We investigate travel changes in partially automated electric vehicles. Partial automation can control vehicle speed and steering using sensors that monitor the external environment. We use survey results from survey responses including 940 users of partial automation, including 628 who have Tesla Autopilot and 312 with systems from other automakers. Autopilot users report using automation more than users of other partial automation systems. Autopilot has the largest impact on travel, notably 36% of Autopilot users reporting more long-distance travel. Respondents who are younger, have a lower household income, use automation in a greater variety of traffic, roads, and weather conditions, and those who have pro-technology attitudes and outdoor lifestyles are more likely to report doing more long-distance travel. We use propensity score matching to investigate whether automation causes any increase in respondents’ annual vehicle miles travelled. For simplicity, we focus only on the impact of Tesla Autopilot and find that automation causes an increase of 4888 more miles driven per year.

In this study, we primarily use a cohort survey of PEV owners in California administered by the authors in November 2019. Respondents had been previously surveyed by the PH&EV Research Center between 2015 and 2018 as part of the four phases of the ‘eVMT survey’ when they originally bought the PEV tracked in the repeat survey in 2019 (households may have other PEVs in their fleet). Respondents for the four phases of the eVMT survey were sampled from the pool of PEV buyers who had applied for the state rebate from the California Vehicle Rebate Program (CVRP). More than 25,000 PEV owners were surveyed between 2015 and 2018. 15,000 of these respondents gave consent to be re-contacted and were invited for the repeat survey in 2019. 4,925 PEV owners responded to the repeat survey. The sample is a convenience sample. Most survey respondents have several vehicles in their household. For the purpose of this study we consider the vehicle most frequently used by the survey taker. We ask survey respondents how they use this vehicle, whether it has automation, and how often they use the automated features of the vehicle.

Please note the data only contains three sociodemographic variables for households. The data was reduced so that is includes no more than three indirect identifiers of households to align with guidance on preparing human subject data.