This repository contains code for the Colorado School of Mines EENG350 Seed Lab Class.
The purpose of this class is to design and build a robot that can track and follow a trail of blue painter's tape on the ground. The project is split into the following subsystems with different tasks and responsibilities:
System Integration: Send data between a RaspberryPi and Arduino using I2C protocol in order to interface between computer vision and localization.
Computer Vision: Use openCV to detect and isolate blue painters tape on the ground, and calculate how much the robot must turn to be centered with the tape.
Localization: Control the motors of the robot using an Arduino, telling the robot to turn a certain amount or go foward a distance, using control systems to keep the robot on track for following commands.
Simulation and Control: Create control system models in Simulink to upload the data read by the motor encoders, and tune the system to ensure the robot performs the desired task.