Dennis Hong: My seven species of robot

So, the first robot to talk about is called STriDER. It stands for Self-excited Tripedal Dynamic Experimental Robot. It's a robot that has three legs, which is inspired by nature. But have you seen anything in nature, an animal that has three legs? Probably not. So, why do I call this a biologically inspired robot? How would it work? But before that, let's look at pop culture. So, you know H.G. Wells' "War of the Worlds," novel and movie. And what you see over here is a very popular video game, and in this fiction they describe these alien creatures that are robots that have three legs that terrorize Earth. But my robot, STriDER, does not move like this.

So, this is an actual dynamic simulation animation. I'm just going to show you how the robot works. It flips its body 180 degrees and it swings its leg between the two legs and catches the fall. So, that's how it walks. But when you look at us human being, bipedal walking, what you're doing is you're not really using a muscle to lift your leg and walk like a robot. Right? What you're doing is you really swing your leg and catch the fall, stand up again, swing your leg and catch the fall. You're using your built-in dynamics, the physics of your body, just like a pendulum. We call that the concept of passive dynamic locomotion. What you're doing is, when you stand up, potential energy to kinetic energy, potential energy to kinetic energy. It's a constantly falling process. So, even though there is nothing in nature that looks like this, really, we were inspired by biology and applying the principles of walking to this robot. Thus it's a biologically inspired robot.

What you see over here, this is what we want to do next. We want to fold up the legs and shoot it up for long-range motion. And it deploys legs -- it looks almost like "Star Wars" -- when it lands, it absorbs the shock and starts walking. What you see over here, this yellow thing, this is not a death ray. (Laughter) This is just to show you that if you have cameras or different types of sensors -- because it is tall, it's 1.8 meters tall -- you can see over obstacles like bushes and those kinds of things.

So we have two prototypes. The first version, in the back, that's STriDER I. The one in front, the smaller, is STriDER II. The problem that we had with STriDER I is it was just too heavy in the body. We had so many motors, you know, aligning the joints, and those kinds of things. So, we decided to synthesize a mechanical mechanism so we could get rid of all the motors, and with a single motor we can coordinate all the motions. It's a mechanical solution to a problem, instead of using mechatronics. So, with this now the top body is light enough. So, it's walking in our lab; this was the very first successful step. It's still not perfected -- its coffee falls down -- so we still have a lot of work to do.

The second robot I want to talk about is called IMPASS. It stands for Intelligent Mobility Platform with Actuated Spoke System. So, it's a wheel-leg hybrid robot. So, think of a rimless wheel or a spoke wheel, but the spokes individually move in and out of the hub; so, it's a wheel-leg hybrid. We are literally re-inventing the wheel here. Let me demonstrate how it works. So, in this video we're using an approach called the reactive approach. Just simply using the tactile sensors on the feet, it's trying to walk over a changing terrain, a soft terrain where it pushes down and changes. And just by the tactile information, it successfully crosses over these type of terrain.

But, when it encounters a very extreme terrain, in this case, this obstacle is more than three times the height of the robot, Then it switches to a deliberate mode, where it uses a laser range finder, and camera systems, to identify the obstacle and the size, and it plans, carefully plans the motion of the spokes and coordinates it so that it can show this kind of very very impressive mobility. You probably haven't seen anything like this out there. This is a very high mobility robot that we developed called IMPASS. Ah, isn't that cool?

When you drive your car, when you steer your car, you use a method called Ackermann steering. The front wheels rotate like this. For most small wheeled robots, they use a method called differential steering where the left and right wheel turns the opposite direction. For IMPASS, we can do many, many different types of motion. For example, in this case, even though the left and right wheel is connected with a single axle rotating at the same angle of velocity. We just simply change the length of the spoke. It affects the diameter and then can turn to the left, turn to the right. So, these are just some examples of the neat things that we can do with IMPASS.

This robot is called CLIMBeR: Cable-suspended Limbed Intelligent Matching Behavior Robot. So, I've been talking to a lot of NASA JPL scientists -- at JPL they are famous for the Mars rovers -- and the scientists, geologists always tell me that the real interesting science, the science-rich sites, are always at the cliffs. But the current rovers cannot get there. So, inspired by that we wanted to build a robot that can climb a structured cliff environment.

So, this is CLIMBeR. So, what it does, it has three legs. It's probably difficult to see, but it has a winch and a cable at the top -- and it tries to figure out the best place to put its foot. And then once it figures that out in real time, it calculates the force distribution: how much force it needs to exert to the surface so it doesn't tip and doesn't slip. Once it stabilizes that, it lifts a foot, and then with the winch it can climb up these kinds of thing. Also for search and rescue applications as well.

