In particular, we can turn the neck base from fixed to floating and consider linear and rotational accelerations undergone by the head assembly during the above tasks.
Also, we can account for external forces/torque applied at the top of the head when in weak configurations (e.g., the pitch is fully bent).
Once done, we could deploy the new PID gains on the physical robot and check the performances while executing the tasks.
Following up on the very nice results obtained in https://github.com/icub-tech-iit/ergocub-design-head/issues/89, we'd like to push further the design of the controller so that we can reject the more complex disturbances as those yielded by:
In particular, we can turn the neck base from fixed to floating and consider linear and rotational accelerations undergone by the head assembly during the above tasks.
Also, we can account for external forces/torque applied at the top of the head when in weak configurations (e.g., the pitch is fully bent).
Once done, we could deploy the new PID gains on the physical robot and check the performances while executing the tasks.