3,628 views | 00:09:37
Search Icon
Invention and innovation are the mark of robotics engineer Dennis Hong’s lab. Watch as he demonstrates some of the remarkable robots that his team has built. The robots walk on three legs, ascend vertical cliffs, and climb over obstacles three times their height.
Episode filmed live at the 2010 World Science Festival in New York City. The full Cool Jobs program from that year can be viewed online.
Dennis Hong:
Thanks for having me. My name is Dennis Hong. I’m a professor in mechanical engineering. I get to build robots. How cool is that? All right. So how did it everything gets started? Well, when I was a young kid, when I was seven years old, I watched Star Wars for the very first time in Grauman’s Chinese Theater in Hollywood. I was completely blown away by all the cool, neat stuff, especially the interesting locomotion of R2D2 and the human/robot interaction with C3PO. Since then, I decided to become a robot scientist and I’m here today.
Dennis Hong:
So I truly believe that I have the coolest job in the world. Well, if you don’t believe me, let me show you. I’ll show you some of the cool robots that we’ve developed. So the first robot I want to talk about is called STriDER. It stands for self-excited, tripedal dynamic experimental robot. It is a robot that has three legs.
Dennis Hong:
Actually, we can see the robots over here. So it walks in a very interesting way. It swings a leg between the two legs and catch the fall. Isn’t that cool? And the body flips 180 degrees every time it takes a step. It’s pretty neat, but why are we trying to do this. Well, because of school? Yeah? But actually there’s a good reason why we’re not doing this. So we call this the concept of passive dynamic locomotion. We’re not doing this joint position control that you see in regular robots, but rather using the built in dynamics. It swings its leg. As a matter of fact, this, even though it has three legs, this is more close to human walking. You swing your leg, you catch the fall, stand up again, potential energy to kinetic energy and potential energy. Thus, it’s inspired by human bipedal walking. So, that is STriDER.
Dennis Hong:
So we don’t have this yet, but this is the next step we want to do. We want to fold up the legs, shoot it out long range motion, deploys its legs, almost like Star Wars, and it lands with this long legs, absorbs the shock. Then what you see over here, this yellow thing. This is not a death ray. Just want to show that if you have cameras because it’s tall, it’s 1.8 meters tall, it can see over obstacles like bushes. So it’s great for sensor deployment. So is it all animation and stuff? No, we actually have two prototypes. Just going to see our prototype number one, and prototype number two.
Dennis Hong:
One of the problems that we had with the early prototype number one is the top body was just too heavy. We had so many mortars and gears and those kind of things. So it was too heavy. So for the second version, we came up with this ingenious mechanical solutions. We synthesized these mechanisms and we got rid of all the motors and with just a single motor now we can align all the joints and now the top body is a light enough so it’s now walking in our lab. So this was our very first successful step. It’s pretty cool. Still, it’s a little bit unstable, so we still need to continue our work on STriDER.
Dennis Hong:
So the second robot I want to talk about is called IMPASS, intelligent mobility platform actuated spoke system. You can actually see our early prototype over here with a carbon fiber body. So this is a interesting new type of locomotion robot. It’s a wheel/leg hybrid robot. So it’s a rim, this wheel, or a spoke wheel, but the spoke individually moves in and out of the hub, as you can see. So it can rotate, but also moves in and out. So it’s a really a true leg wheel hybrid. We’re literally reinventing the wheel.
Dennis Hong:
So why are we trying to do this? Well, for example, if you have a rough terrain, it can adjust by using the contact touch sensor in the foot. You can adapt to the changing train, in this case it’s trying to walk over this cushion. But if it encounters a really, really tall, difficult obstacle, in this case, we have a big obstacle more than three times of its nominal height, then it switches to this deliberate mode and using the laser range finder and camera system it senses the obstacle size and shape, and it does intelligent planning and actually controls a spoke so it can actually surmount, go over these very challenged obstacles. You probably haven’t seen anything like that out there.
Dennis Hong:
This is probably the highest mobility robot for … application could be search on rescue missions or on Mars scientific exploration, anywhere that the wheels cannot go. This is probably going to be the perfect locomotion robot for those kind of situations.
Dennis Hong:
Okay. This next row is called HyDRAS, hyper degrees-of-freedom robotic articulated serpentine. It’s actually a family of snake type robots that not only slithers, but it actually climbs up structures by wrapping itself up. Among the HyDRAS robot this one is particularly interesting. It’s called HyDRAS Arm. So we have a standard snake type robot arm, but the cool thing is not the robot, but rather the user interface. As you can, see this black cord, that’s a simple optical fiber. This student, probably the first time using it, she just grasped it and just moves it, and the robot mimics the motion so it can actually generate a lot of type of complex motion. Again, first time using it, you can use this for doing complex operations.
