The National Highway Traffic Safety Administration (NHTSA) defines Distracted Driving as “any activity that diverts attention from driving, including talking or texting on your phone, eating and drinking, talking to people in your vehicle, fiddling with the stereo, entertainment or navigation system — anything that takes your attention away from the task of safe driving.”
To align the capture of distracted driving events with the above definition, Samsara’s Dual-Facing AI Dash Cam (CM32) uses the inward facing camera, computer vision, and AI to analyze the head position of a driver in real-time. Distracted Driving Detection analyzes driver behavior for changes in the driver head position to help improve safety by proactively identifying distracted driving behavior and sending optional in-cab alerts that encourage drivers to keep their eyes on the road.
Distracted Driving Detection triggers when:
-
Driver speed is above 25mph (by default).
-
A driver lowers their head outside of their normal range, off and on, for a sustained period of time.
You can Configure Distracted Driving Detection time threshold and the driver speed threshold. Increasing the severity threshold may reduce the triggering of some or all these events.
The following examples provide different scenarios where Samsara expects to detect distracted driving and also scenarios where Samsara does not expect to detect distracted driving.
Note
Detection is not guaranteed in any scenario.
|
Scenario |
Example |
Triggers Detection |
|---|---|---|
|
Driver lowers head to look at phone |
 |
 |
|
Driver lowers head to adjust radio |
 |
|
|
Driver’s head bounces up and down |
 |
|
|
Driver looks down at food/hands/lap/phone repeatedly |
 |
|
|
Driver adjusts hat/scratches head/looks left or right while lowering head |
 |
|
|
Driver uses cell phone without lowering head pose |
 |
 |
|
Driver appears drowsy, yawns, or has eyelids closed while head remains straight ahead |
 |
|
|
Driver is eating or drinking, but head remains straight ahead |
 |
|
Distracted Driving Detection can sometimes trigger false positives or inaccurate alerts. To enable accurate detection, encourage your drivers to observe the following best practices:
-
Avoid covering the driver's head by removing glasses, hats, masks, and objects hanging in the view of the camera.
-
Ensure there is adequate lighting in the driver cab. Extremely dark or bright conditions can hinder detection.
-
Ensure the driver keeps their hands away from their face. Any obstruction of the face can confuse the detector.
Comments
0 comments
Article is closed for comments.