Philosophers have argued about the nature of consciousness for many centuries. In Impossible Minds
Igor Aleksander looks at the question of whether a machine can be conscious, and more generally, what experiments with machine consciousness might tell us about the human variety. Aleksander describes the work he has been doing in trying to model consciousness, describing the features he thinks a conscious being will have and how he has included these in his neural network Magnus. But the book isn't just a description of the author's work, it also discusses the ideas about consciousness which people have put forward through the centuries. Thus it would suit anyone looking for a general overview of the nature of consciousness and the question of whether a machine can possess it.
Sometimes the book seemed to be a bit vague - it's not really suitable as an introduction to neural nets and robotics, but neither is it particularly persuasive regarding ideas about consciousness - Aleksander seems keen not to argue too much with ideas put forward by others. But this isn't so unreasonable as the real point of the book isn't to promote a particular view of consciousness. Rather, it's to make the point that the best way to understand consciousness isn't through endless arguments, but through experimentation, and especially experiments with neural networks.