share interactive transcript request transcript/captions live captions
download
|
MyPlaylist
FRANCIS C. MOON: This is the third lecture in Kinematics of Machines and Mechanisms. And today, we're going to talk about two different technologies that use mechanisms and they are robotic medicine and also the aviation. Let's see if this is working.
So we're going to talk about robots in medicine and the evolution of flying machines. And we're also going to talk about Cornell and the early history of aviation. And then we're going to end with rules on how to invent a new machine. And that's part of what you're learning in this course.
It's interesting. I've probably only put up one formula, all right? And that formula was Gruebler's formula for given the number of links and joints, how many degrees of freedom do you Have But the main ideas are quality. Main ideas are topological. And one message that I give in the course I teach in robotics is that geometry rules.
We can think about artificial intelligence, and all kinds of computer algorithms, and electronic sensors, but in the end, the way you put together your systems in some sort of a geometric or other type of architecture, whether that's a algorithmic architecture or a physical architecture, that determines a lot what's going on. So sometimes we get caught up in analysis in the details. So these lectures are about big topological questions.
And I start with robots in medicine. And in the other lectures, we talked about how Leonardo da Vinci had, not necessarily invented mechanisms, but he recorded them, which suggests that in the Renaissance, people were using all kinds of machines to either manufacture textiles, or to fight wars, or trebuchets. And in his book he had many sketches of machines.
Really, they were not even books. They were collections of his drawings. And he was also interested in anatomy. And so he did a lot of the drawings in anatomy. And the left figure shows his anatomical picture of the arm.
Franz Reuleaux of Berlin, in the late 19th century, was also interested in animal and human anatomy. But by that time, because we had so many machines that could develop energy and move around, scientists began to think of the human body-- or the animal body-- as like a machine. So you could think of the walking animals and humans as a collection of links and joints.
And so you could think of the organs in the body also as machines. For example, if you could think-- here's a spherical pump model. And I know some of you have to design some sort of pump, and most of you are going to use the cylinder configuration. But this was a design that used a spherical configuration. And you could almost think of this as kind of a heart like. You could run this as a heart type of machine.
So of course by the 20th century, we were now replacing the image with mechanical things. We were replacing them with cells and all kinds of electrical circuits and connections. But at some level, the human and animate beings are still have some mechanical aspects, and some of it in terms of links and joints.
We also talked about, for example, using a four-bar mechanism to replace the motion of a knee. So if we had a four-bar system like this, this was attached to the lower leg, this to the femur here, then this can act as a knee. And instead of having fluid, you could replace the function of the fluid with a four-bar.
And up on the right-hand side you can see a modern four-bar knee prosthetic. And you can Google this and find a number of manufacturers who make these four-bar knee prosthetics. So we are, in the 21st century, still using links, and joints, and mechanisms to replace human function.
Now, in recent years-- and this is really now a 21st century invention, although it started in the late 20th century-- and that is robot-assisted surgery. And the idea is we have a patient here and we have one or more arms-- robotic arms-- and there may be a camera on the arm or a camera up above. And the surgeon never looks at the patient.
So when you are criticized by your parents for playing video games for hundreds of hours, you just say listen, I'm training to be a modern 21st surgeon, all right? And I've been to lectures with surgeons who were more my age and who learned to use this here and a younger surgeon who had grown up playing Pong or some other video game, and the younger surgeon said it was easier for him to learn this new robot-assisted surgery than the older, who had been used to the open surgery.
So the idea is that this is a so-called master/slave concept where there will be an image. The image will be much clearer. The image will be 3D. And the surgeon's looking at that image, and the surgeon's operating some controller here, and that's operating the arms. And we'll see the topology of these machines.
There were several earlier ones, one called AESOP, approved-- this is FDA, 1994. The most popular one right now is the so-called da Vinci Surgical-Assist Laparoscopic Robot. The company is Intuitive Surgical, is a big commercial for them. The prices have now gone up, probably to more like 1 and 1/2 million. And there are now hundreds of these now in the United States and around the world. This other one, Zeus was an earlier one, which was bought out by Intuitive Surgical, and I don't think it's in service anymore.
