Is it because of the VIO failure that caused losing control?

Hi,

I take off in an indoor warehouse using Intel t265 (EKF_AID_MASK = 24 & EKF2_HGT_MODE = vision). After a few seconds, the quadrotor lost control and continued to rise. I attempted to lower the throttle (at about 12:56) but was unsuccessful. The quadrotor then collided with the wall (at about 13:01) and crashed to the floor. Here is the flight log.

https://logs.px4.io/plot_app?log=1f6e310c-feb8-4455-8cc2-88412f531c71

My friend suggested that the reason could be due to the IMU being calibrated during the flight, as the CAL_GYRO_ROT was set to its default value of -1 (figure 1). However, I believe that the IMU anomaly was actually caused by the crash. The loss of feature points in the visual odometry was the primary reason for the collision.

What is the reason for the accident? How to avoid it? :face_exhaling: