I have been trying to make the Finite State Machine (FSM) walking controller work for the OSL. However, I am encountering a critical failure where my ankle motor shuts down after a while, under the following error:
I am using the legacy-opensourceleg codes.
For reference, it’s always the ankle joint that shuts down (I have not seen this behavior in the knee joint so far). I have implemented a torque clipping logic to avoid spikes in the commanded torque and thought that would solve the issue, but in my latest walking trial, the ankle shut down after ~1.30 minutes - although there was no spike in the motor torque.
After the first trial in which this happened, all subsequent trials were terminated within the first few gait cycles, as the ankle joint shut down. I can confirm that all cables are in place during the shutdowns. Also, right before the ankle shuts down there is a distinct sound coming from the OSL, as if something is “cracking”. It’s almost like the joint is resisting the movement.
Has anyone else encountered this issue or has an idea what could be wrong?