In the 20th century, humans had invented and developed so-called industrial robot arm. This is one that's still being made by EPSON. This is like a two-link. There's one link here and then there's a vertical prismatic joint here. It's generally used for pick-and-place operations in electronic assembly.
And as you can see, this evolves into the robot assistant. So you have, again, a kind of SCARA robot arm, but now there are three of them. And in some cases, there are four of them, OK? And so this is going to hover over-- now these mechanisms here are very interesting mechanisms. And we'll talk about that in a minute.
If we look at the comparing-- let's say the traditional open surgery with the da Vinci surgery-- this is gynecological surgery-- here you have an open surgery. We had eight to ten-inch incision. With the da Vinci arm laparoscopic, you had about five one-inch incisions. The operation takes longer, however, because you have to set up the machine and the patient.
The operation with the abdominal takes less time. However, recovery is longer in the case of the open surgery, and is a shorter-- maybe even one-day stay for the new type of surgery. And laparoscopic surgery has been around a long time.
In most of the hand-operated systems, you have the surgeon operating physically, and you may have some sort of a camera inside to see what you're doing. And oddly enough, the country which has the most skill in this is the French. And it's interesting how the da Vinci machine got approved in the United States.
One surgeon told me that the company had developed this surgical robot but couldn't get FDA approval. And they went to Europe where it was more easier to get-- and some French surgeons asked if they could play with it. And so they played with it, played with one of these machines in France, and eventually this reached FDA approval.
So, again, interdisciplinary international cooperation, we'll see that later when we talk about the aviations. So here is the so-called Zeus. Again, you'll see one link here, another link here, and this other device here.
And now this made the prostatectomies, which is removal of the male prostate for prostate cancer. Enlarged prostate is another problem which this has been used. Hysterectomies for female problems, heart surgery, mitral valve repair, brain neurosurgery, and orthopedic surgery, hip replacement has been used.
So here is the-- this comes from the patent office. And this may be an early one. And this is a three-arm with so-called RCM mechanisms. So these things here are called RCM mechanisms. And this is a so-called remote center of motion.
So during the operation, these arms do not move. The image of robot surgery is not that you're being grabbed and sliced by these arms, OK? The arms get into position. The arms get into position. They make the incision. And now the laparoscope is going to go into something called a trocar. And the robot is going to move it this way.
And the whole idea is can you move this where there's a center which is not moving? The center is attached to the patient. You don't want this center-- you'll tear the patient apart. So you want to create a mechanism no matter what you do here, this remains-- this center remains fixed, see? So that's called a remote center of motion mechanism. And there are lots of different types of these.
This is one here. So no matter what happens, when this goes through the patient, some point here, when this moves, this can go in and out of some point here. It's not going to move. And when I read about this, I wondered if we had some of these in these 19th century mechanisms in the remote collection. And low and behold, I found that we did.
We did have one right here, which-- this would be the remote center here. So I can go this way, and I can go this way, and mechanically, this point is not going to move. And they're probably a dozen or more other types of remote center of motion mechanisms. Maybe they were invented a century or more ago, and may they'll be rediscovered again in the present technology.
So from a mechanism point of view, that's a very important part of the design of these machines. The other thing is end effect. And so it's one thing to get the laparoscope in. What are you going to do inside? And so that involves also mechanisms.
And this is from Intuitive Surgical. You see a sort of a clamp-type of mechanism. And the other thing here is these companies brag that they can get wrist-like motions, OK? They can get wrist-- your pitch and also roll, OK?
And so the interesting thing about the laparoscopic idea, when you just-- when you're going in like this, if you move it up, this goes down. So you have to get your brain turned upside down. But with these new systems, you have a kind of a gimbals-like mechanism.
So here you see there's one axis there. There's another axis there, and another axis there. So it's almost like a gyro. Perhaps I'm missing-- oh, here. It's right here. So if you look at the gimbals that's used in the gyro-- so if I have rotation about this axis, and I have rotation about this axis, and rotation about this one, that's the same topology.
And by the way, this type of topology can be traced back to Tibet and China because people had-- and also mariners on boats, when they wanted to have a lamp, the lamp had maybe oil in it. When the ship pitched, they didn't want the oil to drop out, so they invented a kind of a gimbal. So when the ship was rolling and pitching, the lamp would remain fixed.
