IEEE RAS Robotathon 2024

Overview

Robotathon 2024 was my first competition robotics team. I programmed an autonomous maze robot in C++ on Arduino, navigating using distance, color, and line sensors with PID-based line following.

We participated without placing first. It is on this page because it is where the rest of my robotics work started.

The problem

A maze robot has to decide where to go using only what its sensors can tell it about its immediate surroundings. It has no map. Distance readings say how far a wall is, color readings identify markers, and line sensors say whether it is still on track, and from those three signals it has to produce a route.

Approach

Line following used PID control: the robot measured how far it had drifted from the line and applied a correction proportional to that error, damped so the correction itself did not cause oscillation.

Distance and color sensors handled the parts of the maze that line following alone could not: knowing when a wall was close enough to require a turn, and recognizing markers that indicated something about position within the maze.

The whole thing ran on Arduino in C++, which meant working inside real constraints on memory and timing rather than abstract ones.

Working with the team

The software ran on hardware that other people were designing, including 3D-printed parts from the mechanical team. This was the first time I experienced a mechanical change altering what my code could sense. Moving a sensor changes its readings, and no amount of correct logic compensates for a sensor that is now looking at the wrong thing.

Result

We competed and did not place first. The source is public.

What I learned

This is where I learned that PID is not a formula to copy but a behavior to tune, and that robotics is a team discipline before it is a software one. Every competition I have worked on since (Makeathon, Robotathon 2025, and now IGVC) built directly on what went wrong and right here.