search

Jaeyoung-Lim / InRoL-Quadrotor-Platform-for-Research

Project for developing a quadrotor platform for research platform used in Seoul National University interactive networks Laboratory
1 stars 2 forks source link

APM-Copter won't stay in posion in Loiter mode #4

Closed Jaeyoung-Lim closed 9 years ago

Jaeyoung-Lim commented 9 years ago

APM won't stay in position is loiter mode

default

Diverges after circling the location

Jaeyoung-Lim commented 9 years ago

Log shows EKF-1 / EKF-2 Errors

Cases of the problem could be

  1. The oscilation of external compass sensors
  2. Pre arming process of gps not performed
Jaeyoung-Lim commented 9 years ago

For Case 1. : did a test flight indoor and when unplugging the external compass, the quadrotor did not show EKF errors.

Case 1 could have been the problem

Jaeyoung-Lim commented 9 years ago

Found out that Err: EKF-Check-0 and EKF-Check-2 errors are from the link below http://copter.ardupilot.com/wiki/configuration/ekf-inav-failsafe/

Interestingly link says as below

The EKF check runs only on the Pixhawk and only when the EKF is being used as the primary source for attitude and position estimates (i.e. AHRS_EKF_USE = 1).

Firmware version maybe wrong?

Jaeyoung-Lim commented 9 years ago

Arducopter 3.3 does not support APM 2.x boards

http://copter.ardupilot.com/wiki/common-autopilots/common-apm25-and-26-overview/

Jaeyoung-Lim commented 9 years ago

Currently trying to upload firmware from a previous developer program that was cloned.

(Version : Arducopter v3.3)

Jaeyoung-Lim commented 9 years ago

Changed board to HKpilot32

still problem seems to be the terminology of Toilet Bowling

http://copter.ardupilot.com/wiki/flying-arducopter/flight-modes/loiter-mode/

"The vehicle circles (aka “toiletbowls”). This is normally caused by a compass problem the most likely being magnetic interference from the power cables under the flight controller. Running compassmot or purchasing a GPS+compass module normal resolves this. Other possibilities include bad compass offsets set during the live calibration process or incorrect compass orientation."

The copter is currently experiencing bad compass offsets

Jaeyoung-Lim commented 9 years ago

loiter_150728_working

All previous experiments were done around building 301 in Seoul National University. Calibration was conducted in Building 35 and loiter performed working.

The environment in which the calibration was done seems to be the problem. Building 301 is a metal structure high rise building. Calibrating the compass sensors outside in an open field may inhance performance.