Federal Highway Administration National Dialogue on Highway Automation: June 26-27, 2018 Policy and Planning Workshop Summary

Collaboration Corner

Format

The Collaboration Corner was an interactive session designed to gather input from stakeholders on a range of topics. It consisted of a career-fair-style setup with six stations for collecting different types of information from stakeholders. USDOT staff members were located at each station to encourage participation, clarify the exercise, engage in discussion, and ask follow-up questions. Information was collected at each station through two methods:

• Post-it exercise—Attendees used post-it notes to respond to a specific prompt, which was presented on a wall-hanging poster at each station. This was a public form of communication that allowed attendees to view and engage with submitted suggestions.
• Suggestion box—Participants wrote their questions, suggestions, or other input on an index card and placed it into a suggestion box. This was a more private form of communication that allowed attendees to provide information that they may not have been comfortable sharing in a public forum.

Stakeholders provided input on the following topics:

• Communities and AVs: How can communities prepare for and shape the impacts of AVs?
• State and Local Issues: What do organizations need to prepare for an automated future?
• Developing Policy: What new policies are needed and what existing policies need to be modified to enable the development, deployment, and adoption of AVs?
• Research Needs: What research needs to be conducted, and when and by whom?
• Terminology: What terms are frequently used when discussing AVs, and are these words useful or confusing?
• Parking Lot: What are additional questions or comments that don't fit into the other categories?

The following section summarizes key themes and takeaways for each topic.

Communities and AVs

This station focused on how AVs could affect communities and how communities can prepare for and shape these impacts. Input was solicited in five categories, each of which was represented on a separate flip chart at the station: (1) Environment and Air Quality, (2) Mobility and Congestion, (3) Freight and Economic Impacts, (4) Land Use, and (5) Multimodal Safety.

Table 1 Participant Input: Communities and AVs

Environment and Air Quality
Impacts AVs could increase the number of vehicle miles traveled, leading to increased emissions. Sprawl could increase as AVs enable people to travel further at all times of the day, potentially impacting water quality, natural habitats, and social isolation.
Strategies To mitigate the potential impact of greater emissions, encourage the use of alternative fuels and consider limits on hours of operation.
Mobility and Congestion
Impacts Congestion could increase or decrease depending on a number of factors, but the deployment of AVs could lead to a greater number of vehicles, both occupied and unoccupied, traveling greater distances on the road during a broader range of peak hours, potentially leading to multiple rush hours throughout the day. Transit use could decrease or become obsolete, increasing the equity gap between those who can and cannot afford AVs.
Strategies Encourage more efficient use of AVs (e.g., to enable first or last mile connections with transit, to facilitate ride-sharing, etc.) through new revenue models, pricing mechanisms, or other incentives. Proactively identify activities and policies in support of equity and mobility goals.
Freight and Economic Impacts
Impacts Automation will impact jobs in the trucking industry. Job retraining programs may be necessary to adapt drivers' skills to new technologies. AVs could lead to new freight business models, and consideration will be needed for factors such as pickup or delivery and parking and idling as delivery options increase. Impacts of automated freight could be unevenly distributed between urban and rural locations.
Strategies Explore opportunities for job retraining and workforce development.
Land Use
Impacts There are questions about the need for parking in different situations (e.g., stadiums, airports, driveways, etc. AVs will eventually lead to a decrease in private vehicle ownership and reduce the need for parking. AVs may contribute to increased sprawl, which could result in both environmental and community impacts due to decreased density (e.g., stretching emergency services across greater distances).
Strategies Develop guidance for designating safe pickup and drop-off areas and developing adaptable or convertible parking structures. Fiber-optic links are able to communicate information, such as surveillance of intersections and traffic signals, further and more quickly in transportation networks. Also, encourage the installation of fiber-optic technologies in all new construction projects to facilitate the creation of an intelligent transportation system.
Multimodal Safety
Impacts Impacts on vulnerable road users are unclear.
Strategies Collaborate with the Federal Railroad Administration (FRA) about potential benefits from AV interaction with railroad infrastructure and potential challenges at grade crossings. Design infrastructure to protect vulnerable road users, including improving signage and striping. Educate drivers, the general public, the planning community, and others on AV operations and impacts.

