DARPA’s four-legged, load-bearing robotic mule was already impressive. AlphaDog could trot around on all fours for 20 miles without refueling. It could carry up to 400 pounds of gear up hills. It could even pick itself back up if it got knocked over. Now, it can think. Start stocking up the fallout shelter, folks. Things could get ugly if Alphadog decides it’s not our friend any more.The Pentagon’s been working on different versions of AlphaDog, formerly known as Big Dog, since 2005, and it’s getting close to being battle ready. The latest upgrade includes smart software that enables the robotic mule to follow human instructions. Using the Marine Corps Tactical Robot Controller (TRC), a ten-pound touchscreen device that works as an interface for the soldiers in charge, AlphaDog will follow a platoon at “tight” range or trail further back. It’s also ten times quieter than it used to be, so it’s feasible this thing could accompany infantry squads on their patrols.In the near future, the Marines want to hook their radios up to AlphaDog so they can bark verbal commands in the field, and the robo-beast will do their bidding. “We want the operator to be able to command it to come here, sit, stay. Likewise, the robot would be able to tell the operator OK, I'm stuck, I'm here, wait,” Lt. Col. Joe Hitt, Darpa's program manager told Wired. The next generation of AlphaDogs will also be twice as fast, capable of running seven to 10 miles per hour. (The Pentagon already has a sprinting robot-beast called Cheetah that can run over 28 miles per hour.)This is all well and good until the AlphaDog’s thinking chip shorts out, and the robo-beasts turn on us. Just imagine these cow-sized monsters with machine guns strapped on their backs, trotting through the streets of Manhattan, looking for victims.But seriously, that will never happen. After all, these things are getting pretty sophisticated, but we’re a long ways off from some Matrix-level robot takeover. After all, they run on go-kart engines.
Advertisement