Notifications
Clear all

RTK: Things that can go wrong

46 Posts
17 Users
0 Reactions
8 Views
(@francish)
Posts: 378
Registered
 

I do case #1 all the time if all I want is to have a basis for azimuth/bearing for the survey. I just adjust in the office to move points measured into a local coordinated plot.
If I want to connect to an existing reference station, I also do case #1. Only difference is I have another base located on top of an existing GPS control point with knownpublished coordinates.
The autonomous base with no nail is there to add accuracy to the local points when post-processing is done later in the office. So instead of getting coordinates of local points coming all the way from the base/network station say 10 km away, another vector is added into the mix from the autonomous base with no nail under it.

 
Posted : August 24, 2016 9:08 pm
(@nate-the-surveyor)
Posts: 10522
Registered
Topic starter
 

I have NO problem with case # 1 above, IF you know what you are doing. However, as with all things, pushed too far, things can go wrong.
Javad LS has software, to run base coordinates in reverse, with a shift.
Set base autonomously. (This is USUALLY within 3 ft, of true SPC, if you let it settle down, before grabbing a "Here")
Take rover to another point, with genuine SPC on it. Occupy, and store the shot. Now, there are 2 coordinates based on the autonomous base. Tell the software the 2 point numbers that are supposed to be the same, and it tells you the SHIFT to move the autonomous base, and all it's sideshots.
I have no problem with that... as long as you 2x check into something else. (ie, to be sure someone did not move the point you used to get on true spc.)
Or, that something else did not go wrong... This is a great feature, when setting your base in a cow pasture, where cows or animals would knock your base down.
(hint, set a control nail, nearby, to check on)
But, without the double check... you are trusting too much.... I can hear the mother tell her son... "you go put that surveyor's point back..." yup.... he shore did!
N

 
Posted : August 25, 2016 4:09 am
(@yuriy-lutsyshyn)
Posts: 328
Registered
 

It was bugging me what error in coordinates it would be if we found all ambiguities right but only one of them is one DD cycle off, and the answer is:0.326m in my case. Of course if we have more than one ambiguity wrong than wrong coordinates would meters.

What is really very important that epoch to epoch solutions are 1-2 cm off as long as we hold the DD ambiguities that we think are true, so we get perfect error estimation while the being a foot off from true position (when one amb is 1 dd cycle off)

In reality ambs can be many cycles off and not only from one satellite, I would not rely much on RTK in bush.

Attached files

OnceCycleOffEffect.pdf (37 KB) 

 
Posted : August 26, 2016 9:25 am
(@nate-the-surveyor)
Posts: 10522
Registered
Topic starter
 

Yuriy, have your ever seen a Javad work in the woods?
N

 
Posted : August 26, 2016 2:51 pm
(@monte)
Posts: 857
Registered
 

Yuriy, I am asking because I am curious, please don't take offense, I am asking to learn. Did you have English as a language course in school, or as a hobby, or do you use a computer translator? My question comes from my trouble in learning other languages, and feeling a translator doesn't correctly carry the message across.

 
Posted : August 26, 2016 3:18 pm
(@warren-smith)
Posts: 830
Registered
 

Monte,

Are you referring to speaking to the rest of us non-Texicans?

 
Posted : August 26, 2016 3:20 pm
(@holy-cow)
Posts: 25292
 

[USER=9900]@Warren Smith[/USER]

My thoughts were similar to yours. West Texans and East Texans barely comprehend one another. How can the rest of us ever hope to master their dialects?

 
Posted : August 26, 2016 3:30 pm
(@monte)
Posts: 857
Registered
 

Well... I am still not happy that my cell phone cannot understand me, my text tries to correct my typing when I know exactly what I meant, and that for some reason they teach my child Texas History but they also teach English but not Texican.

 
Posted : August 26, 2016 3:33 pm
(@warren-smith)
Posts: 830
Registered
 

Monte, post: 388238, member: 11913 wrote: Well... I am still not happy that my cell phone cannot understand me, my text tries to correct my typing when I know exactly what I meant, and that for some reason they teach my child Texas History but they also teach English but not Texican.

Claro que si!

 
Posted : August 26, 2016 3:39 pm
(@monte)
Posts: 857
Registered
 

Warren Smith, post: 388240, member: 9900 wrote: Claro que si!

That looks Spanish. I can get by in TexMex. With gestures. And Slang.

 
Posted : August 26, 2016 3:57 pm
(@yuriy-lutsyshyn)
Posts: 328
Registered
 

Monte, post: 388233, member: 11913 wrote: Yuriy, I am asking because I am curious, please don't take offense, I am asking to learn. Did you have English as a language course in school, or as a hobby, or do you use a computer translator? My question comes from my trouble in learning other languages, and feeling a translator doesn't correctly carry the message across.

Hi Monte,

I was not good at English in school at all, it was later on, in 1995 when when we received 3 Trimble 4000SSE units and all manuals were in English, I realized the importance of the language the world speaks. Internet just started to be a common source of information back than and that was another reason to learn English.

