Working with Robots!

In the future, humans will work closely with robots in all aspects of life, including in space. Construction of the International Space Station is clear evidence of this, as astronauts have performed many spacewalks in tandem with the Canadarm2. As we look to the future, these operations will continue and will become more complex; we will continue to work on the ISS using Dextre, the latest robotic addition to the station, we will build colonies on the moon, we will eventually send humans to Mars, where they will work with rovers to explore the surface.

Our job in the research lab is to plan for these activities and figure out ways to make them not only possible but more efficient and safer for the astronauts. We've got a lot of ideas and have installed a few of them in the MX-2, including voice recognition and a Heads-Up Display.

Voice recognition software running onboard the Mac Mini in the suit allows the suit subject to control suit functions, robots, or really anything we can dream of using voice commands. The subject is wearing the Communication Carrier Assembly or "Snoopy Cap", which includes headphones and a microphone. The voice signal is amplified inside the electronics box, and split with one line going into the Mac Mini, and the other line hard wired into the lab's communication system. This is done for safety reasons, if the Mac Mini crashes, the suit subject will still be able to communicate with the safety divers and all the technicians and engineers on the deck. The onboard software meanwhile parses the voice signal and recognizes certain specific commands, which it then translates into commands for the suit or robots. For example, the suit subject can obtain information about the suit, can change the video feed in his Heads-Up Display, can start various programs on the Mac, etc.

The Heads-Up Display (HUD) uses a very small LCD screen in the upper right half of the helmet to display information to the subject. Checklists can be displayed dynamically, so that the subject can go through tasks as needed. As the subject performs a wiring task, for example, he/she can use voice to ask for a wiring diagram to be displayed on the HUD. This is all done autonomously, with no input from the technicians on deck. The Mac Mini recognizes the request, and displays the wiring diagram. When the subject is finished, they again with voice request the checklist to reappear as they no longer need the wiring diagram.

In the next few months we plan to integrate this system with our team of robots, Ranger and SCAMP. Ranger is a two-armed robot that also can provide a mobile platform for the subject. With some development, the suit subject will be able to use their voice to ask Ranger to bring them anywhere in the tank, and then work with Ranger on construction tasks. SCAMP meanwhile is a flying camera platform, a floating eyeball which can obtain a camera view anywhere in the tank. The subject can use their voice to request this view in their HUD, and then use the voice to maneuver SCAMP around, seeing whatever SCAMP sees. This gives the astronaut another set of maneuverable eyes, thus reducing the amount of time and effort spent translating along the structure.

Eventually these teams of humans and robots will be able to work seamlessly together to perform very complex tasks. These are the first steps towards such Human-Robotic Interactions.