State and Local Agency Needs

Participants were asked to provide input on what States and local agencies need to prepare for an automated future. Input was solicited in five categories, each of which was represented on a separate flip chart at the station: (1) Technical Assistance, (2) Information and Tools, (3) Guidance, (4) Workforce Training and Skills, and (5) Other. Participants used color-coded postit notes to delineate between near-term (within 5 years) and long-term (beyond 5 years) needs.

Table 2 Participant Input: State and Local Agency Needs

Technical Assistance
Near-term Standards for data management and sharing.
Needs Guidance to support planning for mixed vehicle fleets. Increased funding to hire technical experts. Peer exchanges and training opportunities to support education on AV technology.
Long-term Development of a training curriculum for State DOTs and MPOs.
Needs Funding for pilot programs and demonstrations in different operating environments. Guidance for State and local agencies on cooperative automated vehicle (CAV) infrastructure implementation. Continuing education for drivers on rapidly changing technologies. Guidance on best practices for data collection and management.
Information and Tools
Near-term Scenarios and performance measures that can be adapted to varying local contents
Needs Tools to support long-range planning. Comprehensive information about Federal funding sources and project eligibility for AVs. Comprehensive information about traffic codes, including State traffic and driving laws and regulations. Updated planning tools to add automation into local Transportation Improvement Programs and Statewide Transportation Improvement Programs, including better public access to these programs to increase understanding of projects. Standardization of evaluation tools and requirements.
Long-term Tools for evaluating land use impacts through scenario planning.
Needs Information about demographics and technology adoption. Strategies to planning for accessibility and inclusivity.
Guidance
Near-term Lessons learned from early deployments to guide future deployment.
Needs Information that is applicable or adaptable to diverse contexts (e.g., population, geographic location). Policy guidance from FHWA.
Long-term Clarification of responsibilities between different agencies.
Needs Guidance for local law enforcement on implementing policies related to AVs. Development of national standards.
Workforce Training and Skills
Near-term Strategies to manage the transition to automation.
Needs Incorporation of transportation technology education into university programs. Capacity building for installation and maintenance of systems. Encouragement of more diverse job categories in public service (e.g., data analysts, computer scientists, etc.).
Long-term Development of a culture of innovation.
Needs Increased hiring of data scientists and cybersecurity experts, among others. Workforce development and apprenticeship programs.
Other
Near-term Guidance on AVs and disaster and incident management.
Needs Consistency in State traffic and motor vehicle laws.
Long-term Funding source flexibility.
Needs

Developing Policy

Participants were asked to identify policies or policy issues that they felt were important in certain categories. They were asked to organize polices into those that could accelerate AVs, manage AV impacts or represent an existing barrier to enabling AVs. Input was solicited in five categories: (1) Multimodal Safety; (2) Mobility and Access for Users; (3) Infrastructure Investment and Funding; (4) Federal, State, and Local Roles.

Table 3 Participant Input: Developing Policy

Multimodal Safety
Accelerate AVs Requiring slower speeds to mitigate complex interactions between AVs and pedestrians or bicyclists could accelerate AVs.
Manage Impacts Policies that address right-of-way use, emissions standards, information security, liability, right to repair, and risk exposure tolerance could help manage AV impacts.
Existing Barriers AVs will need to be able to interface with emergency responders during traffic incidents. Interactions at railroad at-grade crossings is a concern.
Mobility and Access for Users
Accelerate AVs Policies should address current non-licensed drivers to promote mobility for those with specific needs (e.g., elderly, low-income, people with disabilities, etc.).
Manage Impacts Policies could encourage equitable distribution of AV benefits. Policies could promote efficient use of AVs through pricing mechanisms. Agencies should consider how routing algorithms will be developed and applied to AVs. Policies should provide more clarity about AV deployment timelines.
Existing Barriers AVs could lead to unequal access to technology tools and benefits.
Infrastructure Investment and Funding
Accelerate AVs Linking Federal funding to technology investment could accelerate AVs. Policies should improve infrastructure for both conventional vehicles and all levels of automated vehicles (not just Level 4 and 5).3 Policies could provide financial support for communities managing AV deployments to fund costs when partnering with AV technology companies. Policies could expand pilots on alternative funding methods, like VMT pilots, to prepare for the lack of traditional revenue sources as more AVs are deployed.
Manage Impacts Public agencies could evaluate new mechanisms for toll collection, payment,and enforcement. Public agencies may need to address equity and environmental justice issues to manage infrastructure investments that only benefit high-end vehicles. Policies could incentivize public-private partnerships.
Existing Barriers The current infrastructure funding system is insufficient. Prioritizing smart corridors for infrastructure investment could help overcome existing barriers.
Federal, State, and Local Roles
Accelerate AVs Expediting standards for AVs and connected systems could be a government role. Enabling testing and deployment of AVs while continuing to ensure safety is a government role. Public agencies may need to change State laws to enable testing and deployment.
Manage Impacts Public agencies could:  Develop guidance for law enforcement, Encourage public participation in the policy decision-making process, Define responsibilities for collecting and managing data, Develop training standards for drivers and operators, and Regulate safety-critical communications
Are Existing Barriers Current rules prohibiting the use of proprietary products on program and highway projects could be an existing barrier. State licensing and insurance requirements could be a barrier to AVs. Out-of-date Manual on Uniform Traffic Control Devices (MUTCD) requirements could also be a barrier to AVs.

