...
- Location: Lobby Uni Tower
- Supervisor: ssm
- Protocol: fma
- Attendees:
- apo
- fma
- lan
- lpe
- ssm
- End: 12:00
Agenda
| Table of Contents |
|---|
...
Errors
The first flight in a larger free area ends with an emergency stop via the computer. The quadcopter pushes to it's side hardly. After some the problem is identified:
The longer some difference between the actual and the intended angle persists, the harder the I-value of the PID-Controller tries to correct it. With starting the Quadcopter's Arduino on a not 100% flat surface and waiting some time before flight, even a small difference of 0.4 degrees can cause a strong counteraction when lifting off.
As a solution, with every start of the motors all controller values are being reset now. This way we are able to start and land and start again without resetting the Arduino every time.
Adjusting the controller values
The next phase is finding the optimal values for all P, I and D controllers. There are some rules about how to proceed the best way. For further informations there will be some information in the future.
Result of this phase are values for our controller with which we can reach a quite steady flight. For steering against any drifting the keyboard on the computer is used.
[Video coming]
What comes next?
Next tasks are:
...
recent test flights led to unpredictable errors while flight. Without a known reason, suddenly within controlled flight the quadcopter starts to oscillate hardly, that we need to stop the flight.
What we can see so far is that this behavior might be triggered somehow by using the keys on the keyboard to control the flight. It has nothing to do with the amount of serial data, sent to the computer.
See Testflüge.xlsx for all serial data transmitted.