< Back to Session View

Passenger Acceptance of Remote Intervention Following System Failure with ADS Dedicated Vehicles in Shared Mobility Applications

Date and Time: Tuesday, July 11, 2023: 5:30 PM - 7:00 PM

Matthew Marchese 

Human Factors Engineer, National Highway Traffic Safety Administration

Presentation Description

The present work involves Automated Driving System – Dedicated Vehicles (ADS-DVs) remote operation research. When an ADS-DV experiences system failure, the driving automation should return the vehicle to a minimal risk condition. Once the driving automation completes this task, there may be instances when human intervention is required. With shared mobility applications, vehicle occupants have short-term access to vehicles and may not possess the necessary level of ADS-DV familiarity to successfully complete this task. This issue can lead to scenarios that require remote intervention. For Level 4 driving automation, SAE International refers to remote users who execute these interventions as Driverless Operation Dispatchers (DOD; see SAE J3016_202104). This completed project explores human factors concepts to support remote DOD intervention and occupant informational needs in shared mobility applications. When considering the use of DODs, occupants may have concerns about the level of DOD intervention, messaging provided by a DOD, and even the number of vehicles a DOD is assigned to monitor. This DOD presence also has the potential to increase overall trust for an ADS-DV-equipped rideshare, the availability of human assistance may assuage some mistrust around driving automation. However, occupants may also be concerned with privacy considerations regarding the camera views necessary for successful remote operation. Furthermore, the types of communication and coordination (e.g., direct voice versus updated in-vehicle displays) may differ based on the communication needs’ urgency. In all cases, a best practice for safety and design is provision of redundant means of coordination. For example, vehicle-specific methods for coordination could be supported by direct contact to participants via smartphone application, text message, or telephone communication. Published research focused on passenger/occupant needs when using ADS-DV applications is limited. To address this gap, focus groups were conducted to gather information about the needs of passengers. Three focus groups were held via teleconference, and each focus group included four participants that were rideshare users (e.g., Uber, Lyft). Videos of ADS-DV scenarios were created to provide focus group participants with an exposure to ADS-DVs and DOD intervention scenarios. Two scenarios were presented to participants: an introductory scenario to the ADS-DV platform and a road closure scenario requiring DOD intervention. After viewing the videos, participants were asked questions related to the video scenarios. Focus group analysis centered around how participants answered each of the primary questions posed to them during the discussion after reviewing a series of videos and/or hearing about situations the ADS-DV might encounter. The questions were asked to gain understanding of participant information and communication needs. Using an iterative approach, an analyst reduced the data through summarization and synthesis while maintaining links to the original data, thus allowing for a comprehensive and transparent analysis. The types of information that focus group participants reported wanting during an unexpected road closure or an emergency stop were primarily related to the situation, next steps, and how the route/time will be impacted. Participants also discussed how an emergency stop situation may involve a medical emergency, specifically, the role of the dispatcher in a 911/medical situation. Communication between DODs and passengers was an important theme in participant responses which encouraged providing communication multi-modally (audibly, visually) via multiple methods (e.g., text, app, intercom). Safety and security concerns were raised by participants when asked about their acceptance of the vehicles. Results from these focus groups reduce the gaps present in the current research body of knowledge and help gain a better understanding of human factors issues that emerge as ADS-DV applications develop. In addition, they also provide some guidance toward prioritization of addressing these gaps.

Speaker Biography

Matthew Marchese is a Human Factors Engineer for NHTSA's Vehicle Safety Research Office of Vehicle Crash Avoidance/ Electronic Controls. He has a B.S. and M.S. in Industrial and Systems Engineering from Virginia Tech. Mr. Marchese's area of expertise is related to ADS and ADAS human machine interaction in both light and heavy vehicles. He has also worked with the Federal Highway Administration's Offices of Safety and Operations where he conducted human factors research related to novel traffic control devices and cooperative driving automation. Mr. Marchese began his career in transportation safety at the Virginia Tech Transportation Institute where he assisted in the collection, reduction, and analysis of naturalistic and experimental driving studies.

Presentation File