods to verify. I guess I'll still have that option if needed, though.
They keep this up....
It shouldn't effect usability at all. For one this setting can be turned on and off and it doesn't effect the RTK engines ability to obtain a fix. It will just take longer to get a shot when verifying positions where it is difficult to initialize.
They keep this up....
Exactly. As far as software engineering goes, I'm far more intrigued by the "6 Pack" RTK algorithmic processor.
I am impressed by the PRACTICALITY of the auto-verification idea.