Diving is a very safe sport, as long as you follow all the rules. And there are a lot of rules: about how deep you can go, how long you can stay down and when not to throw altitude changes into the mix. It's about the accumulation of gas in your blood, and DARPA wants to use technology to help Navy divers turn into Aquamen.
Here's the deal with diving: water weighs a lot. And when you're under the water, your lungs and chest have to be able to expand with all of the water above you pushing down on them. The only way to do this without wearing a pressurized suit is to inhale air at the same pressure as the water around you, which is what divers do. This is generally not a problematic thing, except that the high pressure air forces gas into your tissues, and when you go up and the pressure on your body decreases, that gas will start to expand and form bubbles. If you're not careful and those bubbles get too big and end up in your joints and brain and eyeballs, they can cause you to pop like a bottle of champagne.
Recreational divers try to prevent all of this bad stuff from happening by carefully keeping track of how deep they go and for how long, by ascending slowly to allow pressurized gas to be exhaled before it gets dangerous, by spending time on the surface between dives so that gasses don't build up in their tissues, and by not flying for 24 hours after a dive. The military, however, isn't diving recreationally, and it's looking for ways to use technology to make it safe for humans to do things underwater that rec divers would consider to be crazy and suicidal. Here's what DARPA wants to be possible in an "extreme combat dive profile:"
- Insertion via military free fall from 35,000 feet altitude.
- Combat dive down to 200 feet for at least 120 minutes.
- Surface and immediately begin a second dive of variable, increasing depth to 200 feet for at least 20 minutes without needing to decompress.
- Extraction in an unpressurized aircraft to 14,000 feet altitude.
Arguably, the free fall insertion from 35,000 feet is the least dangerous thing on this list, which should tell you something about how risky this sort of thing is. What DARPA is looking for is some sort of technological solution that combines "control algorithms, physiologic sensors, gas sensors, and gas control components" in an "integrated microsystem device" that "enables extreme combat diving with limited risk of complications." The device will accomplish this by continually analyzing a diver's blood with a compact ultrasound array to check for bubble formation, and then administering small amounts of nitric oxide when necessary to help counter the problem.
This is part of DARPA's 2013 solicitation for Small Business Innovation Research proposals, which is how DARPA communicates that they have something really cool that they'd like to have happen and someone should just go out and make one already. Often with things like this that do somehow manage to happen, commercialization eventually follows, meaning that at some point, recreational scuba diving could get much safer for everyone.