Page tree
Skip to end of metadata
Go to start of metadata

Meeting Details

  • Location: Lobby Uni Tower
  • Supervisor: cmot
  • Protocol: lpe
  • Attendees:
    • cmot
    • fma
    • lan
    • lpe
  • Start: 10:00
  • End: 12:30

Agenda

Testing reworked Flight Controller

We started the day by testing the reworked flight controller. We needed to redo it as we noticed erroneous behaviour a lot - e.g. jerking movement while steering the copter by hand. These tests started very promising. A lot of the changed behaviour worked as we wanted and there were no crashes and no unpredicted behaviour.

We also adjusted the P, I and D values of the yaw axis which lead to a fairly stable axis and much less turning while starting the flight (and during flight as well).

Crashing

After testing the flight for a while we encountered a devastating crash. While testing the "steering reset" button, a button to reset the values that control the desired angle, the connection to the Arduino was suddenly lost and the Copter drifted very fast into the wall of the lobby. This resulted in critical damage to the quadcopter:

  • One leg broke off
  • Two protectors for the rotors broke
  • One motor smells weirdly and was replaced
  • One more motor was precautionally replaced

Afterwards the situation was assessed. We tried to find out why exactly the copter crashed. There are multiple theories:

  • The Arduino Mega itself crashed (Hardware Malfunction)
    • Critical failure, highly unlikely
    • If true, we will need to buy a new Arduino
  • Code on the Arduino is malfunctioning/Arduino gets stuck in infinite loop
    • More likely, will need assessment/review of our code
    • If true, we will need to find the mistake and correct it
    • Other solution for both failures: Watchdog (Code on the Arduino that causes the Arduino to reset after a while, if a function is not called regularly
      • This will require a new bootloader as it doesn't work with our current setup
  • Loss of Bluetooth connection
    • Most likely scenario
    • Either caused by low voltage (hopefully! As this is easy to handle)
    • Or by other causes, this would require a keepalive signal from the computer connected to the arduino via bluetooth
  • No labels