One can easily train more than one classifier to deal with multiple DoFs, and the device can be designed so that it can switch between different positions described by n-tuples (Position-DoF1, Position-DoF2, .... Position-DoFn). So what is the problem?

Similar questions and discussions