Project: #26 Bystander Interactions with Failing Vehicle Autonomy Progress Report - Reporting Period Ending: Sept. 30, 2018 Principal Investigator: Aaron Steinfeld Status: Completed Start Date: Jan. 1, 2017 End Date: Aug. 31, 2018 Research Type: Advanced Grant Type: Research Grant Program: MAP-21 TSET National (2013 - 2018) Grant Cycle: 2017 TSET UTC Progress Report (Last Updated: Oct. 20, 2018, 8:36 a.m.) % Project Completed to Date: 100 % Grant Award Expended: 100 % Match Expended & Document: 100 USDOT Requirements Accomplishments During this project the team developed and ran an experiment protocol that gathered bystander input about AVs across summer and winter periods. Team members waited in areas of Pittsburgh where Uber AVs frequently pass through. After such an occurrence, team members would intercept participants and ask them the protocol questions. This led to a paper at the competitive AutomotiveUI conference and the team presented findings at multiple non-publication events. This work was done in collaboration with Jodi Forlizzi. In parallel with this, a MS HCII student conducted an independent study (Spring 2018 semester) with Steinfeld and Forlizzi on communication methods between pedestrians and AVs. A set of designed interactions were prepared and tested in a field study in Pittsburgh. This work revealed numerous areas of possible confusion in bystanders about the vehicle's intent and what they were supposed to do. For example, do red lights on the windshield indicate the vehicle will stay stopped or that pedestrians should stop? We also tested words and icons, with similar results. These findings have been discussed with industry and other researchers attempting to run similar studies at a larger scale in more controlled conditions (MIT, VA Tech, etc). Steinfeld was also active in outreach and dissemination efforts: - Quoted in: http://www.businessinsider.com/how-self-driving-better-human-drivers-2016-12 - Quoted in: http://www.latimes.com/business/autos/la-fi-hy-driverless-regulations-california-20170126-story.html - Quoted in: https://www.popsci.com/ibm-self-driving-car-patent-prevent-crashes#page-4 - Invited panelist at the Autonomous Vehicle Policy Series #3, "Deepen Your Understanding of Critical Issues with Data Sharing and Processing," which occurred in Washington, DC on June 2, 2017. This event was organized and hosted by ATI21. - Quoted in: https://www.fdiintelligence.com/Sectors/Automotive-OEM/What-self-driving-vehicle-technology-means-for-FDI - Quoted in: http://sandhill.com/article/the-future-of-smart-cars-is-now/ - Conversations with Department of Labor and Society of Automotive Engineers regarding accessibility of AVs for people with disabilities. - Meeting with representatives of the Government Accounting Office regarding the future of work as robots and autonomy become more prevalent. - Hosted a tour of 15-20 students from the STEM Academy of Mt. Lebanon High School. Stops included labs related to the UTC. Impacts The team aggressively disseminated this study's findings to industry and policy makers through small meetings, presentations at conferences, and a high quality conference publication. Insights on these outcomes will help inform research, development, and policy regarding how AVs and associated systems convey failure and status. Two of the primary MS students on this project have joined industry (Google and iRobot), thereby supporting additional knowledge transfer. The lead author of the published paper is a PhD student who has incorporated aspects of this project into her long-term research agenda. All of the students who worked on this project, either funded or through class projects, are women. Other None Outcomes New Partners None Issues None