Tesla is reportedly stepping up driver monitoring in its vehicles, with the system now tracking yawns, blinks, and more in order to detect how drowsy a driver is.
Ever since Tesla introduced the cabin-facing camera in 2021, it has used it to track eye movements to ensure the driver is watching the road and not using a handheld device.
The automaker has made several improvements since then, including stepping up the penalties for distracted driving by disabling the use of Autopilot for those users who may engage in other activities behind the wheel.
Now, however, Tesla is reportedly taking it up a notch. According to Tesla hacker Green (@greentheonly), the automaker is now tracking things like yawns and blinks.
“Looks like Tesla is planning a big boost to (camera based) driver monitoring. They are now tracking additional things like how many yawns the driver had recently, how many blinks and how long they were, leaning. All this is to calculate how drowsy the driver is,” the hacker wrote on Twitter.
Green said the code is being added, but “it is not propagated to anything UI as far as I can tell yet.” Interestingly, he noted that the increased tracking is not happening only when Autopilot or FSD are active, but also when Tesla drivers are driving without ADAS assistance.
“It also looks like they are planning to apply this even when not on AP (which is a very right move!) by seeing how well-centered the driving is, how many lane keep assist warnings and corrections happened lately,” he posted in a reply to his initial tweet.
However, Green said it’s unclear what Tesla plans to do if the driver is deemed unfit to drive when not on Autopilot. “Hopefully they’ll at least dial all the warnings and active safety stuff to 11 (and enabled when disabled?),” he tweeted.
Why is Tesla increasing driver monitoring, though? Well, the automaker used to be confident that it could deliver Level 4 and Level 5 autonomy quickly, but that doesn’t seem to be the case anymore.
Tesla is paying more attention to driver monitoring now probably because it is more useful for Level 2 and Level 3 autonomy, which are more realistic targets in the short term.