Like everything I did it on my own, first by memorizing separate words, later on I used Raymond Murph„Ä - English Grammar in Use to learn grammar, watching BBC helped too.
Later on I moved to Canada and that is where I learned all the swears, and slang . Yelling at each other while surveying in bush is common practice there 🙂 ( like everywhere )

I use google translate jut for words but not for sentences.

Yuriy

 
Posted : August 27, 2016 3:39 am
(@yuriy-lutsyshyn)
Posts: 328
Registered
 

Nate The Surveyor, post: 388231, member: 291 wrote: Yuriy, have your ever seen a Javad work in the woods?
N

Hi Nate, no, only on your pictures.
I am developing RTK program by myself (my hobby) and it is almost impossible to find correct ambiguities in bush because most of the satellites are affected by multipath and I do not understand how it is possible using affected observables to find correct ambiguities even if you have L1/L2/L5 data which are all affected. If one does not see the sky through canopies it means 100% are affected by multi path.

I am trying to develop a procedure to find out which satellite suffers from multipath and than immediately exclude it from contributing to normal equations accumulation process, which is part of my program. but I am far from that. I have only L1 at my disposal and its SNR, which is supposed to decrease once signal comes from different path.

 
Posted : August 27, 2016 5:05 am
(@nate-the-surveyor)
Posts: 10522
Registered
Topic starter
 

Well, Yuriy, Javad has conquered that problem. If you are a tinkerer, then you may wish to continue with your tinkering. But, if you are trying to get a job done, the LS is the most phenomenal piece of multi-path conquering hardware out there, at any price. I don't sell them. But, I sure use them!!

Nate

 
Posted : August 27, 2016 8:45 am
(@williwaw)
Posts: 3321
Registered
 

Nate The Surveyor, post: 388312, member: 291 wrote: I don't sell them.

Really? Sure comes across that way at times. Noisy data is noisy data. I can clean it up, slap lipstick on it, but does that make it something that it's not? Some manufacturer's algorithim may handle noisy data better than others, but what you're pushing is the superiority of one tool over all the others. You're highlighting all the strengths and not discussing any of the limitations. That to me is the epitome selling.

 
Posted : August 27, 2016 9:57 am
(@monte)
Posts: 857
Registered
 

Yuriy, thanks for the answer. I have tried to learn more Spanish (well, Mexican version) since I live and work near Mexico. I tried the CDs, I took classes in school, I watch Mexican TV, but I have trouble getting very far with language. It is encouraging to me to hear from people who have successfully taught themselves new languages, so I will keep on giving it a try.

 
Posted : August 27, 2016 10:31 am
(@mark-mayer)
Posts: 3363
Registered
 

Nate The Surveyor, post: 388312, member: 291 wrote: ...the LS is the most phenomenal piece of multi-path conquering hardware out there, at any price.

You have to admit, Nate. That sure sounds like salesman rap.

 
Posted : August 27, 2016 12:04 pm
 adam
(@adam)
Posts: 1163
Registered
 

Yuriy Lutsyshyn, post: 388289, member: 2507 wrote: Hi Nate, no, only on your pictures.
I am developing RTK program by myself (my hobby) and it is almost impossible to find correct ambiguities in bush because most of the satellites are affected by multipath and I do not understand how it is possible using affected observables to find correct ambiguities even if you have L1/L2/L5 data which are all affected. If one does not see the sky through canopies it means 100% are affected by multi path.

I am trying to develop a procedure to find out which satellite suffers from multipath and than immediately exclude it from contributing to normal equations accumulation process, which is part of my program. but I am far from that. I have only L1 at my disposal and its SNR, which is supposed to decrease once signal comes from different path.

Yuri, folks that develop RTK programs by themselves for a hobby are the one's who go very far in this world.

 
Posted : August 27, 2016 4:58 pm
(@nate-the-surveyor)
Posts: 10522
Registered
Topic starter
 

Mark Mayer, post: 388336, member: 424 wrote: You have to admit, Nate. That sure sounds like salesman rap.

I am not sure what to reply to that. I want to see surveying quality go up.
Every piece of equipment we bought, made our work quality go up.
I am banging my head against a "Survey" done with RTK, abuse Monday, or Tuesday. They consume time. When you bang your head against one, and it eats up your time... it makes you sad.
RTK Abuse. I am against it. Marginal shots need double checks.
Period.
N

 
Posted : August 27, 2016 7:16 pm
(@yuriy-lutsyshyn)
Posts: 328
Registered
 

Thanks Adam, I have not gotten too far 😉

 
Posted : August 28, 2016 8:05 am
 adam
(@adam)
Posts: 1163
Registered
 

Keep a goin, Yuriy. You no doubt have drive and determination.:)

 
Posted : August 28, 2016 3:48 pm
Page 2 / 3