Closed schmurfy closed 8 years ago
It gets a little problematic when trying to actually use it as I just realized, since these are executed immediately instead of being queued it does not work that well ^^
yes, I realized that it is executed asynchronously to the movement commands after having implemented laser control. That requires slightly different command representation in firmware and slightly different handling + make sure that everything still fits in the allowed time frame to not miss to send next byte via SPI to the FPGA. That said, let's still consider merging this (after we decide whether we want two commands or just one for the pump) and improve later. Incremental improvements. FYI, you can still use time.sleep() to delay execution of this command, though it would be horribly inconvenient and a dirty workaround.
Thanks for the code!
The ideal would be to have all the commands queued but I haven't looked at your code too deep so I can't comment on this, the problem is that you need to enable the pump at a precise point otherwise there is not much use to it and it will be the same problem with the claw.
There will be no problem with the gripper as it will be a part of movement command that is sent to FPGA. FPGA is in charge of PWM pins on the top board that controls the gripper. The pump will be fixed by extending the command structure in the firmware and being appropriately queued along with the movement commands. Please create a separate issue to fix laser/pump commands (after this PR is merged) and I'll plan it for the upcoming release.
Merged as I got some time to make them queued.
Please create another PR to add description to those commands. Thanks.
Queueing implemented. See #14
great :)
I am not that clear what are the physical actuators exactly but it works, when you turn both on it grabs things xD