Need some help. We are performing 4 hour static observations on Township Corners. Each station gets two 4 hour sessions. On some, due to canopy, we get results back of OBS USED 5639/10418: 54%,?ÿ# fixed amb: 90/107 : 84% (as an example). Overall RMS: 0.020(m).
My understanding is?ÿ OBS used should be > 85%.
My question is, because the overall RMS?ÿ is .02(m), is this a good solution, or should we scrap it. I have processed the same file through Trimble RTX but the report seems a little vague to me, can't tell if I got a quality solution or not.
?ÿ
8 hours on township corners? Under canopy? Guess I'm wondering why anyone would pay for that. They don't appear to be good solutions, it would be much better to survey points in the open and traverse to the corner. To tell what the quality is, the two sessions should agree within a few millimeters, depending on CORS locstions.
How close are your two 4 hour observations? 84% observations used tell you that you did not have perfect skyview, but you knew that already. If the solutions?ÿare repeatable the data is useable. In most areas you could break that 4 hours into 3 or 4 OPUS-RS files. I would not hesitate to mean 3 out of 4 OPUS-RS solutions and throw out the outlier. I agree with occupying multiple points near each corner. In doing that it is preferable to get simultaneous observations and field survey same.
Paul in PA
I'm with Moe. There are better ways to do this. If the observation conditions are poor you can sit on that corner forever and still not get a good position.?ÿ?ÿ
OPUS is a great resource, but it isn't for every task, this is clearly not one of them. You can chase your tail endlessly. Not trying to lecture, we've all been there with GPS. I've spent way too much time on some points trying to make it work.
85% observations used does not guarantee a good position and less is not necessarily?ÿa bad position.
85% indicates you may have 1.5 bad satellites, but if you have good geometry, less observations used can still provide good useful data. That is why you compare what you have and don't automatically bail out on your task.
Paul in PA
Run two receivers at the same time on adjacent/nearby corners. Make sure one of the receivers is in open. Process in OPUS Projects. The receiver that is in the open will have clean vectors to CORS. The receiver that is shaded will have poor vectors to CORS, but the short vectors to the nearby receiver will resolve well. Then it will all work out in the adjustment.
M
I suggest you do as Mark has described with GNSS receivers at 5-second epochs, but do not process all with OPUS. Send OPUS only the points in open sky. You will get better results post-processing yourself the points in canopy, as OPUS only uses 30 second epochs, and GPS. As far as I know OPUS is not processing GNSS. In canopy you want to take advantage of all the SV's and constellations your receivers can see.
I would scrap it.?ÿ With OBS that low you might have hz solution wrong by feet in my experience.?ÿ Because it's no good even after 4 hours, I go with the suggestion of traversing to the point from better positions in more open area. Still, might get good results if you have equipment and know procedures to follow suggestions above.
In order to check your vectors, you want at least 3 receivers collecting data simultaneously. This way you have a triangle to check closure. After all of this work and time, it still may not process. Therefore a conventional?ÿtraverse with EDM and correct procedures are the only "sure thing".
If you have two four hour sessions taken at different times and they agree with each other at centimeter or sub-centimeter level go with it. You're not going to get the same wrong answer twice.
Run your observations through Javad DPOS.?ÿ You will get a much tighter solution than OPUS every time.?ÿ Its free, also.
https://app.javad.com/jca/#/home
I'm pretty sure?ÿJavad DPOS only works with Javad files. Are they accepting RINEX files now? Sounds like the OP is using Trimble.
If you have two four hour sessions taken at different times and they agree with each other at centimeter or sub-centimeter level go with it. You're not going to get the same wrong answer twice.
True, but if two don't agree do you go for a third with a chance that still no two agree??ÿ Even with todays GNSS systems I still think at times better off doing something other than direct occupation.