Hi.
During a recent experiment, despite homing the joints at the start, the reported ankle output position routinely exceeded its RoM (e.g., reported measurement of -45 degrees, while the limit is -30 degrees).
What could be the cause of this? This seemed like a drift in the encoder readings and/or some sort of miscalculation by the motor driver.
Note: I’m using the motor encoder readings (not the one at the joint output).
Hi @malizade
Did you check if the belt slipped ? Having also data from the joint output encoder would be great to fix the bug.
I’m not sure how to detect that. In any case, I think the motor encoder readings are not affected by belt slippage (since it’s further down the drive train), right?
The output encoder is not used currently in the code.
If the belt slips, your motor can travel further while avoiding the mechanical stop.
First, I would check the ROM at a slow speed and with no load. Then, while monitoring with the joint encoder, I would try to reproduce the shift of the ROM.