MAR24: The Minimal Autonomous Rover Project
Parent: My Projects
The Minimal Autonomous Rover Project
Supervised by: Dr Klaus-Peter Zauner
Disclaimer: This project is under active development. Information on this page may not be up to date or accurate.
This project is to develop a computer vision board that detects the gates formed by pairs of ArUco markers, and output the coordinate of the gate to the rover's microcontroller. The system is designed to use minimal resources (e.g. run on a Raspberry Pi Zero board) and work very reliable under many circumstances (e.g. strong sunlight, high contrast, rain). To make the board boot fast, it may be necessary to build a lightweight Linux using Yocto. The vision module may be used for ELEC2209 Engineering Design module.
ArUco Detector
The Python script to detect the ArUco markers.
- Git via HTTPS: https://git.soton.ac.uk/qs2g22/mar24-aruco-py.git
- Git via SSH: git@git.soton.ac.uk:qs2g22/mar24-aruco-py.git
Serial Communication
The vision board talks to the microcontroller via UART (9600-8-N-1: 9600 baud, 8 data bit, no parity, 1 stop bit). Example use of UART can be found here.
The detailed specification can be found at: mar24-aruco-py/docs
Fast Booting
Remove unnecessary modules to make Linux boot faster on Raspberry Pi Zero.
- Git via HTTPS: https://git.soton.ac.uk/qs2g22/meta-marfb.git
- Git via SSH: git@git.soton.ac.uk:qs2g22/meta-marfb.git