So this type of topology has its origins maybe hundreds or 1,000 years ago. And now it's being used again in this robot surgery. There are advantages and disadvantages. The disadvantages of robot surgery is longer set-up time, longer operation time.
Right now there's no haptic feel. Are some of you interested in med school after you finish? A few. And one of the things with current surgery right now is that when you take the knife and you move it, you can feel something, OK? That's called haptic. That's feeling the forces.
Right now it's all kinematic. It's like playing a video game. You're clicking something and you don't feel anything. It just moves on the screen. But there are companies right now trying to develop that. If you go back to this here, if you could put in some feed backs-- in other words, you're putting in some breaks, some motors here at these joints here.
Then as you're moving something and it's supposed to be cutting something, you're actually going to feel. You're actually going to feel what it-- you can sense it, OK? So that's coming down the line. The other thing is training and learning curve for a surgeon.
So if you have to have one of these operations and you say, well, would you like the real-- the surgeon on top of you or the surgeon sitting in a console, you might say, gee, that'll be really cool if I could have this robot surgery. Well then you better ask the surgeon, how many of these have you performed with this machine? He or she better say a couple of hundred. Then you say, what happened to the first hundred?
[LAUGHTER]
Anyway, and the other thing, one of you had multiple arms? Well, you have some arms that are going to manipulate things. There's another arm that has a camera inside. It has two eyes, and the two eyes give you 3D picture. This is from early patents. And you see this is a slide. And I don't know whether they're using this here.
One of the difficult things for you as engineers-- perhaps in your class, they've talked about benchmarking. What's benchmarking? Have you talked about benchmarking? What's benchmarking? What's benchmarking?
SPEAKER 1: It's making sure your product is competitive enough to--
FRANCIS C. MOON: Making sure your product is competitive with other manufacturers. Well, go online and try to figure out what's behind the da Vinci robot arm. Almost impossible. Intuitive Surgical has put out beautiful color pictures. But it's like looking at a Mercedes Benz. Do you know what kind of engine's inside by looking at the outside of a Mercedes Benz?
Well, one place you can find out where-- if your competitor is patented, you can go to the patent literature and this is what we found. That's how we got that other-- in fact, one of the students in the class found this. However, it's not going to say robot surgical arm. It's going to use the lawyers.
That's why if you're going to have a team invent a new machine with you, you better get not only a venture capitalist to provide you with cash, but you also need some lawyers and maybe some patent attorneys to figure out which jargon would they use, because they use the jargon to hide this in the literature. So that's a little hint in terms of-- because this is another one from the same company. And you can begin to see each one of these numbers is described in the patten description.
So you may be able to get a clue as to how they're getting the degrees of freedom and how they're actuating. By the way, they're actually with cables. This is like 19th century. And even if you go back to Leonardo's time, they were using belt and pulley systems. So again, you see this continuum from the Renaissance 19th century, 20, 21st century. And we each keep-- we keep adding but we don't always throw everything away.
So this is an action, and you say, what is this mound here? That's the patient. Is this an obese patient?
SPEAKER 2: Yes.
FRANCIS C. MOON: Somebody said yes.
SPEAKER 3: No.
FRANCIS C. MOON: Somebody said no. What do you think they've done?
[INTERPOSING VOICES]
Somebody. Somebody. What do you think they've done? Yes?
SPEAKER 4: Filled them with water.
FRANCIS C. MOON: They've filled them with air, I believe. I think it's air. So they put the air in to get some space. All right? So the patient looks heavier than they are. And there are the arms. And basically, the main machine is not moving. What's moving is the laparoscopic tools.
How many people here, if you had a serious operation, how many people here-- this is a good test is how the public will be interested in new technology. If you had to have a surgery, and the doctor says, would you like the robot surgery, the da Vinci robot arm, or would you like to have it done by Dr. X, who's done it 1,000 times? Let's say both ways, the robot surgeon has done it 1,000 times and the conventional one.
Let's show of hands. How many people would have it done the robot way? How many people that the conventional way? Look at this. Not even the engineers trust the technology. Probably two to one. [LAUGHING]
So you can see how difficult it is to introduce. This technology is now almost a decade old. Every hospital in the country wants one of these. By the way, your health care costs are going to go up, because you engineers are inventing all kinds of wonderful things for the doctors. But even then, only one out of two of you would want to have this surgery.
