in the case of a robot, you can explain how it senses whats going on around it, learns things, and takes actions.
but is the robot conscious? is it ok to kill it? think about why it might be ok to kill the robot and not a person. because the difference between them is that the person feels; the person is conscious. it is the consciousness (what's left after all the other explanations) that is precisely the mystery that we seek to explain but have no idea how to do scientifically. that puts it into the domain of philosophy, but that doesn't mean it's not "much use". it's not much use yet, it will be some day. most of what we consider science now began in the past as philosophy. astronomy, physics, medicine, etc..
Or maybe there is no natural way to do it and it will just muddy the issue further? That would be interesting to know too.
I'm saying that whatever they are, they must cause the external behaviors we associate with consciousness. (And furthermore, the theory must explain things like the effects of brain damage, surgery, the effects of drugs, hormones, how children's minds develop, evolution, and so on.) These are just the requirements for any scientific model of how minds work.
I suppose there are alternative definitions but if they're not focused on somehow explaining peoples' behavior I don't think they're much use.