Events are any of a number of risky incidents that can occur during a trip, such as acceleration, sharp turns, or distracted driving.
Following are the different event types:
-
general
-
accelerometer and risky driving
Accelerometer events, such as acceleration and violent left turn, are detected by the camera's motion sensor. The motion sensor has different sensitivities to trigger events depending on the vehicle size, from the least sensitive (private) to the most sensitive (trailer).
-
distract driving
Distracted driving and smoking, are detected by MV+AI. Machine vision (MV) uses an algorithm to determine if the driver is focused on the road or distracted. Artificial intelligence (AI) focuses on face and upper body movements to determine the type of distracted driving.
All events contain the following data:
-
timestamp, in ISO 8601 format
-
location
-
option to retrieve video
In addition, all acceleromter and risky driving events include the following metadata:
-
speed, in meters per second
-
g-force on the X, Y, and Z axes, in meters per second squared
-
timestamp, in Unix format
The metadata is sampled every 100 milliseconds of an event, which is 10 seconds long (5 seconds before the event is triggered until 5 seconds after). This comes to a total of 100 measurements per event.