Vibration

What it reads#

The rule reads accel_vibration_metric from vehicle_imu_status in the flight log, in m/s². PX4 computes the metric from the flight controller's accelerometer, so it describes vibration where the flight controller is mounted. In the INC-0143 flight log a new sample arrives every 200 ms.

topic      vehicle_imu_status (multi_id 0)
timestamp  237,401,000 µs   boot + 237.401 s

accel_device_id             2621474 (ICM-45686)
accel_vibration_metric      24.31 m/s²
gyro_vibration_metric       0.388 rad/s
accel_clipping[0..2]        0 2 9
temperature_accel           41.2 °C

The same record carries accel_clipping, three counters of accelerometer clipping. The threshold does not use them, and they are worth reading beside the metric.

When it fires#

On the T4, a finding is produced when the metric stays above 24 m/s² for 1 s. The event sits on the first sample above the line, and the sample 1 s later confirms it. The Q4 value in the same threshold text is 12 m/s².

The finding's title states the stretch: Vibration above 24 m/s² for 1 s in the obstacle belt.

Wording and evidence level#

Every finding reads Observed: vibration above threshold. The flight log shows the metric directly, so the level is observed.

What the finding does not mean#

• Nothing necessarily broke or came loose. In INC-0143 the radio dropped off USB 1.69 s after the onset. The report keeps the idea that vibration unseated the radio's connector as a hypothesis until the connector is inspected.
• It describes one mounting point. The metric comes from the flight controller's accelerometer. A part mounted elsewhere, such as the radio on the companion computer, can see a different load.
• Flight Review's colors are not your threshold. For reference, PX4 Flight Review shades this metric green below 4.905 m/s², orange to 9.81 m/s² and red above 9.81 m/s², for air vehicles. The T4's 24 m/s² is a program choice for a tracked vehicle, not a PX4 number.
• No finding is not low vibration. If the flight log did not log vehicle_imu_status, the rule has nothing to read. The example dataset's Q4 runs list it as Not logged in the run page's Flight log topics panel.

Threshold and overrides#

The per-platform values live inside the threshold text, Above 24 m/s² for 1 s (T4), 12 m/s² (Q4), not in the Override field. The scope is All vehicles. Set each platform's value from its own known-good runs.

Change the threshold under Rules with Edit. The rule saves as vibration_high 1.1, and findings already produced keep version 1.0 and its threshold. See Thresholds, scopes and overrides.

Example finding#

In INC-0143, UGV-02 crossed the obstacle belt in Breach lane 2 at 1.25 m/s. The metric read 24.31 m/s² at 15:41:00.912 and stayed above 24 m/s² through the sample at 15:41:02.912. It peaked at 38.6 m/s² at 15:41:02.312.

In the 1.4 s before the finding, the samples read 19.8 to 23.1 m/s². All of them are red on Flight Review's air-vehicle scale, and none produced a finding on the T4.

The kernel logged the radio's USB disconnect 1.69 s after the onset and 0.29 s after the peak. The report files the disconnect as correlated with the vibration, and the cause as a hypothesis. The Findings row labels this finding Vibration above threshold, the incident's first.

On the event page, Vibration around this event plots 8 s before to 4 s after the first sample. A dashed line marks the rule threshold, and filled points are at or above it.

Related#

• device_disconnect records the radio leaving the USB bus in INC-0143.
• failsafe_entered records the Hold that followed.
• PX4 ULog lists the topics Foxborne reads from a flight log.
• Check a run shows which topics a flight log contains.