I'd say that is "standard advice" for anyone wanting to make a change to any software system. Your first step is to compile the entire system from source code. Do that before you make any change and actually run the system to verify it works.
Then make one small change and test your change. "Test frequently" is the key. If you have a working system then if you change 12 lines of code and find the system no longer works you have a very good idea about where the error is located. Don't work for more than a few hours before rebuilding and testing. What really helps is to implement automated regression testing. At least at the unit test level and have this test is part of the build process. Again I'll ask if this is a "serial link" arm and if so, does EMC have enough information to do IK? You need a model of the arm, part, and fixtures to do collision avoidance or else how do you prevent the arm from moving through some object like a clamp? This is easy with a normal mill but hard with serial links. For those on the list who have not seen a "serial link". Here is how you make one: Get a rotary table and mount a 90-degree angle block on it, then mount another rotary table to the block, put a block on this table and then another rotary table on that. Continue until you have a stack of *six rotary tables* each mounted 90 degrees from the table they rest on. Yes controlling this is "crazy" and way-hard but 6-DOF serial link arms are pretty much the "stanadrd" and are common. If the six axis are in parallel then it is easier. A parallel 6-axis machine would be a Steward platform. That would be easy to do with just IK. Because you can Google the equations. On Wed, Oct 31, 2018 at 9:01 AM andy pugh <[email protected]> wrote: > On Wed, 31 Oct 2018 at 15:07, yomin estiven jaramillo munera > <[email protected]> wrote: > > > i say this because i tried to install genserkins.c with few changes and > it > > showed me a lot of warnings, then my file didn´t run. > > You probably needed to make some more changes until it compiled > without any warnings... > > > I am worried because maybe i could damage the trivkins files > > what do you think about this? > > Yes, that is entirely possible. > > I suggest that you get the whole LinuxCNC source code and compile a > run-in-place system. Then you can experiment with that without > damaging the installed version, until you have it all working. > > http://linuxcnc.org/docs/devel/html/code/building-linuxcnc.html > > -- > atp > "A motorcycle is a bicycle with a pandemonium attachment and is > designed for the especial use of mechanical geniuses, daredevils and > lunatics." > — George Fitch, Atlanta Constitution Newspaper, 1916 > > > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users > -- Chris Albertson Redondo Beach, California _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