All right. So there's other ways in which you can use robots. One of them is patient services. One of them is in the service industry. And one of the important early pioneers is Joseph Engelberger, who gave a lecture here about a decade ago on the future of robotics, and he says the future of robotics is not an industry where we have maybe a million robots. The future is in service industry.
So one of the areas where you may have heard about-- and maybe you can buy this for your birthday or whatever-- is Roomba. How many have heard of Roomba? So the Roomba is now sort of the largest selling sort of robot that does something.
And this is one from we have a professor of robotics, professor Gazit Cress, and she loaned me this little Roomba here. And it has various sensors that detect impact, but it also senses-- I'm going to stand in front of it. If it bangs into things, you see, it's stuck.
Oh, it detects the edge, too. All right. And what's underneath? Just two wheels. That's it. How many degrees of freedom does this have? How many degrees of freedom would this have?
SPEAKER 5: Three.
FRANCIS C. MOON: Three. Why? What's the three? Two position, X, Y, and rotation. So this is not good for an industrial room. You need another three. So you could get another three if you put another arm on this, all right? So this is one of the types of robots that could be used in hospitals.
One of the big costs is cleaning, keeping the germs out. So this is one of the areas where in the future we might see robots enter, but not at the surgical level or the medical level, but in the service areas. Another one is a so-called exoskeleton, where people have lost their ability to walk, and also military operations.
I'm going to-- you may have heard of CyberKnife. There's no knife. It delivers radiation. Again, it's a six-axis machine. It's oriented so that it can direct radiation to a particular part of the patient with extreme accuracy. But there's no real knife. It's delivering radiation. It's another expensive piece of equipment.
Here's another SCARA. There's one arm, two arm, there's the other one there. This is used to help people in stroke therapy. And there's one introduced at Hopkins earlier. It's your friendly doctor robot. And instead of having the doctor make rounds, the doctor sits in his office or her office. And this machine visits the patient and says, hello, how are you? I'm Dr. Smith. How are you feeling today?
So maybe it delivers pills, but there's no hands-on with this, right? So I don't think this is going to go very far. But some company tried to sell it. So now I want to talk about design issues for medical robots because this is a class on mechanical design and design in general.
First of all, cleanable surfaces, sterilization, no oil discharges, no electrical sparking hazards-- you don't want any arcing inside-- fail safe shutoff-- if something goes wrong, it shuts down-- optimal surgical insertion ports, path planning for surgery, patient-robot registration capability-- how do you find out where the patient is relative to the machine-- and a vital signs monitoring interface, and manual moving of arms in case of power failure. That's some of the things that you have to worry about if you're going to invent these machines.
And by the way, a lot of these issues it's not biology, it's mechanical. So it's possible for you to go into biomedical area and be a mechanical engineer or electrical engineer. Because most of these technical questions are not necessarily bio. You talked some bio person, you communicate. But they want to know how can we design, or improve, or maintain this to operate both mechanically, electrically, and also in terms of computers.
Now I'm going to talk about machine evolution. And one of the things I want to talk about is that how do we invent things? And one of the big questions for the 21st century is who will invent the great machines of the 21st century? Right now, we have these tragedies in New Zealand and Japan in terms of natural disasters.
But another type of challenge is occurring, and many nations-- certainly Europe, North America, China, and India-- are looking to see how can they develop the next great inventions, the next great machines. And so one of the things you find out-- you've talked about working in groups. But there's also a social interaction beyond the group.
For example, this is a picture of this is Leonardo da Vinci here around 1500, but he got some of his ideas from another man named Mario Taccola from 1450. How did he get those ideas? Because Taccola had a student named Francesco di Giorgio, and Leonardo had Francesco di Giorgio's book. And Giorgio was a little older than him. So he may have copied from di Giorgio. Giorgio certainly copied from Taccola.
Then there were many of these other engineers. Some of them were artist engineers. Some of them were architect engineers. And it trickles down. And here's the 19th century. Here's Reuleaux. And Reuleaux would say, well, I learned some of this stuff from Leupold from 1724. And Leupold copied from somebody in the 16th and the 17th century. And so you have all of those connections.