Dennis Hong:
In the war zone there’s a roadside bombs. Currently, they send out these [inaudible 00:04:46] robots. It’s very expensive to train an operator to this. If you have something like this, this user interface is so intuitive, just pick it up and start using it. It’s called HyDRAS Arm.
Dennis Hong:
Okay. This is another different type of project. This project is called the blind driver challenge. Yes, I know what you’re thinking. We have devolved the world’s very first and only vehicle that can be driven by a blind person. So again, this student is completely blind. We do work with blind. It’s not something that we just developed something in the lab and give it to the blind community. We closely work with them. So the vehicle is equipment, many different types of sensors, laser range, finders, and camera systems, and the computer system builds a map around the vehicle called the world model that the computer can understand.
Dennis Hong:
We developed this many different type of non-visual user interfaces that give real time information, high band information to the driver who cannot see. You can see a big smile on her face. So we have a lot interesting things happening with this project. So you’re probably going to see a lot new, exciting things on the blind [inaudible 00:05:53] challenge early next year.
Dennis Hong:
Okay. This is currently our superstar robot called DARwin, dynamic anthropomorphic robot with intelligence. So as you know, I’ve been interested in human walking. So we started this project in 2004. At that time, this is something really new. So we didn’t know if is it even possible. What kind of motors can you use? What kind of control scheme can you use? So this was more of a feasibility study.
Dennis Hong:
Thus, it didn’t have any sensors, no feedback control. So for those who know about these kinds of things, if you don’t have any sensors, if there’s any disturbances, you know what happens, right? So we brought some of our robots today, the soccer playing robots. I want to give you a quick demonstration. Now, these robots, these are not the DARwin series robot that used for the autonomous robot soccer competition, but these are called [biolids 00:06:41], and we use these robots to train … There you go. We use these robots to train our humanly robot DARwins for playing soccer.
Speaker 2:
Now, how is he doing what he’s doing? Is he being directed by anyone from anywhere or is he doing this spontaneously?
Dennis Hong:
Yes. So these are currently just having …
Speaker 2:
So these ones are programmed. These ones [crosstalk 00:07:03].
Dennis Hong:
Yes, these are not fully autonomous. Oops.
Speaker 2:
Lost the ball.
Dennis Hong:
And this is why we don’t use these for playing soccer.
Speaker 2:
Thank you.
Dennis Hong:
Thank you. Oops.
Speaker 2:
These guys are great, Dennis, but I just don’t think they’re going to be beating Brazil anytime soon.
Dennis Hong:
I don’t think so. All right. I think that’s enough. All right, thank you. Okay. This robot is kind of interesting. This robot is called Charli, cognitive humanoid autonomous robot with learning intelligence. I think this is the first time we’re introducing this to the general public Charli, as you can see, as I mentioned, by the year 2050 you want to have full sight, humanly robots playing soccer against the human world cup champions.
Dennis Hong:
We’re not kidding. This is our first attempt to do this. This is a 1.5 meter humanoid robot, completely autonomous, untethered. And it’s the first of its kind in the United States. You’ve probably seen a robot called ASIMO from Japan, Honda. Right? Have you seen those robots? Right? And in Korea they have this humanoid robot called HUBO from KAIST. Now, United States, now we have Charli. So, let me introduce Charli. So this is Charli over here. Let’s wake him up. Hey, Charli.
Charli:
Wow. We have a large crowd today. Hello everybody. My name is Charli. Nice meeting you all.
Dennis Hong:
Nice meeting you too. It’s pretty cool. So unfortunately today were not able to bring the lower body, because all we’re still working on the lower body in the lab. So we only have the upper body.
Charli:
Oh, I knew they forgot something. Thanks, guys.
Dennis Hong:
Yeah. Sorry. So, Charli, what can you do?
Charli:
I can walk like you with two legs. I can see and recognize objects with my two camera eyes. I can track objects and grasp them with my hand. And most of all, I can play a game of soccer.
Dennis Hong:
There you go. So this is a video clip. This is early this year, the first time in running Charli. So, the most interesting thing, the most remarkable thing is that we started this project only a year and a half ago with 12 students with a seed funding of only $20,000 with some donations of parts from companies. As you can see, it’s completely untethered, bipedal walking. It might look simple, but this is something very, very challenging to do. Yet, we were the very first group in the United States that accomplished this with our students. So, very good.
Speaker 2:
Dennis Hong. Thank you, Dennis. Fantastic.
