Most industrial robot systems employ some sort of look-ahead logic when it comes to motion. By looking ahead to upcoming motion commands, the robot controller can smooth trajectories between commands. Linuxcnc and GRBL both employ look ahead schemes.
My question is, does your software employ such logic? I also would like to know if there is any thought of porting the software to Linux in the future?
Thanks again...
Hello Chris...Thanks for the reply. I was thinking that the commands can be sent out to a buffer and then shuffle them out in a FIFO. I have done a lot of work with KUKA's MxAutomation in the past and this has worked ok for me. This scheme is not exactly a look ahead as it does not take into account the complete trajectory of multiple commands and the associated trapezoid or s curve acceleration/de-acceleration of each command but it does help smooth out the stop and starts.
I was wondering if a look at the GRBL code would provide a look into how they have done it.
Thanks for a great project...
Hello, no currently there is no look ahead but its definitely something Ive given a fair amount of thought to. with the current architecture I'm executing one serial command at a time waiting for the arduino to verify the command was completed and then updating the robots position in the AR2 software. Im looking at ways to move position tracking into the arduino so I can send consecutive move commands and execute in a look ahead fashion.