Professor, Computer Science and Electrical Engineering
Gordon Marshall Chair in Engineering
University of Southern California
Los Angeles, CA, USA
- Bachelors, Electrical Engineering, University of Lisbon, Portugal
- Masters, Electrical Engineering, University of Rochester, NY, USA
- PhD, Electrical Engineering, University of Rochester, NY, USA
Requicha is Professor of Computer Science and Electrical Engineering at the University of Southern California, where he focuses on nanorobotics.
Advice to Students:
"I would advise the average high-school student to learn your math and your physics and chemistry as much as you can. But if you're really very smart, and you find that high-school's boring, you can try to hook up with somebody in research, over the summer or after hours, and go do something in the labs, or work in theory."
Q: When did you first find that your career path focused on nanotechnology?
Requicha: In 1994, I was tired of what I was doing, and didn’t think that what I was doing was going to have sufficient impact, so I looked for a change of direction. I ran into a friend that was in the same situation, who said "There's this thing called nanotechnology." We were both working in areas related to robotics, so we read a bit about nanotechnology and decided to try something in robotics at the nanoscale. Then by luck or serendipity, USC got a few million dollars from a gentleman named Mike Markkula, who was one of the founders of Apple. He gave the university the money to be used to support some high-risk, high-payoff, interdisciplinary effort. The university made a call for proposals, and most everyone responded. Our proposal was to do nanorobotics….and our proposal was accepted. It was actually to do something that has still not been done, 17 years later. We were going to take DNA and image it with atomic force microscopes and then cut it and edit it. We didn’t succeed in that, but we succeeded in lots of other things.
Q: What current nanotechnology applications are you working on?
Requicha: I spent 16 or 17 years running a large lab, called the Laboratory for Molecular Robotics, and eventually got tired of trying to keep it funded, and decided to pass the directorship to a younger guy. I am now working on a related area: suppose you would be able to build the nanorobots as autonomous devices. I'm exploring what you would do with them -- or more to the point, how would you program them given that one robot wouldn’t be able to do much of anything. You would need thousands or maybe millions! How do you program a million robots? In principle, this could be done at any scale, but nobody's going to build a million robots with meter dimensions. But, if you are working at the nanoscale, you probably would want to build these large numbers of them. So I am working on self organization of robot swarms. It's very difficult to work on programming swarms unless you have some kind of a specific task. The task I have been investigating is building shapes.
So why did I get into this? First, because before this I did a lot of computational geometry, so I am interested in a lot of geometrical things, and computerized design. The potential practical applications would be to manufacture things, artifacts of that scale. I would want to build some kind of a device of a particular shape, but it's difficult to do that at the nanoscale. However, it might be possible to convince a whole bunch of robots to create that shape, and once the robots are arranged in that shape, you could somehow either glue them all together and say that’s what you want. Or, alternatively, have the robots have some kind of attachment placed in them, and then once they have made the shape, use the shape as a template or as a scaffold, and then throw something over it, like particles, and these particles would attach to the spots you have designated in each robot, and then you could let go of the robots, and use them for something else.
In reality, this started because my knees and my hips are not what they used to be, and I kept thinking that if one could take a whole bunch of robots, inject them in the joint, tell them somehow that the shape that it should have is a knee, never mind what’s there, just cover it, make the result be the right shape, then that would be very nice. One could think of these scaffolds as being for electronics, or for building organs, or repairing bones and organs, and things of that sort. That's the potential application, but we are nowhere near being able to do that. In fact we don’t have the physical robots at that scale, so we have been working in simulation, and we've shown some very interesting things in simulation. We have actually shown that you can build arbitrary shapes pretty much automatically, except for minor details in the simulation. That is the sort of work that I'm doing these days.
Q: What's the most rewarding thing about working with nanotechnology?
Requicha: Well, I have always been working on things that have two characteristics. One is that they have to be intellectually interesting. The other one is that, if you succeed, it should have impact.
Q: Is there an example you can provide that shows how something you’ve worked on has positively impacted the world?
Requicha: Yes, I've worked on geometric modeling -- solid modeling -- and it was mathematically interesting, and it also had a big impact because the work we did is now used by everybody. All the mechanical engineers use solid modeling, which is what I worked on. Now everything is 3-D. As to my nanotech work, I think is too early to have had any major impact.
Q: What do you think is the single greatest impact nanotechnology has had on the world thus far?
Requicha: Mine hasn’t had a great big impact yet. But an example of nanotechnology that is beginning to be used, and will have impact very soon is targeted drug delivery. In some ways this is a "simple" application, but if you can do anything that is even a little bit better then what's being done today -- for example treating cancer -- you'll save a lot of lives. At the moment, there's a lot of people I think who get killed by the cure. Basically what you do for cancer treatment is give people some nasty poison, and that kills cancer cells, but kills the other ones too. So if you can target drug delivery, that's an improvement, and it’s beginning to work.
