From Challenges to Victory

My journey in robotics began in 2015, when I joined the robotics club at my university with no useful skills and a great deal of enthusiasm. The ratio improved slowly.

The motor driver

The first real thing I was responsible for was a MOSFET motor driver. It worked on the bench. It worked during testing. It failed roughly four minutes into every match, which is a very specific kind of failure to debug, because it only happens when it matters and never when you are watching.

The eventual answer was unglamorous: gate drive that was adequate at room temperature and inadequate once the board had been running under load, letting the MOSFETs spend just long enough in their linear region to cook themselves. Nothing in that sentence was visible on an oscilloscope trace taken in the first thirty seconds.

I rebuilt that board more times than I want to admit. Each version failed slightly later than the last, which at the time felt like failure and in retrospect was the entire education.

Robocon 2016

We went to Robocon in 2016 and beat IIT Delhi.

I want to be careful about how much weight to put on that. It was one match in one competition, and the margin was not enormous. But for a team that had spent the previous year watching its own hardware fail, the fact that nothing broke was the achievement. The result was almost incidental.

What the club was actually for

Competition robotics is a strange way to learn engineering. The deadlines are absolute, the budget is whatever the team can scrape together, and nobody grades you on elegance. What you get in exchange is an unusually tight loop between a decision and its consequences. You choose a connector on Tuesday and by Saturday you know whether it was the right connector.

Very little of my professional work has had a loop that tight. Almost all of it has been easier because I once had one.