Tip:
Highlight text to annotate it
X
My name is Jonathan Hurst and I am an Assistant Professor here at Oregon State in Mechanical,
Industrial, and Manufacturing Engineering. But, my work does span across Electrical and
Computer Science as well. We do robotics here in the lab and specifically physical interaction
tasks. The theme of the lab is understanding what we call passive dynamics, and that is
the mechanical components of the robot. So you will see on this machine for example,
there are springs. These are fiberglass springs and these are the same purpose as your tendons
there as stretchy and flexible so when this thing is hopping up and down a lot of that
energy is going into and out of those springs as part of energy storage. That really enables
the behavior of the device if we did not these springs and there were just motors attached
to the leg it would not be physically possible to hop up and down or to run efficiently,
no matter what software you wrote, or no matter what sort of controls you had.
Our big focus right now is building robots that can walk and run outside in the real
world over reasonably rugged terrain. We are going to want robots to be always working
with humans, around humans, in human environments, and even on humans. We are talking about like
locomotion applies directly to prosthetic limbs, to exoskeletons that can help disabled
people get around, or for military applications so that the soldiers can carry a lot more
or go farther. This is our very first prototype of Atreus, is the name of this machine. It
is a monopod robot, means just a single leg, it has been designed specifically to match
a very simple mathematical model which looks a lot like a pogo stick, just a mass on top
of a spring. That model has been used to describe animal locomotion and has been investigated
by a lot of mathematicians and controls people. What we would like to do is make a machine
that matches this model as closely as possible so we can control it, we can understand it
and so on. We built this based on what I learned doing
my Ph.D. work at Carnegie Mellon University at the robotics institute building Thumper
here. So Thumper is also a monopod, and this as part of my Ph.D. work I had hopping around
the room there. Things that I learned from this we were able to apply to the design of
Atreus. Thumper was also built into a bipedal version, so just two of these legs exactly
the same put together. That robot is named, Mabel and is at the University of Michigan.
One of the things that impressed me when I arrived here is that Oregon State students
know how to build stuff. This place is really great for applied engineering. Students can
not only do the math and analysis but they know how to weld and machine and put together
a device that is going to function and work. That has really been helpful, that works well.
It is a good fit for me; it’s a good fit for this lab. That sort of culture doesn’t
exist everywhere. A lot of times now when I watch science fiction movies it is just
ruined because all of the technical parts are wrong. It’s hard when it eliminates
my suspension of disbelief, I can’t do it anymore, when I say the whole premise of the
movie is flawed so this is dumb. We want to have robots moving and doing things,
I like to be in here when there is a robot hopping around and an arm moving over there
and every week we are taking the thing outside maybe to the practice football field once
we have got this off the boom and in 3-D. We are running the tests and the robots going
to run around for a while then something breaks so we bring it back to the lab to work on
it for a week and then take it back out again. That is where I want to be.