Competition Build — Autonomous Systems

Autonomous Vision-Guided Robotics Platform

LANGUAGEC / kiprALGOPID ControlVISIONOpenCVANALYSISMATLAB
Final Competition Chassis
Overview

Wrote the C-based control firmware for an autonomous mobile robot on a Wallaby controller (kipr/wombat.h) with touch sensors, IR reflectance, CdS photoresistors, an E.T. distance sensor, and a USB camera.

Architecture

Labs built up through reactive, Braitenberg, and closed-loop control paradigms. Lab 3 fused touch-sensor obstacle avoidance with IR-based line following into a single behavior-priority loop. Lab 4 implemented a Braitenberg light-follower: raw CdS readings (100–4000 range) normalized to 0–100% motor power, with a meta-sensing counter that detected repetitive bump cycles within a fixed time window and injected a random escape maneuver to break the loop. Lab 5 added a PID wall-following controller using the E.T./IR sensor, with data logged to arrays and plotted in MATLAB to compare P-gain values.

Competition Notes

Final competition: the robot autonomously collected color-coded cubes on a 4×6 ft arena, reading QR codes at the arena corners to determine the target cube color, and used a floor luminosity gradient to identify the home zone. 2-minute autonomous collection window. 1st place finish.

View Code on GitHub
Back to All Projects↑ Top