Is there a way in Carlson to best fit a block of points to another block of points?
I used to use a program called Triad and it was used to average the positions of say field points to the platted tract corners... It would then give you the errors and you could use a scale if needed.
thanks
Could you creat a rw5 file of the the points locations with angle and distance and then use the other points as a control file in Survnet ? may be the long way around but should adjust your first set to your second set and give you the errors .
Thanks for the idea, I'll try that.
Yes, under the coordinte file utilities, there is a routine I used all the time to rotate and check alignment of platted points versus filed shots. Haven't used the program in a few years to remenber the command.
Deed Correlation. Works slick!
Matt
:-X
I saw this mentioned this week in the initial Beta of Carlson 2013:
>Best-Fit Point – New command to average points and report statistics.
I don't have any other details about the routine just yet but it sounds like something should be coming soon (and before the question is asked... it is a bit premature for me to even speculate about the availability release date of Carlson 2013). 🙂
--
Ladd Nelson
Under the points menu, Coordinate File Utilities, Coordinate Transformation.
i remember triad
if you do this often and want to expand, look up PrimaCode's Transform
written by a member of the Mass. Board of Registration, the basis is to scale older subdivision plans to match monumentation located with modern equipment.
It's a tough problem mathematically. I'd be interested in any math links anyone has on how this is done.
I didn't have much luck aligning the rotation and shifting of two blocks of points with my least squares program. It tends to go numerically unstable when two blocks are tight within themselves but loosely tied to the other block. Zero or near-zero distances between points are really tough for the linearization formulas to deal with, and some programs refuse to accept such constraints. My program does not have an overall adjustable scale factor, which it really needs, but that won't fix all the problems.
http://www.primacode.com/
30 day trial
sounds like what you are getting at bill93
sounds scary
makes me nervous to let software do this automatically. I'd rather play around translating and rotating the points manually paying particular attention to how the stable control points line up.
sounds scary
transform that i pointed out above pairs the points between the two data sets, then calculates the distances / bearing between them, it gives you the statistical data and allows you to abandon / unselect pairs to change the rotation/translation of one data set to the other.
i have a nice 3-mile traverse, run once in one fell swoop, then we've returned over the years and ran loops of it and even just re-tied into points. i GPSd every original control point that remained a few years ago (rapid static, each point twice from different base stations with an appropriate time offset for constellation).
this traverse is on Route 28, a 1927 state highway layout. the monuments and the math don't exactly match: most match one solution, in which case some are off by 3' and some are off by 6'. transform gave me these results in a few minutes. i would have been overwhelmed to do this all by hand.
ps, be sure to find the "rollback scaling" item in the menu. i don't like to scale a plan unless I am sure i have the full solution of a record plan.
sounds scary
I'll second the recommendation by spledeus. I use this program, too.
(tell them Don Poole sent you!)
sounds scary
hey, you're just brown nosing dennis!
tell them thadd and don sent you, but mostly thadd.
(i can brown nose too)
What Is Really Scary
Is people using least squares programs to massage poor fieldwork into gold, who could not do the sme manually and cannot explain what happened.
Paul in PA
What Is Really Scary
> Is people using least squares programs to massage poor fieldwork into gold, who could not do the sme manually and cannot explain what happened.
>
> Paul in PA
Doesn't sound like anybody's trying to massage poor fieldwork, it seems more like people are using least squares/scaling try to analyze and retrace boundaries. I believe this is one of the most overlooked and valuable applications of least squares.
Ralph
What Is Really Scary
I agree with both Ralph and Paul. Anyone using LS should know what's going on in the background and realize it has it's own flaws: statistics and probability. With that in mind, LS is an excellent solution for many problems.
What Is Really Scary
The LS analysis report will tell you if the field work is poor.
