After taking a very math-heavy controls course, I thought I'd apply the math I'd learned to make a little robot that can draw arbitrary patterns with pen(s) mounted on its behind. The problem is difficult because the pen is not at the center of rotation for the robot. The project fell victim to a lack of free time and dedicated hardware, but before the end of the semester, I had a motion planning system that worked reasonably well, running in Java and sending data to the differential drive robots in the Moble Robot Programming Lab at CMU. Shown is the robot drawing a cursive S from a few waypoints. The wiggly lines in the image were due to overshoot from the routine's use of just simple proportional control; once the proof of concept was complete, I never found the time to implement a PID loop.