When controlling multiple actuators via Platinum Elmo drives using individual CSV commands, fastcat can get stuck in a state where one actuator is suddenly stopped, while all the other actuators on the bus continue with the last commanded velocity at the time of the "crash". The moving actuators continue at constant velocity and ignore the position command limits and all subsequent commands, until a fault is manually triggered (extremely dangerous situation!).
This is triggered especially frequently if fastcat is not given real-time priority and competing for resources.
Below is an fcat log during which that happened, before being shut down:
fcat_issue_log.txt
When controlling multiple actuators via Platinum Elmo drives using individual CSV commands, fastcat can get stuck in a state where one actuator is suddenly stopped, while all the other actuators on the bus continue with the last commanded velocity at the time of the "crash". The moving actuators continue at constant velocity and ignore the position command limits and all subsequent commands, until a fault is manually triggered (extremely dangerous situation!).
This is triggered especially frequently if fastcat is not given real-time priority and competing for resources.
Below is an fcat log during which that happened, before being shut down:
fcat_issue_log.txt