Isaac Asimov, the father of
robotics, considered this problem and considered that it will be possible to do this (using a science he called
mentalics). Also, he considered robots to be
intelligent beings, and possibly
sentient as well, even if they are not patterned on humans.
On the second count I agree with him, it is possible to create intelligent, sentient machines. However (possibly because I am one generation up on him) I also believe that it is possible to create machines with true feelings and emotions. This is an interesting concept to toy with (hypothetically).
However, I disagree on the first concept. The human brain is many times more complicated than any computer that will be invented in the near future, so mapping the human brain topologically is more or less impossible, for a few years at least. But don't quote me on that.