While I was at Johns Hopkins University during the summer, I found out about the first demonstration of a new chip that can be used to stimulate locomotion in an animal (tested on a temporarily-paralyzed cat, see right). Unlike previous controllers, this one is tiny and low-power. However, it can still take account of the sensory input coming from the movement of the limbs through a tiny neural network.
I find the work very interesting and, potentially, extremely important. Rather than explain it here, I recommend you check out the story I wrote for EE Times on the subject. Let me know what you think.
Diagram: The schematic of the experiment showing locomotion stimulated by a central pattern generator (CPG) chip.