AUV2k19

---

Software stack for the vehicle that participated in SAVe 2019

Note: Major portion of this code was hastily written a month before the competition. Hence, brace yourself for bad coding practices!

Repository Structure

• motion: Includes all codes related to motion
  • movement.py: the main motion module that includes all basic movements. It is intended to be used as a library for all major tasks.
  • line_follower.py: the code we used in the competition. The name is misleading. Its actually bouy detection and following, which was the fist task of the competition.
  • camera.py: camera utility functions and vision-based bouy detection algorithms. This module should ideally be inside the vision folder.
  • thrust.py, thrust_new.py and thrust_bouy.py are just test scripts and can be safely ignored.
• sensors: sensor related modules
  • IMU.py: A utility script to read IMU data. The code here is copied and not written by us.
    • This might help if you want to interface IMU with Pi. Accelerometer, Gyroscope, and Magnetometer Analysis with Raspberry Pi Part I: Basic Readings — Maker Portal
  • NodeRead.py: Need of NodeRead.py and how it works?
    • Our Vehicle had two controllers: Pi and NodeMCU
    • We used NodeMCU because pressure sensor throws analog data which can't be read by Pi.
    • Hence NodeMCU was used for reading analog data from pressure sensor.
    • NodeMCU then sent this data to Pi over a USB acting as a TTY(terminal)
    • NodeRead.py is used to read this data from terminal.
    • Note: You will node nead this if you use and ADC(Anolog to Digital Convertor) to convert pressure sensor data to digital form.
    • A special thanks to Ishan Singh for writing this code.
• vision: camera related code
  • VideoRecord.py: A script to record and save video.
• keyboard.py:
  • A script to manuall control the AUV using keyboard.
  • Generally you have to press enter to give input to a program. To avoid that, I used the curses module.
• main.py:
  • This is the driver script for line_follower.py
  • This is the script that we ran in the SAVe competition.
  • So yes, running python main.py helped us reach finals and secure 4th rank in SAVe 2019.
• Test Scripts: Ideally, they should have been in a separate test folder. But currently they are present in the root folder itself.
  • test_camera.py: check if camera is working
  • test_depth.py: test the code that maintains depth and pitch simultaneously.
  • test_pitch.py: check the code that maintains pitch

Hydrophones

• Read hydrophones_study.md
• Most of the work in this area is done by Jawad Akhtar

Tech Stack

• OpenCV
• Numpy
• Python
• Raspberry Pi 3B+
• Node MCU
• Pressure Sensor
• IMU MPU6050