Open
Description
The performance of the inverse-dynamics control heavily relies on the quality of the identified motor models. Even if these models have already been validated, we cannot be sure that they remain valid in all operating conditions. For instance, a higher room temperature during the summer may significantly affect the friction, inducing an error in our models.
We should create an entity that constantly checks the quality of the motor models and notifies the user in case of significant errors.