Browse Publications Technical Papers 2018-01-0497

Driver Behavior While Operating Partially Automated Systems: Tesla Autopilot Case Study 2018-01-0497

Level 2 (L2) partially automated vehicle systems require the driver to continuously monitor the driving environment and be prepared to take control immediately if necessary. One of the main challenges facing developers of these systems is how to ensure that drivers understand their role and stay alert as the systems require. With little real world data, it has been difficult to understand user attitudes and behaviors toward the implementation and use of partially automated vehicles. At the time of this study, Tesla was one of the few OEMs with a partially automated vehicle feature available on the market; Autopilot. In order to understand how customers interact with a partially automated vehicle, a study was conducted to observe people driving their own Tesla vehicles while autopilot was engaged.
Sixteen Tesla owners (14 males and 2 females) between ages 25 to 60 had their vehicles instrumented with video/audio data collection systems for three consecutive days. These owners were dedicated autopilot users who used the feature daily and primarily on highways. Results from the study show that (i) participants’ eye off-road glance behavior while operating a partially automated vehicle was somewhat similar to eye off-road glance behavior in radio tuning task while driving manually, though substantially more variable, (ii) eyes on/off road glance behavior had no relation to whether drivers kept their hands on the steering wheel or not, (iii) drivers exhibit a bi-modal behavior when operating a partially automated vehicle: those who mostly kept their hands on steering wheel while autopilot was active (i.e. active drivers) verses those who mostly kept their hands off the steering wheel while autopilot was active (i.e. supervisors).


Subscribers can view annotate, and download all of SAE's content. Learn More »


Members save up to 18% off list price.
Login to see discount.
Special Offer: Download multiple Technical Papers each year? TechSelect is a cost-effective subscription option to select and download 12-100 full-text Technical Papers per year. Find more information here.
We also recommend:

Identifying Traffic Scenarios to Evaluate Driver Readiness in Automated Driving: An Exploratory Study


View Details


Additional Findings on the Multi-Modal Demands of “Voice-Command” Interfaces


View Details


Safety-Relevant Guidance for On-Road Testing of SAE Level 3, 4, and 5 Prototype Automated Driving System (ADS)-Operated Vehicles


View Details