Invention and innovation are the mark of robotics engineer Dennis Hong’s lab. Watch as he demonstrates some of the remarkable robots that his team has built. The robots walk on three legs, ascend vertical cliffs, and climb over obstacles three times their height.
Episode filmed live at the 2010 World Science Festival in New York City. The full Cool Jobs program from that year can be viewed online.
Dennis Hong, a TED alumnus, is an associate professor and the founding director of RoMeLa (Robotics & Mechanisms Laboratory) of the Mechanical Engineering Department at Virginia Tech.
Read More
Dennis Hong:
Thanks for having me. My name is Dennis Hong. I’m a professor in mechanical engineering. I get to build robots. How cool is that? All right. So how did it everything gets started? Well, when I was a young kid, when I was seven years old, I watched Star Wars for the very first time in Grauman’s Chinese Theater in Hollywood. I was completely blown away by all the cool, neat stuff, especially the interesting locomotion of R2D2 and the human/robot interaction with C3PO. Since then, I decided to become a robot scientist and I’m here today.
Dennis Hong:
So I truly believe that I have the coolest job in the world. Well, if you don’t believe me, let me show you. I’ll show you some of the cool robots that we’ve developed. So the first robot I want to talk about is called STriDER. It stands for self-excited, tripedal dynamic experimental robot. It is a robot that has three legs.
Dennis Hong:
Actually, we can see the robots over here. So it walks in a very interesting way. It swings a leg between the two legs and catch the fall. Isn’t that cool? And the body flips 180 degrees every time it takes a step. It’s pretty neat, but why are we trying to do this. Well, because of school? Yeah? But actually there’s a good reason why we’re not doing this. So we call this the concept of passive dynamic locomotion. We’re not doing this joint position control that you see in regular robots, but rather using the built in dynamics. It swings its leg. As a matter of fact, this, even though it has three legs, this is more close to human walking. You swing your leg, you catch the fall, stand up again, potential energy to kinetic energy and potential energy. Thus, it’s inspired by human bipedal walking. So, that is STriDER.
Dennis Hong:
So we don’t have this yet, but this is the next step we want to do. We want to fold up the legs, shoot it out long range motion, deploys its legs, almost like Star Wars, and it lands with this long legs, absorbs the shock. Then what you see over here, this yellow thing. This is not a death ray. Just want to show that if you have cameras because it’s tall, it’s 1.8 meters tall, it can see over obstacles like bushes. So it’s great for sensor deployment. So is it all animation and stuff? No, we actually have two prototypes. Just going to see our prototype number one, and prototype number two.
Dennis Hong:
One of the problems that we had with the early prototype number one is the top body was just too heavy. We had so many mortars and gears and those kind of things. So it was too heavy. So for the second version, we came up with this ingenious mechanical solutions. We synthesized these mechanisms and we got rid of all the motors and with just a single motor now we can align all the joints and now the top body is a light enough so it’s now walking in our lab. So this was our very first successful step. It’s pretty cool. Still, it’s a little bit unstable, so we still need to continue our work on STriDER.
Dennis Hong:
So the second robot I want to talk about is called IMPASS, intelligent mobility platform actuated spoke system. You can actually see our early prototype over here with a carbon fiber body. So this is a interesting new type of locomotion robot. It’s a wheel/leg hybrid robot. So it’s a rim, this wheel, or a spoke wheel, but the spoke individually moves in and out of the hub, as you can see. So it can rotate, but also moves in and out. So it’s a really a true leg wheel hybrid. We’re literally reinventing the wheel.
Dennis Hong:
So why are we trying to do this? Well, for example, if you have a rough terrain, it can adjust by using the contact touch sensor in the foot. You can adapt to the changing train, in this case it’s trying to walk over this cushion. But if it encounters a really, really tall, difficult obstacle, in this case, we have a big obstacle more than three times of its nominal height, then it switches to this deliberate mode and using the laser range finder and camera system it senses the obstacle size and shape, and it does intelligent planning and actually controls a spoke so it can actually surmount, go over these very challenged obstacles. You probably haven’t seen anything like that out there.
Dennis Hong:
This is probably the highest mobility robot for … application could be search on rescue missions or on Mars scientific exploration, anywhere that the wheels cannot go. This is probably going to be the perfect locomotion robot for those kind of situations.
Dennis Hong:
Okay. This next row is called HyDRAS, hyper degrees-of-freedom robotic articulated serpentine. It’s actually a family of snake type robots that not only slithers, but it actually climbs up structures by wrapping itself up. Among the HyDRAS robot this one is particularly interesting. It’s called HyDRAS Arm. So we have a standard snake type robot arm, but the cool thing is not the robot, but rather the user interface. As you can, see this black cord, that’s a simple optical fiber. This student, probably the first time using it, she just grasped it and just moves it, and the robot mimics the motion so it can actually generate a lot of type of complex motion. Again, first time using it, you can use this for doing complex operations.
