Sushibot: A Smart Autonomous Mobile Robot

with Harine Ravichandran, Michael Rittikaidachar, and Alexa Siu. Completed for Stanford’s ME 218B Smart Product Design Applications course in Winter 2017.

Project Overview (ME 218B)

Our final project in Stanford's ME 218B, Smart Product Design Applications, was to design and build an autonomous mobile robot that competes in a game against another robot. In the game, each robot autonomously navigates to target staging areas, checks in, and launches construction materials (foam balls) to construction sites (buckets) across the field. The robot was required to be fully autonomous and untethered, powered by two 7.2V batteries. Event-driven embedded software was written in C on a Tiva Launchpad to control the system, and each robot was required to communicate autonomously with the game field to verify its position and request game information. I was primarily responsible for sensing and electrical design for this project.

To follow a designated path on the playing field, it was required to detect the frequency of alternating current in a wire laid under the field's surface. To do this, I designed a sensing circuit using two pairs of inductors (on in front, one in back) that change their voltage with their orientation in the wire's magnetic field. The inductors' signals were filtered, amplified, and passed through peak detection circuits. The difference between the two inductors' signals was used as an analog reference to determine if the robot was to right or left of the wire.

We also used Hall-effect sensors to detect the frequencies emitted by multiple staging areas throughout the field, and photo-transistors to detect the frequencies of various target beacons which indicated where the robot should launch the foam balls.