Seems to me we started off trying the wrong foot, and just make it exponentially worse by doubling down on a bad solution...repeatedly.
Function: Safety feature to modulate speed and lane maintenance.
Problem: Use of the function while being inattentive results in unsafe consequences.
Root Cause: Inattentiveness/Lack of driver attention
Implemented Solution: Require [minimal] driver input
Result: Inattentiveness/Lack of driver attention resulting in unsafe consequences.
Root Cause: Inattentiveness/Lack of driver attention
Implemented Solution: Require [minimal] driver input more frequently
Result: Annoyance to customers resulting in less frequent use of the safety feature.
Root Cause: Driver inputs required too frequently
Proposed Solution: Clearly state that driver input is minimal.
Expected Result: [Predicted] Annoyance to customers resulting in less frequent use of the safety feature.
Root Cause: Driver inputs required too frequently
Dear Tesla/Elon, how about we address the root cause? You were so great as a company at getting past the problems and just getting things to work in the past. It's as if you've forgotten how to solve problems in the present.
Working backwards.
Root Cause: Driver inputs required too frequently
Proposed Solution: Require less driver inputs!
Root Cause: Inattentiveness/Lack of driver attention
Proposed Solution: Implement an actual solution that measures driver attentiveness.
Here's just one man's suggestion. I'm sure there are thousand others.
Measure reaction time of a driver to an alert/notification.
Since Tesla's already gone with steering input, I'll stick with Tesla's baseline but modify it.
- Step 1. Alert for driver input every (old limit of 2min)
- Step 2. Measure response of the driver (How many seconds until input recieved)
- Step 3. Does it exceed average reaction time + X buffer? If so, retain alert frequency or increase frequency of alerts. If not, reduce the frequency of alerts. Heck if you want, put a cap on maximum time for alerting based on no detected driver input (5min? 10min?)
Repeat Steps 1-3 until the drive is over. Then reset to your baseline on the next drive and repeat steps 1-3.
This allows for Tesla's butt to be covered in the event of an accident. Because you tested the drivers attentiveness. The system adjusts based on inputs that indicate driver attention and there was a reasonable amount of data to support whether someone was paying attention.
If the driver has shown to be an inattentive driver, you're taking safety actions to require more frequent inputs. But if there is no evidence of inattentiveness and the driver is responding and alert to the notifications; reward them and give them some rope.
PS. I also think there should be no limit at all and people should stop being blaming Tesla. If I stopped paying attention and screwed up, that's on me. But if we're going to get all nanny based on this; I think the above would be much easier to swallow. Otherwise, next up: Maximum speed on all cars will be the federal speed limit.
