The other day I was running a static session on a base point and a strong gust knocked over the GPS while I was away. Is there a way to look at the static data to determine when the fall might have happened so I can truncate the file? I looked at the observation data in Magnet tools but didn't see anything like an abrupt loss of satellites being tracked from when the unit was laying on the ground. I was away doing RTK on a mountain job for about 6 hours and have a pretty good idea when the base might have fallen but would like to try and confirm my suspicions.
I'd suggest chopping the data into 2+ hour sessions and sending to OPUS.?ÿ I assume that the base station has previous occupations that you can compare to.?ÿ If you were running another base (RTK base?) you could chop it into 1/2 sessions and post-process (or OPUS-RS) to see when the position changed.
I don't know from your post what the brand is, but for Leica and its post-processing software you can look at the residuals.?ÿ That will likely indicate when your receiver decided to take a nap.
If there is no abrupt loss of data, you can track the location of the receiver by processing the file. Process 10am-11am see where the receiver location is, process 11-12, ect, When it moves then you can get closer if you want too. I've been able to isolate it that way when cattle tip or lean the receiver. Problem is that even while leaning the dang thing will track all the satellites.?ÿ
I know in my receivers when I create a RINEX file I sometimes see the comment "Antenna is moving". I work from just before that to the time I noted I reset the antenna. Also the processing software sometimes call a time period Kinematic that actually is not. In early January I had one project with gusts, 3 out of 4 antennas blew over and one a second time. I rest them with a traffic cone sitting on the upwind leg or two. I did not have enough cones to cover all of my 3 legged bipods. Once I was solving in my processor I had to shorten a few parts once or twice to get normal residuals. A couple of hours extra office work sure beats a 312 miles return trip. For my base I got a 2Hr 10Mn OPUS and a 1Hr 03Mn OPUS-RS. Two short duration vectors got excluded from my adjustment.
I do know in the older Leica Ski Pro software you could see satellite residuals epoch by epoch and I improved one project by only eliminating 1 satellite for several minutes. In that particular case that satellite was passing in front of the Sun at that particular time.
Does your processor have a "View Raw Data" option?
Paul in PA
In Magnet Tools change the data from static to kinematic, download the nears CORS RINEX data and post-process. Then look for the abrupt elevation change of ?ñ2 meters. This should identify to point of impact.
Point and time of impact does not always happen. Of my 4 blowovers 2 hit the ground, the other 2 were leaners to 45° so I had to backtrack to when the data started looking off.
Paul in PA
Try the Canadian PPP service. If I remember correctly, it gives a plod of position vs time.
Thanks for all the suggestions.
It got me heading in the right direction. I had a third receiver collecting static data so I could post process short 5 minute static sessions from a different point to check my RTK shots. it was a mountain survey with some tree cover and wanted a verification shot without having to go back.
What I ended up doing was use TeQc to split the file into 15 minute sessions and process the blow over against the third receiver. That got me the 15 min period when it fell. I then had to break that down to 1 minute sessions because I had a RTK shot about that time. I found the base blew over a few minutes after I got my last RTK shot. Nice to get lucky every once in a while.?ÿ
As Bill93 mentioned, it is sometimes possible to process epoch to epoch in PPP.
I have had too many blow-overs in the 34 years that I have been doing GPS, including one in to the river (too deep to recover without a diver).
When doing photo/lidar control on pavement, blow overs happen way too often. I had one a couple of months ago blow over into a traffic lane on I-95 in NJ. Not sure how but nobody hit it. I saw it wobbling, jumped out and picked it up pretty quickly, then moved the photo point away from the active lanes. Could have been disastrous.
You were in a traffic lane on I-95 in New Jersey and are here to talk about it? I don't believe you.
I had a blowover once on to the hood of a brand new PT Cruiser. Paid the dealer ( actually where I have bought a majority of my cars) to buff it out. The car owner was satisfied (plus he did not no his property corner was there) and the dealer picked up a new customer for the future.
Paul in PA
I'm glad you were able to isolate the issue and save your work.?ÿ I have a Base Guard function on my Javad rover that is an on-screen button that tells me how level the base still is and if it has moved.?ÿ That gives me peace of mind, especially when working in areas where people may be inclined to "help".
@paul-in-pa I was in a cross over between NB/SB on the corner of an inlet. It blew over in to the left lane (of three in that area). I was very leery about it to begin with, I didn't like where it was, but I figured it was only for three minutes...needless to say I moved it to a safer location.?ÿ
?ÿWe do a lot of work for the PA Turnpike, getting them to close down a lane for us is difficult, usually only happens when we need to measure inverts for CB's against the jersey barrier.?ÿ
I have a Trimble Alloy that I use as a base. It has a Position Monitoring function, and also has RTX. I can enable it and I will get an email if the position changes by more than the tolerance which is configurable.?ÿ
?ÿ
This may help if you are running TBC. There are likely comparable tools in other software suites.
I was doing a LONG Static Observation on a pillar once and had a passer by adjust the leveling screws.?ÿ I saw a 2 cm difference in horizontal position from one day to the next.?ÿ We had the antenna mounted on a 1 meter pole on a Tribrach.?ÿ By looking at the results of the 24 hour files, I split them up into 3 hour files and submitted them to OPUS.?ÿ From that I was able to determine which 3 hour window it occurred in.?ÿ I took that 3 hour file and split it into 20 minute sessions and submitted all fo them to OPUS RS.?ÿ By comparing the positions,, I figured out when ot was messed with.?ÿ Since I had 14 days of data to work with, I was able to use the data before and after the re leveling.?ÿ We were doing a study to determine if the structure was moving so it didn't ruin our project.?ÿ I?ÿ made a cover the hide the leveling screws and observed for several more months.?ÿ Over a period of 6 months, that point moved 3 mm or less.?ÿ We went back and did another optical survey to verify.?ÿ We were looking for large movements like 2 cm over a week, and not expecting 2 cm in one day.
Just process the data as kinematic in PPP.
Should be straight obvious, unless it spent 99% of the time akimbo.?ÿ
Process the raw data as kinematic. You will then get height values at every epoch that you used in the observation session. Look at when the height dropped by 1.xx meters. The resulting data should have a time stamp for every epoch recorded whether it was fixed or float. Trim the original data to before that?ÿ time stamp.
?ÿ
horizontal should be just as obvious. Height is just one of three returned values. Even in ECEF
Sounds like not many guys keep sandbags in their trucks.
Working up in the mountains where it can go from calm to gusts of 70 mph and having experienced a couple of blow overs, I usually drive three 18" No. 5 rebars and wire the tripod legs to them.?ÿ