Dennis Hong:
In the war zone there’s a roadside bombs. Currently, they send out these [inaudible 00:04:46] robots. It’s very expensive to train an operator to this. If you have something like this, this user interface is so intuitive, just pick it up and start using it. It’s called HyDRAS Arm.
Dennis Hong:
Okay. This is another different type of project. This project is called the blind driver challenge. Yes, I know what you’re thinking. We have devolved the world’s very first and only vehicle that can be driven by a blind person. So again, this student is completely blind. We do work with blind. It’s not something that we just developed something in the lab and give it to the blind community. We closely work with them. So the vehicle is equipment, many different types of sensors, laser range, finders, and camera systems, and the computer system builds a map around the vehicle called the world model that the computer can understand.
Dennis Hong:
We developed this many different type of non-visual user interfaces that give real time information, high band information to the driver who cannot see. You can see a big smile on her face. So we have a lot interesting things happening with this project. So you’re probably going to see a lot new, exciting things on the blind [inaudible 00:05:53] challenge early next year.
Dennis Hong:
Okay. This is currently our superstar robot called DARwin, dynamic anthropomorphic robot with intelligence. So as you know, I’ve been interested in human walking. So we started this project in 2004. At that time, this is something really new. So we didn’t know if is it even possible. What kind of motors can you use? What kind of control scheme can you use? So this was more of a feasibility study.
Dennis Hong:
Thus, it didn’t have any sensors, no feedback control. So for those who know about these kinds of things, if you don’t have any sensors, if there’s any disturbances, you know what happens, right? So we brought some of our robots today, the soccer playing robots. I want to give you a quick demonstration. Now, these robots, these are not the DARwin series robot that used for the autonomous robot soccer competition, but these are called [biolids 00:06:41], and we use these robots to train … There you go. We use these robots to train our humanly robot DARwins for playing soccer.
Speaker 2:
Now, how is he doing what he’s doing? Is he being directed by anyone from anywhere or is he doing this spontaneously?
Dennis Hong:
Yes. So these are currently just having …
Speaker 2:
So these ones are programmed. These ones [crosstalk 00:07:03].
Dennis Hong:
Yes, these are not fully autonomous. Oops.
Speaker 2:
Lost the ball.
Dennis Hong:
And this is why we don’t use these for playing soccer.
Speaker 2:
Thank you.
Dennis Hong:
Thank you. Oops.
Speaker 2:
These guys are great, Dennis, but I just don’t think they’re going to be beating Brazil anytime soon.
Dennis Hong:
I don’t think so. All right. I think that’s enough. All right, thank you. Okay. This robot is kind of interesting. This robot is called Charli, cognitive humanoid autonomous robot with learning intelligence. I think this is the first time we’re introducing this to the general public Charli, as you can see, as I mentioned, by the year 2050 you want to have full sight, humanly robots playing soccer against the human world cup champions.
Dennis Hong:
We’re not kidding. This is our first attempt to do this. This is a 1.5 meter humanoid robot, completely autonomous, untethered. And it’s the first of its kind in the United States. You’ve probably seen a robot called ASIMO from Japan, Honda. Right? Have you seen those robots? Right? And in Korea they have this humanoid robot called HUBO from KAIST. Now, United States, now we have Charli. So, let me introduce Charli. So this is Charli over here. Let’s wake him up. Hey, Charli.
Charli:
Wow. We have a large crowd today. Hello everybody. My name is Charli. Nice meeting you all.
Dennis Hong:
Nice meeting you too. It’s pretty cool. So unfortunately today were not able to bring the lower body, because all we’re still working on the lower body in the lab. So we only have the upper body.
Charli:
Oh, I knew they forgot something. Thanks, guys.
Dennis Hong:
Yeah. Sorry. So, Charli, what can you do?
Charli:
I can walk like you with two legs. I can see and recognize objects with my two camera eyes. I can track objects and grasp them with my hand. And most of all, I can play a game of soccer.
Dennis Hong:
There you go. So this is a video clip. This is early this year, the first time in running Charli. So, the most interesting thing, the most remarkable thing is that we started this project only a year and a half ago with 12 students with a seed funding of only $20,000 with some donations of parts from companies. As you can see, it’s completely untethered, bipedal walking. It might look simple, but this is something very, very challenging to do. Yet, we were the very first group in the United States that accomplished this with our students. So, very good.
Speaker 2:
Dennis Hong. Thank you, Dennis. Fantastic.
