It works with same code. Its electrically identical, just smaller. I try to make the arduino code this week and test the buttons, will post here when its working.
My program has 2 modes, survey and grade, you switch by click on the bottom task bar.
In survey mode you can only survey, it will build an .ags file when done. It have to be done in one shot, you cannot exit the program and resume the survey later, it will overwrite the older one (probably need to add time/date stamp to the file)
This .ags file cannot be used in opengrade3D, you have to pass trough optisurface to make the design and create an .agd file.
In grade mode opengrade3D can only work with an .agd file, all others functions (except laser mode) have been removed. You cannot modify the cutaltitude once in opendrage3D you can only adjust the overall height (like commercial systems)
All the big calculations are made the first time the .agd file is openned, after that you can close and open the field or opengrade3D and all the data will be there.
@KentStuff I didn’t take time to look at your AOG survey yet. I think you should export an .ags file for your survey only (only existing altitudes) and if you add cutaltitude values to your point you should build an .agd file (only with the points that have an altitude and a cutaltitude).
This could be used in opengrade3D.
Opengrade3D can handle survey pts 30m apart right now and is optimized for point equally spaced (like a grid) but I will add more setting to handle random spaced pts and greater distances.
Here are example of files
http://support.optisurface.com/knowledgebase/articles/1941271-3-1-3-survey-file-fomats-accepted
That’s good information. I’ll change the extension. *,ags - as the survey only, *.agd - for design.
When you get a chance, post few line files from each type to make sure that it’s the same.
Edit: Try this one. It was saved in AOG. Based on your agd reader. I’m setting the cut and fill to the same edited value. Just save the file and drop the txt. It is just a simulated file with a lot of variations. I tried to open it in your OpenGrade3d. It put it way to the right of the center of the screen. I had to zoom back to the moon to see it. I did not attempt to drive to it to see what it looked like. Must be something to do with that simulator offset.
Survey.agd.txt (15.2 KB)
@Pat how are you sending the RTCM corrections to rover with opengrade? Do you have a separate bluetooth module or another USB cable to it?
I have a Ardusimple simpleRTK2B and their bluetooth module. I use the Lefebure ntrip client but I have to keep my phone sreen on to keep it from dropping.
I would like to have it trough the tablet, I was looking to build ntrip into OpenGrade but seam to be a lot of work.
Do you think we could use the second usb port to connect to uart2? Or we need to jump cable. I was thinking about it but not made extensive research. Lefebure seem to also have a good client for windows.
Manual control with ino + pcb now working, it’s now only in the man mode. Takes the max pwm value for the manual stuff, ino attached
ValveControl.ino (8.6 KB)
I now managed to massage the NTRIP client (somehow and dirty) into your latest version @Pat
Not sure yet if it’s sending the corrections correctly, would somehow need to sniff the COM port going to ardusimple to see what’s being sent to the receiver. But getting there!
And now it looks like it’s working Added a label to the top bar to keep track what’s going on with NTRIP. Stupid, stupid mistakes when integrating the code last night, was just missing the line of code that calls for updated data… It’s found as menu item in config menu.
In case you’re wondering, the antenna is inside on a windowd board for the test, hence only float fix.
I got my opengrade to open the agd file created in AOG. It was a simulator offset problem.
Same file opened in Opengrade3D.
I’ve added Pat’s contour map to AOG. Busy for sure, but should be able to toggle them off and on.
Just surveyed a 5 hectare field to test and the OG seems to be working, as well as the NTRIP, RTK fix full time. Huge difference to the previous single-band antennas I had, no more constantly dropping to float RTK and find the RTK fix about 10 times faster.
Modified the elevation map painter a bit. Still makes the map with Pat’s routine, but draws it a bit different. Also, added a display option in the display settings that you can turn off or on the survey marks and grade map. When making a new field and surveying the field, using the survey button at the lower left, when unclicked it draws the grade map. It still takes a few seconds to process the whole field. Once in the ABDraw screen, once you click select, the survey marks come back. Right click on any of them and you can edit the elevation point or the cut point. It redraws the grade map once you click the edit this point button. Cut and fill can be grouped with a lower left to upper right window.
![image|690x215]
I have made a new release (1.0.3-beta) with @nut 's work on ntrip integration. It’s a beta, there were some little bugs in the gps/simulator/ntrip combo but I think they are resolved.
I have integrated the AGS file survey into AGopenGPS, so it should also use the roll compensation.
It’s really light, sould not interfere with the rest of the app.
But not sure all the sin/cos are OK, neither the altitude.
It’s independent from others boundary and fields.
Click on the button under the martian to make the buttons appear.
Right now the survey can be made with a field open or closed, it will create a survey.ags somewhere in the field directory, crushing the previous one!
When surveying, you have to drive the boundary first then the field. It make the survey pt on the pivot axle, the boundary pt at a half tool width from pivot axle.
I will give a try in a hilly field in some days and compare results with the opengrade survey.
To install, copy this AgOpenGPS.exe to your folder.
It’s a beta, some little visual bugs maybe.
If someone look at the code or run it in the field give me a feedback.
Yes, I had trouble with mine, when to subtract the antenna height. Finally, decided on only when building the point from the driving. Because when loading an old file it would add that back. You can start and stop survey whenever, but in adding a boundary, do it first. Working on reducing the make the map routine time. I believe by sorting the pt list northing, then easting, then once the loops find the closest point, It should be able to focus on that area for the others. And once the distance begins to increase after finding the closet point, to exit that routine. In AOG, we already have the max and min field dimensions so no need to do that again. Just something I’m working out in my mind right now.
I updated my AgOpenGPS.exe
Changes:
It create a new ags each time, no more overwrite
It also create an agd file to verify the points with OpenGrade3D in the field right after survey (only for existing elevation check)
Some visual improvements
Still not tested in field.
@KentStuff I was thinking the last few days that we could develop a program only for design like optisurface, like what you do with AOG. I would like to have it all separate, AOG for survey, a design app and OpenGrade3D for levelling.
Are there any free design software available?
My goal was a simple field editing software that could handle a small leveling job all in one package in one field. But that is growing. My next project was to be able to slope the selected points to one direction. Or create a flow direction. Been also working on mapping the cut and fill. I’d like to export the maps to a jpeg file for printing. Not much success in drawing the contours yet. Finding all the even points is easy enough. Finding a closed circle, not so much (at least for me.). There are other using system.windows.bla.bla that can build meshes and contours, But each I have tried interferes with the main drawing functions.
I tried to look for one, without success. I would go for python to develop that part since all the math and surface visualization stuff is way easier and faster to develop that C++/C# and you can build a GUI as a web app quite easily.
Fitting a plane is quite easy, but doing a free form design like optisurface is bit more involved. Was trying out something with fitting a surface with least-squares and then putting a constrain on the gradient. It’s an optimization problem of finding a solution with no local minima (i.e. no ponding) along with a gradient constraint but couldn’t quite get it right on the first go.
Yeah, using optisurface is the fast solution for now, 2.50 USD/Ha for planes or 7.50USD/Ha for 3D design is not so bad.
A multi-plane software to begin would be great even if you have to manually set the height of each reference point(corner). Each design pt being calculated form is 3 closest reference pt.(My next opengrade should be able to handle also only the reference points) With a cut/fill calculation for each plane and the whole field. So you can adjust the pts to have a decent cut/fill ratio.
After some thought, I will personally devote time to other things (always too many projects here! ) but will definitely give a try if anyone is working on this.
@KentStuff here is one of my field ags. 1.7Ha, Survey65.ags.txt (42,5 Ko) very flat, need to remove the ponds
I will make many surveys this fall so I will have a lot of data.