First I would like to say that I certainly do appreciate Ashtech's involvment with this board. It is an excellent learning tool. I have a question concearning editing a site i.d. in LOCUS processor. As we are still using V. 1.1 (patiently awaiting V 1.2 or if you like go ahead and send 2.0) this question may not apply any more.
I was testing the kinematic abilities of the receivers and as I was shooting I could not remember what the site id was for a point I had previously shot, so I just auto incremented to what ever the next available i.d. was (at present we think we will probably stick with numeric only i.d.s). I thought then that I would just edit this in the office but once I began processing it appeared the processor would not accept the changes. First I attempted to change the site I.D. in the Sites Tab which understandably I was not permitted to do. Then I attempted to change the site i.d. in the Occupations Tab. The processor allowed this apparently but did not use those vectors when I reprocessed. I know this because repeat vectors were not shown for these additional vectors. If possible, where does one change an incorrect site id? I would assume it is possible because the blunder detection caught that the two sites may have been the same. I'm just not sure how to remedy the blunder that it correctly caught.

Thank you for your help,
Shawn P.C. Texas

P.S. Unfortunately we haven't had much time to play with the kinematic since we got it, but so far what I have seen is incredible. It took about two minutes to get from processed vectors to ACAD and have the points scaled, rotated and translated, and since your handheld supports it, descriptions as well!

Shawn,

The correct place to edit an occupations Site ID is in the Occupations tab of the Workbook. When processing stop&go kinematic data, Locus Processor combines multiple observations of the same site into one vector solution. This is why you do not see repeat vectors when processing kinematic data.

At the moment, I cannot remember why we took this approach (producing one vector with multiple observations of the same site). I suspect it is one of two possible reasons:

1. Since multiple observations of the same site could be due to either a desire to repeat an observation or could be due to a need to re-initialize, the processing engine may not be able to distinguish between the two, causing the need to always combine multiple observations.

or

2. We may have decided on this approach to eliminate the need to perform an adjustment on kinematic data. If using only one base station is used to collect kinematic data, there is no redundant measurements on your rover points unless you observe a specific point more than once. If multiple observations of the same point produced multiple vectors and therefore multiple positions for this point, an adjustment would need to be performed to produce one solution for the point in order to get the benefit of multiple observations. By combining the data during processing to produce only one vector, the need to adjust is eliminated.

The drawback of producing only one vector solution for multiple observations in that you cannot use the Repeat Vector analysis to as a QA test of the data collected. With current approach, the vector will fail if multiple observations of the same site do not compare well. This is not as easy to analyze.

I will look into the reasoning behind our approach. If combining multiple observations of the same site into one vector solution is not a requirement (reason 1 above), then we may want to consider allowing the processing to produce multiple vectors to allow easier QA of kinematic data.

Hope this helps.

Bill Martin
Ashtech Precision Products

Bill,

Assuming we do two observations on each point would the following be a reasonable approach to QA the points:

1. Use the point I.D.s for the first observations and question mark the second set, process and print out for "visible" analysis.
2. Question mark the first observations and leave the point I.D.s for the second observations, process and print out for comparison to first print out.
3. If results of first "set" agree within user tolerance to second "set" then set all point I.D.s back to their original settings and process all data together, allowing the processor to "blend" the results.

We haven't tried this yet, but have sufficient data files to play with. We'll try when we have time if you think this is a reasonable approach for QA.


Thanks

J.D. Billings, TX RPLS

By the way, we can't get enough of the I.R. down load. Wish we could make it work with our TDS 48.

J.D.,

Your approach would work but you would need to be careful when setting an observation to ????. When you do this, the observation will blend in with the ???? before it and the ???? after it. It will then be difficult to get it back.

Instead, I will give the second observation a different site ID when you first process the data. This way you will get 2 vectors and 2 sets of coordinates. You can then compare them. If this looks good, change the second site ID back to the first and process again. Examine the combined vector and coordinates compared to the first. They should be similar.

Bill Martin
Ashtech Precision Products

Bill,

I believe the instructions you just stated are more or less what Shawn accidentally stumbled on to. We were thinking that it would be proper to I.D. points 0005 - 0010 first time around, then 0105 - 0110 the second time around, in a way they could be readily noted to be the same point. I think our confusion was in that we thought when we applied the same site I.D. to the points in reprocessing the data we would see repeat vectors. You answered that question. I also now see that the procedure, as you just outlined, would allow us to QA the data. That was our main concern with repeat Kinematic obs.
I do think it would be wise on topo projects to do multiply obs. on about every 20 or so points. Just personal preference though.

Thanks for your help as always.

J.D. Billings, TX RPLS