Is there such a thing as the genius working in his or her basement, working by and discovers something new? For example, if we look at Leonardo's social-- I'm sorry this gets a little blurry. Here's Leonardo right here. In his own time, there were a whole bunch of other engineers, including Francesco di Giorgio, in which he had contact with. He would meet them. He would meet them in the street. He would meet them in his workplace. And so there was a whole network of these types of engineers.
So the question is, are machines created by geniuses? And in particular, we'll look at who invented the airplane. So the common theory of who invented the airplane-- What's the common name? If I say, who invented the airplane, what would you say?
SPEAKER 6: Wright brothers.
FRANCIS C. MOON: Wright brothers. Right away. But that's not the answer you'd get if you were in Great Britain. That's not the answer if you were in California, by the way, or Connecticut. That's not what you would get-- the answer you would get if you were in Germany or France. However, the current popular idea is yes, there were the Wright brothers and maybe there was the German Lilienthal.
And then there was the British Cayley. This was around 1800. And maybe they all got their ideas from Leonardo. But anyway, this is the [INAUDIBLE] there's a book written. We stand on the shoulders of giants. What does that imply? We're not worthy of the people who went before us? Yes?
SPEAKER 7: Well, it's more like we have great people that sort of formed our basis on which we come up with our own ideas.
FRANCIS C. MOON: So do you think if we got rid of all geniuses, if they just say-- if like Ayn Rand. Have you ever read Ayn Rand, The Fountainhead? All the industrialists, the movers and shakers decide to opt out. they go to some place in the west and say, we'll see what happens to civilization.
So if all the geniuses sort of said, we'll opt out of here. We're going to play video games out in the desert, all right? Do you think civilization would progress? Well, it's a setup question, right? Because this is the network.
So here is the-- where's the Wright brothers? There's the Wright brothers right here. Here is a man named Octave Chanute. Octave Chanute. And the reason I introduce Octave Chanute-- you've probably never heard of Octave Chanute. Now you're all students in The Sibley School of Mechanical Engineering. In 1890, there was the Sibley College of Mechanical Engineering. And Robin Thurston was the director, and he invited his friend Octave Chanute to come and give lectures at Cornell on progress in flying machines to students just like you.
And Chanute then collected information from earlier. This is Cayley. This is around 1810. And he wrote and looked up all kinds of books, and he put out a book, Progress in Flying Machines. And when the Wrights decided they were going to get interested in flying machines in 1899, they wrote to the Smithsonian. And they wrote to the Smithsonian, and Rathbun-- right there-- was a Cornell graduate in history, not an engineer.
And he gets the letter from the Wrights. And the Wrights say, we're interested in flying. Do you have any information on flying machines? Now Rathbun was kind of a secretary to the director. The director was Samuel Langley, who himself was working on flying machines.
Rathbun could have taken that letter and said, this is a bunch of cranks and just thrown it away. He didn't. He took the letter and gathered material from the libraries, including the book by Chanute, and gave them to the Wrights. And that is the beginning of the Wrights beginning in flying.
Not only that, then they wrote to Chanute and said, by the way, we're trying to develop ideas about flying. Well, Chanute, ten years earlier, had already started doing gliding. So he said, they exchanged about 400 letters over five or six years. So the point is here that the creation of the airplane involved a network.
On this page here, there's about 50 different nodes. These are called nodes. This is very similar to Ruleaux's idea of a mechanism. So these are like the joints, and these are the links. A very similar idea. And a network of invaders from 1810 to 1910. If you really flesh this out, it's probably near 100. There's probably at least 100 nodes, and the links probably are in the hundreds. I traced at least 300 links.
So the decisions that had to be made included the lift device. Do we use wings or air screw? Propulsion, flapping or propelling? Engine, steam or gasoline? Do we push-- the early ones-- or we pull? How many wings? Mono, bi, triplane? Control, wing walking, rigid mechanisms, air lines. So all of those, in fact, some of the things that people were interested in is--
I brought this umbrella in. My mother told me it was bad luck to open an umbrella indoors. But guys, if anything happens today, you can blame it on me or my mother, God rest her soul. So the question is this is a parachute-type of thing. There were people in the early 19th century who said, OK, I'll get up here and I'll jump off this thing. Why didn't it slow me down? What was the problem? Yes?
