Tesla Autopilot Survey (from MIT)

[tonglaji]

took your survey today. The questions about % miles and % hours driven on autopilot were the most difficult ones for me. I see autopilot maturity level significantly different on highway from local road and thus usage pattern differs too. So the question would be easier to answer if you broke them up in 3 or 5 questions: 1) % of highway driving and % of local driving (100% total), 2) % miles or % hours driven on autopilot on highway, and 3) % miles or % hours driven on autopilot on local road.

 

[wdolson]

What do you think would be a good question to ask about the instrument cluster?

Good question. I've kicked this around today, but don't have any solid suggestions. Maybe something along the lines:

Do you think the warnings for autopilot/TACC are useful?

If "No", which warnings are lacking? (check all that apply)
a) Visual cues are inadequate
b) Visual cues are hard to see
c) Audio cues are inadequate
d) Audio cues are hard to hear
e) Other Specify:

Would it be helpful if there were some options for the users to have some options in the car settings?

This is just my best suggestion at the moment.

 

[Lex_MIT]

I've added two new questions on the topic of HCI:

1. How useful do you find the visual display of what Autopilot sees (lane markings, other cars) on the instrument cluster (Model S/X) or center display (Model 3)?
   
2. How useful do you find the sound (audio signal) you hear when Autopilot is turned on or off?

Thanks everyone for filling out the survey.

 

[CK_Stuggi]

Just filled it out. Thank you for the enlightening lectures, my pleasure to pay back a bit this way.

 

[Tezlanian]

Some more suggestions I thought of since completing the survey

• In the case of not just autosteer but warning systems would it be useful to have different sounds for different events? Nearly all have the same sounds.
  
• Would you find it useful to further utilize the speaker system to denote the direction of a threat? (examples: forward collision warning uses front speakers, rear end collision warning uses back speakers, et cetera)
• Would a higher contrast color scheme make the environmental screen more effective? (blind spots as light gray on white is difficult to see)
  
• If you circumvent the autosteer nag, how often do you do so? (if so, duration of trip prompting this)
• Would a capacitance sensing steering wheel, which merely needs to be touched like your smartphone screen, make you less likely to circumvent the autopilot nag?
• How comfortable would you be with a cabin camera performing eye tracking while autosteer is enabled?

 

[Lex_MIT]

Some more suggestions I thought of since completing the survey

• In the case of not just autosteer but warning systems would it be useful to have different sounds for different events? Nearly all have the same sounds.
  
• Would you find it useful to further utilize the speaker system to denote the direction of a threat? (examples: forward collision warning uses front speakers, rear end collision warning uses back speakers, et cetera)
• Would a higher contrast color scheme make the environmental screen more effective? (blind spots as light gray on white is difficult to see)
  
• If you circumvent the autosteer nag, how often do you do so? (if so, duration of trip prompting this)
• Would a capacitance sensing steering wheel, which merely needs to be touched like your smartphone screen, make you less likely to circumvent the autopilot nag?
• How comfortable would you be with a cabin camera performing eye tracking while autosteer is enabled?

Great questions! I especially like the last one. If not in this survey, I'll add in future versions.

 

[bmah]

Great questions! I especially like the last one. If not in this survey, I'll add in future versions.

It sounds like you are modifying the survey in response to feedback. In a methodological sense, are you concerned about these changes having some unintended side effect on the survey results?

Bruce.