Why I love robotics
Why do you want to be a roboticist? Is it for money? For fame? Or for both?
For me, it was neither. In my early twenties, I decided to devote my life to robotics after reading two books by Norbert Wiener: Cybernetics (Wiener, 1948) and The Human Use of Human Beings (Wiener, 1950). Cybernetics is, in many ways, where the history of robotics began (Slotine, 2006).
One of Wiener’s piercing insights changed how I saw the world: as information-processing entities, humans and machines are fundamentally the same. The deeper we understand ourselves, the better we will be able to build machines that resemble us, and the line between us and our machines will begin to fade.
That idea became the heart of my Statement of Purpose. I was fortunate to be accepted to MIT, the very place where Wiener once taught in the Department of Mathematics.
There is plenty to learn from biology
Wiener’s insight also works in the other direction. If humans and machines are the same kind of thing, then biology is a textbook for robotics, and one we have barely begun to read (Slotine, 2006).
Consider how slow we are. The fastest signals in our nerves travel no faster than about 120 meters per second, roughly a million times slower than their robotic counterparts. Our muscles respond at well under 10 Hz, while electric actuators reach hundreds of hertz. And a feedback loop that runs through the cortex takes more than 100 milliseconds to close (Kandel et al., 2014; Nah et al., 2025). By any engineering measure, we should be clumsy.
And yet we are not. A child can catch a ball, open a door, or crack an egg without breaking the yolk, and these remain hard problems for today’s fastest robots. Whatever the nervous system is doing, it is not winning on speed. It must be winning on organization: how it breaks behavior into simpler pieces, how it uses the mechanics of the body instead of fighting them, and how it stays stable while in contact with the world.
One compelling answer is modularity. The nervous system may encode behavior with a small set of dynamic primitives (submovements, oscillations, and mechanical impedance), a simplification that the slow response of the neuromuscular system may make essential (Hogan & Sternad, 2012). My doctoral work carried this idea over to robots, building a modular robot controller from a finite set of motor primitives (Nah et al., 2025). To me, what biology can still teach robotics is one of the most exciting open questions in the field, and one of the reasons I love it.