Five years ago I actually worked at NASA JPL during the summer as a faculty fellow. And they already had a six legged robot called LEMUR. So, this is actually based on that. This robot is called MARS: Multi-Appendage Robotic System. So, it's a hexapod robot. We developed our adaptive gait planner. We actually have a very interesting payload on there. The students like to have fun. And here you can see that it's walking over unstructured terrain. It's trying to walk on the coarse terrain, sandy area, but depending on the moisture content or the grain size of the sand the foot's soil sinkage model changes. So, it tries to adapt its gait to successfully cross over these kind of things. And also, it does some fun stuff, as can imagine. We get so many visitors visiting our lab. So, when the visitors come, MARS walks up to the computer, starts typing "Hello, my name is MARS." Welcome to RoMeLa, the Robotics Mechanisms Laboratory at Virginia Tech.

This robot is an amoeba robot. Now, we don't have enough time to go into technical details, I'll just show you some of the experiments. So, this is some of the early feasibility experiments. We store potential energy to the elastic skin to make it move. Or use an active tension cords to make it move forward and backward. It's called ChIMERA. We also have been working with some scientists and engineers from UPenn to come up with a chemically actuated version of this amoeba robot. We do something to something And just like magic, it moves. The blob.

This robot is a very recent project. It's called RAPHaEL. Robotic Air Powered Hand with Elastic Ligaments. There are a lot of really neat, very good robotic hands out there in the market. The problem is they're just too expensive, tens of thousands of dollars. So, for prosthesis applications it's probably not too practical, because it's not affordable. We wanted to go tackle this problem in a very different direction. Instead of using electrical motors, electromechanical actuators, we're using compressed air. We developed these novel actuators for joints. It is compliant. You can actually change the force, simply just changing the air pressure. And it can actually crush an empty soda can. It can pick up very delicate objects like a raw egg, or in this case, a lightbulb. The best part, it took only $200 dollars to make the first prototype.

This robot is actually a family of snake robots that we call HyDRAS, Hyper Degrees-of-freedom Robotic Articulated Serpentine. This is a robot that can climb structures. This is a HyDRAS's arm. It's a 12 degrees of freedom robotic arm. But the cool part is the user interface. The cable over there, that's an optical fiber. And this student, probably the first time using it, but she can articulate it many different ways. So, for example in Iraq, you know, the war zone, there is roadside bombs. Currently you send these remotely controlled vehicles that are armed. It takes really a lot of time and it's expensive to train the operator to operate this complex arm. In this case it's very intuitive; this student, probably his first time using it, doing very complex manipulation tasks, picking up objects and doing manipulation, just like that. Very intuitive.

Now, this robot is currently our star robot. We actually have a fan club for the robot, DARwIn: Dynamic Anthropomorphic Robot with Intelligence. As you know, we are very interested in humanoid robot, human walking, so we decided to build a small humanoid robot. This was in 2004; at that time, this was something really, really revolutionary. This was more of a feasibility study: What kind of motors should we use? Is it even possible? What kinds of controls should we do? So, this does not have any sensors. So, it's an open loop control. For those who probably know, if you don't have any sensors and there are any disturbances, you know what happens. (Laughter)

So, based on that success, the following year we did the proper mechanical design starting from kinematics. And thus, DARwIn I was born in 2005. It stands up, it walks -- very impressive. However, still, as you can see, it has a cord, umbilical cord. So, we're still using an external power source and external computation.

So, in 2006, now it's really time to have fun. Let's give it intelligence. We give it all the computing power it needs: a 1.5 gigahertz Pentium M chip, two FireWire cameras, rate gyros, accelerometers, four force sensors on the foot, lithium polymer batteries. And now DARwIn II is completely autonomous. It is not remote controlled. There are no tethers. It looks around, searches for the ball, looks around, searches for the ball, and it tries to play a game of soccer, autonomously: artificial intelligence. Let's see how it does. This was our very first trial, and... Spectators (Video): Goal!

Dennis Hong: So, there is actually a competition called RoboCup. I don't know how many of you have heard about RoboCup. It's an international autonomous robot soccer competition. And the goal of RoboCup, the actual goal is, by the year 2050 we want to have full size, autonomous humanoid robots play soccer against the human World Cup champions and win. It's a true actual goal. It's a very ambitious goal, but we truly believe that we can do it.

