OpenGrade

[Briantee]

A few guys were chatting about adding big elevation numbers to AgOpenGPS for doing dozer work and drain/leveling work, so i thought i had most of the code already written for changing it to turn it into dirt work software.

So OpenGrade has been started. It is open source of course and will be downloadable for free off of github once i get some more details worked out. Not sure if i will make it 7 languages like AgOpenGPS, but who knows! A couple videos on the start. The plan is to make it fully automatic to run hydraulics too.

Fun project and an alternative to specialized commercial software costing thousands.

And multipoint cut line creation.

 

[Northern farmer]

I've implemented the control SW on a Raspberry (but have not been able to test it because the season is gone). My programming skills are less than 0.00000001% of yours, don't have any graphics, just the altitude figures and status data (on the tractor screen). Having seen this thread, it is better for me to start studying how to adapt the existing tractor interface to reuse your project.

I have no real experience about field levelling but my opinion is that it is too time consuming and inaccurate to define the target level manually. Instead I've assumed a process where the field level is first measured (at a sufficient sampling rate) and then a two-dimensional target surface should be calculated. This could be partly manually edited but preferably automatic, the volume to be moved should be minimised while targeting to a smooth surface. For a surface reference (instead of a line) the operator is free to chose the path to drive the machine and the application can still control the working depth of the tool.

This OpenGrade would obviously take care of the last step but the preparatory steps could be built to be independent. One could then mix and match steps too with different implementations.

I'm quite surprised about the horizontal accuracy from and RTK system and a base near by. I guess commercial systems use lasers or some other measurement systems instead of GNSS receivers? Our fields would need levelling of a few cm only, nothing like half a metre or even 20 cm but the standard RTK receiver seems to be accurate enough.

 

[Briantee]

Agree. Commercial systems are complex and expensive for a reason. Hoping to be able to drive over with RTK, record altitude, and then make an appropriate slope and resultant line. I'm thinking no matter what, its better then eyeball or a few sticks with sharpie marks.

Emlid Reach and Swift Navigation both make cm accurate RTK systems for under 2000 US$ so a lot of these projects are becoming quite accessible.