Q: Please give an example of what you envision nanotechnology applications leading to in the future.
Requicha: My long term view, and why I actually got into the nanorobotics area, was this vision that you will be able to build swarms of these things and put them into the body. What you do is build a very large sensor actuator network. You'll have these sensors all over your body, and they'll be continually looking for lots of things, infection, bacteria, nasty stuff -- it's almost like having another immune system, looking for something that doesn’t belong there. And since you have a lot of them, and they're all over the body, the system will be very sensitive. You will find, for example, that when there is some infection, you would find the infection when there is only a very small number of bacteria. What you do today is wait until there are millions of bacteria, and then you go to the doctor and say "Oh, I feel terrible." And the doctor says "Yeah, you got a big infection, take these antibiotics," and a week later you're okay, hopefully. If you had the ability to detect diseases very early by having a sensor network, and your devices could not only sense but be able to actuate, to do something, like kill the bacteria very early on, you would have basically changed the current medical paradigm, which is that you wait for people to feel terrible, and then give them some heavy-duty medication. With the nanosensors and nanoactuators, if you had some bacteria in your body, the system would find them and kill them, but you would never know that you were going through it. We are pretty far from doing this -- but I believe it's possible.
Q: Do you find yourself working more in a team situation, or more alone?
Requicha: Nanotech is very interdisciplinary, which is good and bad. I personally find interdisciplinary work very rewarding because you learn all sorts of things that you had no idea about, and you never had studied carefully, but you learn it from your team members. But we don't have renaissance people anymore, things have become too complicated. You need a team!
Q: If you work more as a team, what are some of the other areas of expertise of your team members?
Requicha: Well, in my lab I had myself and another guy, we both are robotics and computer science people, and there was a chemist, a material scientist, another chemist, and an electrical engineer, so we have a big spread of skills, and they are all necessary. And there is not one person in the team who really has depth in all of these. The interdisciplinary worker in my view is somebody with depth in a particular field, but with superficial knowledge about the rest of what's going on. No one can have depth in all these disciplines -- but you have to be able to understand what the other guys are saying, and to interact with them, and to adapt to their culture. It's not just knowledge, there are cultural things.
Q: Did your university training help you in your nanotechnology work?
Requicha: Yes, and no. First of all, nanotechnology didn’t exist when I went to school, which can be said for a lot for other things; geometric modeling didn’t exist either. And most of the interesting things that people are working on now, probably did not exist when I went to school. My university did prepare me with a lot of math and physics and basic science. And that's actually what I tell people who ask me "how do I get prepared to work in nanotechnology?" Just get some serious basic science. I took my first electrical engineering degree in Portugal, a six year course. I'm trying to remember if there is anything in the technical stuff that I learned, that I have ever used. We were kind of a backwater country; we were learning electronics with vacuum tubes. By the time I graduated, there wasn’t a vacuum tube anywhere in sight, everything was solid state. Professors hadn’t gotten around to changing their courses to teach solid state. So when you get into technology, per se, technology can get dated very quickly.
Now math doesn’t get dated very easily -- and even physics. But some disciplines have changed in amazing ways, the best example is probably biology. Think about all this stuff that people now take for granted, DNA and genes, and the fundamental dogma of genetics. When I learned biology in high school, it didn’t exist; people didn’t know any of this. But, math for example, yeah it keeps evolving, but the basic stuff is there, and believe me, you may need all that stuff. And it's very hard to say that this part you're going to need and that part you're not going to need.
Q: Do you have a mentor? Did you in your college years?
Requicha: Yeah, I've had several, some had a more direct impact than others. Your professors and mentors have a very profound effect on you, and the curious thing is that, when I started on something, I never knew what I was going to do. It’s an evolving process: you start doing something and maybe you like it, maybe you don't, you evolve as you do it. And if I think about it, I think I always ended up moving into areas where there was a professor or some mentor that I liked, who had the knowledge and the passion to do something. And you say to yourself, "oh that looks interesting." There are other areas that abstractly might be more interesting, but you don't get exposed to them, or you get exposed to them with someone who really turns you off. My relationship with mentors was always fairly distant, except for my thesis advisor, with whom I had a very close relationship.
Q: If you had to do it all over again, would you still focus on nanotechnology applications?
Requicha: Yes, it's interesting because the potential impact is mindboggling. I like the story about the physicist who was working on some theory or another and was telling one of his friends about it. He said, "this will never work……but if it did!"
Q: What advice do you have for pre-university students?
Requicha: It's difficult to envision what to do in nanotechnology at the high-school level. And there is a big range of capabilities in people. I would advise the average high-school student to learn your math and your physics and chemistry as much as you can. But if you're really very smart, and you find that high-school's boring, you can try to hook up with somebody in research, over the summer or after hours, and go do something in the labs, or work in theory.