So, this is last year in China. We were the very first team in the United States that qualified in the humanoid RoboCup competition. This is this year in Austria. You're going to see the action, three against three, completely autonomous. There you go. Yes! The robots track and they team play amongst themselves. It's very impressive. It's really a research event packaged in a more exciting competition event. What you see over here, this is the beautiful Louis Vuitton Cup trophy. So, this is for the best humanoid, and we would like to bring this for the very first time, to the United States next year, so wish us luck. (Applause) Thank you.

DARwIn also has a lot of other talents. Last year it actually conducted the Roanoke Symphony Orchestra for the holiday concert. This is the next generation robot, DARwIn IV, but smarter, faster, stronger. And it's trying to show off its ability: "I'm macho, I'm strong. I can also do some Jackie Chan-motion, martial art movements." (Laughter) And it walks away. So, this is DARwIn IV. And again, you'll be able to see it in the lobby. We truly believe this is going to be the very first running humanoid robot in the United States. So, stay tuned.

All right. So I showed you some of our exciting robots at work. So, what is the secret of our success? Where do we come up with these ideas? How do we develop these kinds of ideas? We have a fully autonomous vehicle that can drive into urban environments. We won a half a million dollars in the DARPA Urban Challenge. We also have the world's very first vehicle that can be driven by the blind. We call it the Blind Driver Challenge, very exciting. And many, many other robotics projects I want to talk about. These are just the awards that we won in 2007 fall from robotics competitions and those kinds of things.

So, really, we have five secrets. First is: Where do we get inspiration? Where do we get this spark of imagination? This is a true story, my personal story. At night when I go to bed, 3 - 4 a.m. in the morning, I lie down, close my eyes, and I see these lines and circles and different shapes floating around. And they assemble, and they form these kinds of mechanisms. And then I think, "Ah this is cool." So, right next to my bed I keep a notebook, a journal, with a special pen that has a light on it, LED light, because I don't want to turn on the light and wake up my wife.

So, I see this, scribble everything down, draw things, and I go to bed. Every day in the morning, the first thing I do before my first cup of coffee, before I brush my teeth, I open my notebook. Many times it's empty, sometimes I have something there -- if something's there, sometimes it's junk -- but most of the time I can't even read my handwriting. And so, 4 am in the morning, what do you expect, right? So, I need to decipher what I wrote. But sometimes I see this ingenious idea in there, and I have this eureka moment. I directly run to my home office, sit at my computer, I type in the ideas, I sketch things out and I keep a database of ideas. So, when we have these calls for proposals, I try to find a match between my potential ideas and the problem. If there is a match we write a research proposal, get the research funding in, and that's how we start our research programs.

But just a spark of imagination is not good enough. How do we develop these kinds of ideas? At our lab RoMeLa, the Robotics and Mechanisms Laboratory, we have these fantastic brainstorming sessions. So, we gather around, we discuss about problems and social problems and talk about it. But before we start we set this golden rule. The rule is: Nobody criticizes anybody's ideas. Nobody criticizes any opinion. This is important, because many times students, they fear or they feel uncomfortable how others might think about their opinions and thoughts.

So, once you do this, it is amazing how the students open up. They have these wacky, cool, crazy, brilliant ideas, and the whole room is just electrified with creative energy. And this is how we develop our ideas.

Well, we're running out of time. One more thing I want to talk about is, you know, just a spark of idea and development is not good enough. There was a great TED moment, I think it was Sir Ken Robinson, was it? He gave a talk about how education and school kills creativity. Well, actually, there are two sides to the story. So, there is only so much one can do with just ingenious ideas and creativity and good engineering intuition. If you want to go beyond a tinkering, if you want to go beyond a hobby of robotics and really tackle the grand challenges of robotics through rigorous research we need more than that. This is where school comes in.

Batman, fighting against bad guys, he has his utility belt, he has his grappling hook, he has all different kinds of gadgets. For us roboticists, engineers and scientists, these tools, these are the courses and classes you take in class. Math, differential equations. I have linear algebra, science, physics, even nowadays, chemistry and biology, as you've seen. These are all the tools that we need. So, the more tools you have, for Batman, more effective at fighting the bad guys, for us, more tools to attack these kinds of big problems. So, education is very important.

Also, it's not about that, only about that. You also have to work really, really hard. So, I always tell my students, "Work smart, then work hard." This picture in the back this is 3 a.m. in the morning. I guarantee if you come to your lab at 3 - 4 am we have students working there, not because I tell them to, but because we are having too much fun. Which leads to the last topic: Do not forget to have fun. That's really the secret of our success, we're having too much fun. I truly believe that highest productivity comes when you're having fun, and that's what we're doing. There you go. Thank you so much. (Applause)