By tenth grade, I had it figured. I’d been captivated by science and technology since before I could remember. My formative years were spent experimenting, building electrical and mechanical things, and dreaming of traveling in space. Thinking carefully about how I might best satisfy those passions, I constructed a plan for my career. First, I would graduate from high school. Then attend college and graduate school. I would get a PhD in physics and use my knowledge to invent a method of faster than light drive. That would earn me a place on a spaceship bound for Alpha Centauri, our sun’s nearest stellar companion. Anything after that would just be gravy.
Plan
Things began auspiciously enough. I completed high school, first in my class. (That becomes less impressive when you learn that my tiny high school graduated only 38 seniors my year.) Then, through the most unlikely of circumstances, I went to MIT. After college I began graduate school. But at that point the first glitch in my well-honed plan cropped up. Which branch of physics should I pursue—plasma physics, astrophysics, condensed matter physics? While my classmates were busy reading research papers and finding their niches, I was adrift. Getting a PhD is all-consuming, and I hadn’t yet found a subdiscipline of physics that compelled me to work that hard.
Patch
I came up with a patch for my plan. I would leave graduate school, get a physics-related job, and then after further reflection, find my niche and return to the plan.
The job I found was on the staff at the William H. Bates Linear Accelerator. Bates was operated by MIT and located in Middleton, MA. For a technology nerd and life-long tinkerer, several wonderful aspects recommended life at Bates. There were massive magnets for steering and focusing the beam, I learned the cool physics of linear accelerators, there was an enormous, multi-ton spectrometer that separated the results of particle collisions for analysis, and there were computers and analytic instruments galore.
Working at Bates was often fun—my last project there was to develop a control system (written in assembly language) that replaced manual magnet adjustments with central computer control. But particle physics was the lab’s focus and sadly, it failed to excite me. After three years I felt I was settling into a rut at Bates. I hadn’t found my passion, so I needed to try something different.
Trip
I decided to take a trip around the world. What would come afterwards, was unclear.
Leaving Boston in June 1981 I traveled west, visiting friends across the United States. In September I arrived in Anchorage, Alaska. From there I flew to Tokyo. To stretch my funds, I tried to live on only $5 per day. That rendered the Tokyo subway too expensive so, I did a huge amount of walking to see sights around the capital.
After visiting several cities across Japan I took a ferry to Okinawa (encountering a typhoon along the way that transformed a one-day passage into a four-day exercise in nausea control) and then a plane to Taiwan. From there I went to Hong Kong. The month was December. When I’d left home, China had been closed to travelers, but by the time I got to Hong Kong things had changed. A visit to a small office in Hong Kong produced a visa that allowed me to travel anywhere in China!
By pure luck I was among the first visitors to the country after it opened. I journeyed to Beijing and walked on the Great Wall; I went to Chongqing and took a boat trip down the Yangtze River through the Three Gorges. (Before the new dam made that impossible.) I saw the amazing landscapes and caves around Guilin.
After China, I visited Singapore for a few days, then spent a month traveling around New Zealand, from the Bay of Islands in the north to Christchurch in the south. Next, in Australia, I visited a “hippy town” (Nimbin) and saw wild koalas, kangaroos, and an echidna in Halls Gap. I hitchhiked from Adelaide to Perth.
I spent four days in Bangkok and then four weeks trekking in Nepal. In remote Muktinath I saw the natural eternal flame, flickering above water (where gas bubbled from the earth) in a temple.
I flew to New Delhi and visited the Taj Mahal and then spent 24 hours on a train traveling to Mumbai. From there I flew to Athens, with a stopover in Dubai. I spent a month on a kibbutz in Israel, saw the pyramids in Egypt then flew to Rome. I’d met several young European travelers along the way, and I visited some of them in their homes as I traveled across Europe. But what I failed to encounter in my twelve-month trip was my passion.
Passion
By June 1982 I was back in Boston. My trip had consumed my savings so my only career plan at that point was to get a job. I found one at the MIT Artificial Intelligence Lab. Probably I was offered a position on the research staff because of my having built the accelerator control system for Bates.
Surprisingly, when I arrived at the AI Lab—seven years after graduating from college—I found my passion waiting patiently for me there. The excitement of the lab, the huge promise of the new technology, and the chance to be a part of it all quickly redirected me from physics to robots.
I formulate a new career plan. I would help to bring about the robot revolution by participating in research at the Lab for the rest of my career. I spent nine happy years at the Lab during which time I worked on a motion planning project, wrote two books, and built the earliest direct ancestor of the robot that would become Roomba. That robot and I even appeared on TV. [“Ten O’Clock News; Robot Talent Show,” 2/3/1989, WGBH-TV]
Then in 1991 fate messed with my career plan. The lab ran out of money, and I was laid off.
But well before my last day at the Lab I found another robot job at Denning Mobile Robotics. Sadly, Denning wasn’t doing well, and my job lasted just six months. It didn’t help that the president was thoroughly unimpressed with the little floor cleaning robot I proposed and prototyped during my brief stay. That triggered me being laid off.
Luckily, iRobot had just been founded a year earlier and soon after losing the job at Denning they hired me to do the thing I loved: work on robots. The approach iRobot took to robotics was founded on the behavior-based robot control paradigm Prof. Rod Brooks had invented at the AI Lab during my tenure there. (I’d concluded that that was the correct way to program robots.) I changed my career plan yet again, to one where I would help lead the robot revolution from iRobot.
A colleague and I proposed Roomba and then an inspired team built it. Roomba was a big success and seemed to point the way to the greater revolution to come. But again there was a glitch. iRobot was reluctant to pursue the risky path of developing more revolutionary robots after Roomba. That meant the lead-the-robot-revolution-from-iRobot part of my scheme wasn’t going to work. By this point my once simple, linear career plan resembled Brownian motion. Maybe my “plan” hadn’t been all that helpful.
As chronicled in Harvest Automation, Part 1, Hatching Tertill, and Dancing with Roomba, a History, what I did after iRobot was to start two robot companies and, most recently, write a book.
Summary
It’s been great! The career I’ve enjoyed these past 40+ years has been better than the one I imagined for myself so long ago. My high school-formulated plan was short sighted—failing to account for the fact that I would learn and grow and that events would necessarily intervene. (Not to mention the fact that my plan flouted the laws of relativity.) In some ways, the plan I intended to inform my choices held me back. In a world filled with change, learning, and growth, rigid adherence to a static plan ignores the important developments that come along the way. An example: there are many occasions where having had a control theory class or business class would have benefited me greatly. But when I was in school, I never considered broadening my education in such ways, it wasn’t in the plan.
In my career, lucky accidents and unplanned events have brought the greatest rewards. I could not have planned to be an early visitor to newly opened China. And I never imagined that robots would trump physics or that I’d help build an iconic, much-loved product.
It’s the learning, growing, and changing that have made life fun. That and robots.
Check out Dare You Dance with the Robot? for more information about a career in robotics.
