Define physical metrics that are to be measured

[chrisngan24]

• Define which metrics we want to measure. Focus on:
  • Importance in contribution to safe driving
  • Techniques we want to use to measure it
  • Sensor we want to use to measure it
  • current research

https://docs.google.com/document/d/1EiYMJyWXAOXZig904wvGrX0pbSJny1usYfHBZQyh7G4/edit?usp=drive_web

cc: @angelagu

[joshuroy]

what's the green highlighted for? I added speeding in bad weather and was thinking we should include distracted driving - majorrr cause of accidents

[angelagu]

The yellow is the top 2 using metrics we established, the green is the next 4 that are tied. We have "Using phone/radio while driving".

[joshuroy]

my 2 cents is that if you can solve speeding, bad breaking is less of a problem and less lethal if unfollowed. Also, reducing the heuristic in bad weather is pretty impt to (based on top causes of accidents)

[joshuroy]

@angelagu @chrisngan24 how is oversteering defined, and how is it distinguished from a normal lane change? Do we wait to see if it's an exaggerated "oversteer", hence lane change - anything less is bad oversteering?

[angelagu]

I think for oversteering we're focusing on turning and not lane change.

[joshuroy]

Oops I was thinking of veering all along. Lemme change that