Research Needs

Participants were asked to identify what research should be conducted, by whom, and when. Research needs were solicited in three categories, each of which was represented on a separate flip-chart at the station: (1) Urgent (by 2020), (2) Medium-Term (by 2025), and (3) Long-Term (by 2030 or later). Participants used color-coded post-it notes to delineate between public and private sector research responsibilities.

Table 4 Participant Input: Research Needs

Research Areas: Urgent (by 2020)
Public Sector Public opinion and values Impact on vulnerable road users Timelines for automated vehicle adoption and fleet penetration Challenges and implications of Level 3 automation4 Certification standards Societal impacts Transit impacts Accuracy of computer vision
Private Sector Work zone safety impacts Human Machine Interface and public awareness Liability Roadway safety devices that serve both human and automated driving Innovative payment methods Pavement and bridge impacts
Research Areas: Medium-Term (by 2025)
Public Sector Impact of AVs on traditional infrastructure investment Impact on rural communities Infrastructure and maintenance needs as traffic patterns change Areas where AV markets and technology cause economic disruption Impacts of AVs on tribal lands
Private Sector Household location and mode choice impacts Impacts on mobility for people with disabilities Opportunities and considerations for proprietary data sharing Use cases and business models for AVs
Research Areas: Long-Term (by 2030 or later)
Public Sector Use of AVs for disaster or incident management Land use considerations as transportation infrastructure changes Impacts on driving skills when AV adoption is widespread Updates to the MUTCD
Private Sector Standardized model for redundant systems and testing Updates to Highway Capacity Manual

Terminology

Participants shared common AV terminology and indicated which terms are helpful and which are confusing. They placed these terms along two axes to show how these terms are used. The vertical axis represented the frequency with which these terms are used, and the horizontal axis represented the level of confusion surrounding their use. The table below illustrates the terms placed into each quadrant. Some of the most confusing and frequently encountered terms included "autopilot", "autonomous", and "connectivity". Participants were also unsure of the distinctions among pilots, testing, and full deployment or operation.

Table 5 Participant Input: Terminology

	Confusing	Clear
Frequency	Autopilot Autonomous Advanced Transportation and Congestion Management Technologies Demonstration FHWA's role Operator vs. Driver Sensor needs vs. High Definition mapping needs State of good repair Vehicle to Infrastructure (V2I)/Vehicle to Vehicle (V2V) Dynamic mapping elements Connectivity Self-driving car Digital infrastructure Levels of automation Level 4 Automation (needs subcategories)	Original Equipment Manufacturers Users Roadside Unit Market penetration of AVs different levels Driverless Connected vehicle Robotaxi Society of Automotive Engineers J3015
	Highway Operational Design Domain AASHTO Green Book Manual for Uniform Traffic Control Devices Traffic control Level 1-5 Connected Automated Vehicle/Automated Vehicle Mobility as a service (MaaS) vs. Mobility on demand Shared mobility	Fully autonomous Partially automated Driverless vehicle Connected/Automated Autonomous ≠ Connected

Parking Lot

Participants shared any remaining questions and comments that did not cleanly fit into the other topic areas. Topics included:

  

• Data ownership and management considerations
• Public health effects (e.g., more access to healthcare, but also potential increase in sedentary behaviors)
• Need for more diverse stakeholder engagement
• Comparison with international activities
• Consistency of standards and policies
• Accessibility and equity considerations
• Workforce development and capacity building
• Need for new organizational and institutional models
• Cybersecurity considerations