SPEAKER 8: You didn't develop enough air resistance.
FRANCIS C. MOON: Air resistance is one thing, but what's the other problem? I'm too heavy. So there's some ratio between the weight and the wing area, right? So they had to figure out basic things like that. And by the way, they studied birds. So you have ornithology was very important to the development of the aircraft.
Now one of the things-- they said Leonardo sort of invented the airplane. He was mainly interested in flapping devices. So he was interested in OK, if I have two wings like this, and I flapped the wings like this-- because he's looking at small birds kind of flapping. But he didn't get-- he never built it. But later in the 19th century, there were people who built flapping machines, including Lilienthal.
Lilienthal believed that you could do with flapping machines, but humans are too big. And so people began to look at the plane. Why is it called an airplane? Because people were looking at a plane. At first it was just a flat plane. And then they looked at cambers, and they learned about lift over cambers. Then they learned about dihedral like this. So that took a whole century to figure that out.
The other thing here is the topology. Right now, we put the vertical and horizontal stabilizers at the back. In some of the earlier designs, like in the Wright brothers, part of it was in the front. They had one stabilizer in the back, another in the front. The other was where were you going to put the propeller? In the front was called a tractor. But the first ones were pushers.
So you had all of those topological decisions that had to be made. And because the solution was not necessarily the ordinary flapping birds, the solution was the big soaring birds. The big soaring birds. And both the Wrights and others had looked at eagles, and buzzards, and cranes. And then they saw the warping of the wing when they needed some control.
This is George Cayley, his design for a early machine. This was around 1810- 1850. This is 1850. This is Chanute here, and this is his network right here. And you see one of the nodes in this [INAUDIBLE] is Cornell, another is Thurston, another is a man named Zahm, who was here just like you are here right now. And he worked with Chanute to develop an international meeting.
He also became a friend of the Kurds. One of the ones I want to talk about here-- this is the Chanute glider developed from a triplane to a biplane. And here is Lilienthal. He thought you should shape the wings like a bird.
But that, again, was the wrong decision. He was killed in 1896 after several hundred trials, maybe 500 or more trials, and then he was killed. If he was not killed, he was ready to put an engine on this. He would have been the first.
By the way, how does the connection get back to Cornell? You've heard of six degrees of separation, right? Lilienthal was a student of Ruleaux at the Berlin Technical University. And Ruleaux knew Thurston, and Ruleaux had saw the models of these mechanisms, too.
So these models here may have been used-- or some of the one-- by Lilienthal at Berlin, and then copies of that are sent to Cornell. And then you are now using mechanisms to learn about machines. So let's see how connected all of this technology is. Another person is Charles Manly-- a month before the Wright brothers successful test-- we're going to end in three minutes--
SPEAKER 9: Yes.
FRANCIS C. MOON: Professor Langley from the Smithsonian had built a full scale plane and he needed an engineer to build the engine. And he asked his friend, Robert Thurston, do you have an outstanding young engineer? He said, a moral engineer. And that man was Charles Manly.
He graduated in the class of 1900, just a century behind, and he became the engineer for Langley. And though the plane went down in the Potomac-- Manly almost lost his life-- Manly's contribution is he built the lightest gas engine. He built a five-cylinder, radial engine and it had the greatest horsepower per wing of that time.
So let me say that the development of heavier-than-air machines was a natural outgrowth of the industrial revolution. And it grew out of a social network of craftspeople, engineers, and scientists. So the airplane was not the invention of a few geniuses. It was an invention of a group of talented engineers, craftspeople, and businesspeople working together. So thank you very much.
[APPLAUSE]
This three-part lecture series was given by legendary retiring Joseph C. Ford Professor of Mechanical Engineering Francis Moon on the topic of kinetics as it relates to robotics on June 10, 2011, at Cornell University.
Professor Moon focuses on the kinematics of machines as applied to aviation and robotic medicine. He discusses the early history of aviation at Cornell as it relates to the evolution of flying machines and covers the development of robotic medicine going all the way back to Leonoardo da Vinci and Franz Reuleaux.
