We experience and interact with the world through our body. The founding father of computer science, Alan Turing, correctly realized that one of the most important features of the human being is the interaction between mind and body. Since the original demonstration that electrical activity of the cortical neurons can be employed to directly control a robotic device, the research on the so-called Brain-Machine Interfaces (BMIs) has impressively grown. For example, current BMIs dedicated to both experimental and clinical studies can translate raw neuronal signals into computational commands to reproduce reaching or grasping in artificial actuators. These developments hold promise for the restoration of limb mobility in paralyzed individuals. However, as the authors review in this chapter, before this goal can be achieved, several hurdles have to be overcome, including developments in real-time computational algorithms and in designing fully implantable and biocompatible devices. Future investigations will have to address the best solutions for restoring sensation to the prosthetic limb, which still remains a major challenge to full integration of the limb into the user's self-image.