Hit enter after type your search item
Wzy Word


How to Start a Robot Revolution | Part 1 | Breaking the Wheel


(light clicks) – [Interviewer] And what's this little, what's this? – This is a fun little thing This is a thing that we call Tinybot, and we were working on a project that was actually about security in ROS, and we needed to demonstrate that we had made a secure system that you couldn't hack into

And so we actually built an entire Lego town And then had robots like this running around, and we showed when security was turned off, that the bad guy could take over the robot and then mow down the Lego people And then when the security was turned on, the bad guy couldn't take over the robot So we saved the Lego town (upbeat music) – There's really, there's technology, there's the software and then there's the people

(upbeat music) – Willow Garage was this really kind of strange place, just strange enough to be able to bring the community together – It was to make robotics better as fast as possible – The results we got were because it was open source, it was a really good idea It was the right value proposition for those grad students, and they told their friends, and it just, spread through the entire robotics community – The next thing you knew, we had a huge project that half the world was using to do robotics

(upbeat music) – We were working on trying to make robotics happen as an industry So moving it out of just research labs and into the real world That was our goal And one of the great things that came out of it was the Robot Operating System (upbeat music) (soft music) – So this started for me when I was starting my PhD program with my partner in crime in this, Eric Berger

Now, we studied a bunch of things going on in robotics at the timeyou talked to Brian Gerkey, right? So at the time, there was a platform called Player/Stage, which was a very successful open source project in robotics, and basically, like robots would drive around So we talked of reusing that we're like, why isn't this used in more advanced robotics? And that's when we got this message of like, look, nobody's even close to standardizing how did this, the same system should work

– So for a long time, in robotics, it had been the case that you would get your hand on, get your hands on a robot, and then you would write a whole bunch of what we would call infrastructure: software to go with it And that was just to make the robot vaguely work Then after that, you could do what you actually wanted to do, which was to do an experiment, solve a problem, test a hypothesis And the goal with ROS was to make it so that we took care of all that (upbeat music) – ROS is what makes it possible for body parts of the robot to talk to other body parts, and for humans to filter through all of those messages that come spewing out of the machine, picking out the ones that we care about, and using that to build applications on top of

That's how I think of it – Software has to combine together and if you want to do a big, multi-person effort, you have to have a framework for that ROS was that framework – What it means is that if you want to develop a new robot, you don't have to start from scratch You don't have to do the fundamental, sensors talking to computers, you don't have to do the fundamental navigation

You don't have to do the fundamental perception That's all built into ROS – That notion, that core idea that I can use ROS and take my piece of this, but combine it with everybody else's best of breed stuff That's a really powerful idea and that idea spread like wildfire (keyboard clacking) – They really started to cement this idea in a strong way that you can have platforms that are beyond just a standalone arm, or standalone mobile base, you can have a very complex, standard platform that has all of this capability

And I think giving away the first 11 PR2s had a big impact in that way (upbeat music) (crowd applauds) (crowd cheers) – PR2 is a robot designed from the ground up, to enable software developers to focus on new ideas and new technology – PR2 stands for Personal Robot Two – It was a combination of like, the sci-fi idea of a robot and the reality of what is required to make a robot that can do all of the things that you want a robot to do – I think of the PR2 as really a research platform for trying out ideas and prototyping, but no one wants that thing in their house Like it will destroy your carpets first of all, and freak out your dog

The PR2 was not the point It was practice for ROS (upbeat music) – This week, doing the normal things, working out a few new processes with the ROS 2 Technical Steering Committee – My memory recalls a, I believe it was a Friday afternoon might have been happy hour, sitting outside of Willow Garage, and Brian proposed this idea to me about splitting off to form the Open Source Robotics Foundation (upbeat music) – And that's really the point where we took responsibility for ROS

– They took up the gauntlet, right? They kept the ROSCon conferences going they kept the community growing – And that's really accelerating, not just robotics research, but robotics industries everywhere – One of the robots behind me we created in five months from scratch, like hardware, software, everything It was running and it was riding an elevator, right? How do you do that that fast? The answer is ROS, right? There's no way we could have done it without that (upbeat music) – ROS really transformed everything

– It was changing a culture that was very insulated to a culture of, how do we make things reproducible – The community would not be at the size it is and people would not be having the commercial impact they're having, certainly, if we had gone a different way with how we licensed it – And then without that one common platform that everybody's building on, and contributing to, and enabling people to build on each other, without that common platform that brings the community together, robotics would definitely not be where it is today (upbeat music)

Source: Youtube

This div height required for enabling the sticky sidebar