I had the privilege to work for one of the heroes in this video, Richard Little. My software was running in the battery pack of that machine. AMA
For some reason I just see a blank space where the video should be, so I watched it directly on youtube instead. It's easy to get inspired by videos like this, but I can't help but think that they offer an overly optimistic vision of science. The video itself involuntarily offers a cue: one scientist is trying to find a way to feed 2 more billions of people in a world where many are already starving to death, while another has dedicated his life to fighting malaria in Africa - if he's successful, his invention is going to make the overpopulation problem worse. He might be a hero to the people he saved, but what part is he playing in the grand scheme of things? To paraphrase Homer Simpson: "To science! The cause of... and solution to... all of life's problems." Anyway, back on topic: am I right to assume that the energy requirements are the main limiting factor on the speed that REX can currently reach? About a month ago I shared this article, that blamed poor battery efficiency for the robots' lack of stamina, and I would expect powered exoskeletons to run into similar problems. Do you think there's any chance that we're going to get a "running machine" any time soon or are we still decades away from that? I'd also like to know what it was like to be part of a start up, was it as stressful as they say?
I chopped some of the URL embellishment and it seems to display now; sorry about that. Energy requirements are not at all the speed-limiting factor in the REX device. It's a combination of the non-natural gait of the device, and the requirement to never injure the user (the two are intimately related). The battery pack can drive the device for over two hours continuously, so there's plenty of juice.
A primary design constraint is that the device must not fall - not in the event of system crash, or even in response to user movement (and wheelchair users often have quite violent muscle spasms). Its gait is entirely inhuman, because when taking a step, the device shifts to balance on one leg, then moves the lifted leg, then transfers centre-of-weight to balance on the other leg. In contrast, human walking is a controlled fall - if you were to 'freeze' in mid-step, you will topple over, and that's just not acceptable for this machine. Some other exoskeletons are meant to augment human movement, but that is a very different problem.
Its purpose is not to walk fast, but simply to walk (this machine can do stairs as well, though). Sitting all day is unhealthy for a walking person - so imagine the problem caused if you literally have to sit all day, every day. Most wheelchair users are given a "standing frame" that they are told to use for an hour or so every day for this very reason, but few people use them as much as they should. So the reason for this is primarily health benefits from walking and leg movement, and also for rehabilitation after injury.
It was amazing to be a part of this organisation in the early days - we literally worked in a warehouse the first two years, cold in winter and hot in summer; and we were not allowed to talk about what we were working on in public forums (until the company went public). Just a great group of people, and if they still had work for me to do, I'd be there still. The only stress was concern that we'd run out of funding and all have to find new jobs (which actually happened to me, but fortunately the company was able to survive).