DARPA and the Human Machine Interface

In a BAA released yesterday, DARPA announced their Reliable Peripheral Interfaces (RPI) program:

DARPA seeks to develop reliable in-vivo peripheral motor-signal recording and sensory-signal stimulating interfaces. Such efforts will involve design, fabrication, testing, and analysis of new materials and technologies to demonstrate substantial improvements in reliability and quantity of peripheral motor-signal information. Ultimately DARPA desires to develop clinically viable technologies, enabling wounded service members to control state-of-the-art prosthetic limbs.

Here's the good stuff:

Technical Area 1: Demonstrate clinically viable reliable-tissue interfaces to peripheral nerves or muscles. These interfaces must enable stable, robust, and high-performance recording of motor-signal activity.

Technical Area 2: Demonstrate the clinically viable tissue-interface electronics necessary to enable the development and testing of reliable peripheral interfaces designed to control many-DOF prosthetic limbs.

Technical Area 3: Demonstrate clinically viable algorithms and subsystem for reliably decoding motor-control signals from detected peripheral signals.

How ironic this BAA releases in TAE's Year of the Human-Machine Interface! What was it I said? Oh yes. "All you need is drivers." That would be technical area 3. I wrote, in a previous blog post about hijacking the nervous system, that you could essentially read and write all five senses if you found a way to directly interface with the nervous system. That would be Technical area 1. And here, in February, I discuss the idea that electronics that integrate with the human nervous system could revolutionize information retrieval. That would be technical area 2.
So don't worry, DARPA, I've got you covered.

Seriously, though, this is good research. While it is probably impossible for most federal agencies to get ridiculous research funded, not so for DARPA. However, by cloaking what I really think is their ultimate goal: cyborg soldiers, behind a veil of beneficent "help the walking wounded" research, they can really kill two birds with one stone. Sure, they want to have prostheses that are more actively and reliably responsive to the actual neural commands of the amputee. That would be fantastic if concepts like Dean Kamen's "DEKA" arm could have better musculoneural feedback. Or the RIC arm could have a better interface than pectoral stimulation.
However, what would be even better would be if a soldier could plug into a supersuit and control it with his muscles directly, and receive external information, like ally/enemy positions via integrated electronics. He could control targeting systems with his nerves directly, without moving. He could send vocal commands and feedback to his teammates via radio...without speaking into a radio. He could plug himself into a USB port at the end of a mission and download his memories of the mission as a high-tech version of a mission report.

While I really laud DARPA for funding this research to help injured vets, what I really feel could be the bread and butter here would be breaking the code of human nerve to digital electronics communication. That would revolutionize and open up a whole new frontier in bionic interfacing. Sign me up.


_