Abstract: There are provided techniques including a system, method, and computer-readable medium for clock drift monitoring and correction for autonomous vehicles. The method may include the steps of: determining that a first clock or a second clock of a vehicle is unable to synchronize with a common master clock of the vehicle according to a clock synchronization protocol, wherein the first clock and the second clock are associated with a first device and a second device of the vehicle respectively; and in response: determining a first value indicating a relative clock drift between the first clock and the second clock; determining a first time associated with data provided by the second device to the first device, the first time being determined according to the second clock; and determining a corrected first time according to the first clock based at least in part on the first value and the first time.

Abstract: There are provided techniques including a system, method, and computer-readable medium for clock drift monitoring and correction for autonomous vehicles. The method may include the steps of: determining that at least one of a first component associated with a first clock or a second component associated with a second clock of a vehicle is unable to synchronize with a common master clock of the vehicle according to a clock synchronization protocol; operating the first component or the second component without synchronization of the first clock or the second clock with the master clock; determining a first value indicating a relative clock drift between the first clock and the second clock; determining that the first value exceeds a first threshold value; and operating the vehicle in a degraded state or performing a mitigating action based at least in part on determining that the first value exceeds a first threshold value.

Abstract: Techniques described are related to determining when a discrepancy between data of multiple sensors (e.g., IMUs) might be attributable to a sensor error, as opposed to operating conditions, such as sensor bias or noise. For example, the sensor data is passed through one or more filters (e.g., bandpass filter) that model the bias or noise, and the filtered data may then be compared for consistency. In some examples, consistency may be based on residuals or some other metric describing discrepancy among the filtered sensor data.