In academic research and industrial practice, it can be observed that computers are used to control machines. The concept of computers are so integrated, in which for each unique machines that act as robots posses unique design of their own intellectual core a.k.a brain. Those brains are well integrated, in which if we pull out the brain from a robot A (that works as an welding inspector) to robot B (that works as diver), they can’t be operated, simply because each connection to the sensors and actuators are so unique.
Unique VS Unifying Framework
I would suggest a unifying paradigm, where in a case where the brain or the core of robot A are taken, it should be able to control the sensors and actuators in robot B without a hassle. Why? Because that’s how probably human works. We posses unifying design, in which, metaphorically, if the brain of person A can be transferred to person B, the transferred brain should work just like that.
Hence, a standardized controlling core that can be termed as ‘the Core’ or Modular Intelligence can be proposed. It would be great to see this adopted in reality.