Problem about fixed wing headwind takeoff

Flight Testing & Log Analysis
1

Hello everyone,

We are experiencing an issue with the ground behavior of our fixed-wing aircraft in headwind conditions. When we align the aircraft on the runway facing into the wind, the control surfaces deflect as follows: down elevator (nose-down), down left aileron, and up right aileron. This happens both in STABILIZED mode while stationary on the runway, and during takeoff rotation in MISSION mode. During this test, the wind was around 8–10 m/s, which can be seen in this log along with other details.

During this test, the aircraft was unable to take off even though it had sufficient airspeed because of the pitch-down command. It ultimately veered off the runway and suffered a prop strike (The takeoff attempt via MISSION mode is at the end of the log). We also observed that in these conditions, the throttle jumps instantly to 100%, ignoring the ramp-up time we have set (RWTO_RAMP_TIME).

For comparison, we did a quick test on the runway (before the accident) by placing the aircraft in the opposite direction (tailwind). In that case, the abnormal control surface deflections and the instant throttle jump did not occur. (This test is not included in the log above.)

We were also able to reproduce the behavior on the test bench by blowing air into the pitot tube - the control surfaces responded similarly.

This aircraft has flown successfully (aside from a few unrelated, minor issues) with a fully autonomous mission from takeoff to landing using the same parameters before. The log from one such successful flight is here.

In that flight, I recall the wind was around 6 - 7 m/s, but the log does not show significant wind during takeoff (but it can be clearly seen during flight if you compare the airspeed and groundspeed), in contrast to the recent problematic flight (maybe the wind was not that strong at the runway - I don’t remember clearly.)

Any help would be greatly appreciated, as I haven’t been able to come up with a good explanation or solution. Is this behavior normal, or is there something I should configure differently? I would expect normal takeoff behavior even with headwinds of this magnitude, as the aircraft has sufficient power and cruise airspeed for these conditions.

2

Based on the log data, extended periods of remaining unlocked on the ground may contribute to this issue. Minor geometric deviations in the flight controller could accumulate in the integrator over time, potentially leading to unexpected control surface deflections.

To address this, you might consider:

  1. Performing a precise flight controller calibration and taking off promptly after arming to minimize ground time.
  2. Adjusting the controller’s maximum integral limit downward, though please note that setting this value too low could impact inflight stability.