Center for Automotive Research offers municipalities long-term recommendations for creating on-road connected and automated vehicle test environments.

Report discusses potential approaches to attract CAV testing activities as well as potential solutions to regional mobility challenges and funding opportunities to pursue.

Saginaw County; Michigan; United States

Background (Show)

Lesson Learned

The following long-term strategies were identified for the successful deployment of CAVs:

Provision general data to contribute to the production of digital maps. Highly automated vehicles will likely require digital basemaps to navigate a given operational design domain. A good digital map for Automated Driving Systems (ADS) will likely include important real-time information on lane closures, work zones, weather, and other dynamic factors. This could be done at the state or even local levels, and could potentially augment information that is already gathered. The foundations for such a model are already established (e.g., dozens of agencies have partnered with the navigation app Waze for data exchange).

Certify infrastructure condition. Policies that reduce or redistribute liability away from industry could accelerate introduction and adoption of automated driving. One potential method to distribute liability is for transportation authorities to certify that designated routes are appropriate for automated driving. Such a certification could include commitments from the transportation authority that the route will not be subject to closure, construction, or other non-typical events without providing appropriate information to mapping services used by automated vehicles. It is likely ADS providers would be more comfortable allowing automated vehicles to operate without direct human supervision if they could be confident that fundamental details of the driving environment do not unexpectedly change.

Invest in digital communications infrastructure. There are a variety of investments municipalities can make in terms of promoting digital communication required for connected vehicles, as detailed below.
  • Global Navigation Satellite System (GNSS) / Global Positioning System (GPS). The precision of GNSS is limited by the availability of a clear signal from multiple satellites. Thus, GNSS localization can be impaired spatially (certain locations receive a weak signal), temporally (the strength of a signal varies as satellites relocate), and environmentally (impediments such as tree cover or buildings block the signal). To improve the accuracy of GNSS, governments have deployed ground-based augmentation systems known broadly as differential-GPS (DGPS). Further proliferation of DGPS stations could improve the performance and reliability of automated vehicles.
  • Fixed Broadband internet backbone. The local availability of reliable broadband may encourage testing and deployment in an area. Much like the U.S. Interstate system provides high-speed, high-capacity, long-distance travel for vehicles, there is a high-capacity digital infrastructure for internet communication formed by long-haul, broadband fiber optic cables. Most longhaul fiber has been installed within transportation rights-of-way (ROW) a practice encouraged by government 'dig once' policies
  • Cellular Vehicle-to-Everything (V2X) and 5G. 4G/LTE networks will likely be the initial means for automated vehicles to communicate with external data systems even direct vehicle-to-vehicle communications. However, opportunities may exist to engage with providers for trial deployments of next-generation cellular technologies (including 5G and cellular vehicle-to-everything (C-V2X)).
  • Dedicated Short-Range Communications (DSRC). DSRC is an ITS-dedicated wireless network that has been piloted by USDOT's Connected Vehicle Program.
Provision Electric Vehicle Charging Stations. The co-evolution of Automated Vehicles and Electric Vehicles (EVs) brings some synergies due to complimentary features of each system. For example, shared automated vehicles could be programmed to automatically return to a base station for charging when the batteries are running low. Furthermore, the integration of EVs with smart grid technologies could save energy and improve the reliability of electric power delivery. Provision of these systems may incentivize automated vehicle networks services providers.

Allocate parking and loading zones. In high traffic/high-service areas, cities often provide bus turnouts and taxi stands so that these vehicles may service passengers without blocking traffic. These facilities work well when use is limited, but if a sizable percentage of travelers expect door-to-door transport via shared automated vehicles, additional provisions may be necessary. The availability of such facilities may be helpful or even necessary for self-driving taxi service providers.

Allocate dedicated lanes. Deploying automated on-road vehicles in dedicated lanes on fixed routes greatly simplifies the task of automation, even if these routes are intersected by other traffic. It is reasonably possible that a bus-rapid-transit (BRT) service could be fully automated in the near future if the infrastructure is designed as to simplify the automation task and limit conflicts with vehicles and pedestrians. City and transportation planners may consider future systems where self-driving, on-demand buses provide convenient and efficient public transit. In the meantime, providing dedicated bus lanes and BRT are proven methods of increasing use of public transit. High-traffic areas may benefit from such infrastructure regardless of whether or not vehicles are driven by a human or computer.
    Ensure that there are provisions to achieve minimal risk conditions. NHTSA states that a "minimal risk condition means low-risk operating condition that an automated driving system automatically resorts to either when a system fails or when the human driver fails to respond appropriately to a request to take over the dynamic driving task." A minimal risk condition will vary accordingly to the type and extent of a given failure, including automatically bringing the vehicle to a safe stop, preferably outside of an active lane of traffic. An appropriately wide shoulder would seem to be the optimal choice for a minimal risk conditions. In areas where providing a dedicated shoulder is excessively difficult, perhaps automated vehicles could be provided dedicated pullout areas at strategic locations, such as immediately preceding exit ramps.

    Incorporate localization beacons where necessary. Autonomous vehicles can have difficulty locating themselves if the road and roadside environment are relatively featureless or monotonous (e.g., on bridges). If the vehicle does not have a reliable GPS location, it could become very confused about where it is in the world. The localization problem is even more difficult in tunnels—where GPS signals are entirely blocked. If road authorities wish to facilitate autonomous vehicles, they could integrate unique landmarks into the infrastructure in areas where the environment is relatively monotonous. Something as simple as bolting 2x4 pieces of lumber to a guardrail in distinct patterns could allow an a to recognize where it is. Localization can also be augmented with digital signals (e.g., Bluetooth localization beacons).

    Lesson Comments

    No comments posted to date

    Comment on this Lesson

    To comment on this lesson, fill in the information below and click on submit. An asterisk (*) indicates a required field. Your name and email address, if provided, will not be posted, but are to contact you, if needed to clarify your comments.


    Opportunities to Encourage On-road Connected and Automated Vehicle Testing: Recommendations for the Saginaw Region

    Author: Brugeman, Valerie Sathe; Eric Paul Dennis; Zahra Bahrani Fard; Michael Schultz; and Richard Wallace

    Published By: Center for Automotive Research

    Source Date: 05/30/2018

    URL: https://www.cargroup.org/wp-content/uploads/2018/05/Opportunities-to-Encourage-OnRoad-CAV-Testing_Saginaw.pdf

    Other Lessons From this Source

    Lesson Contacts

    Lesson Analyst:

    Kathy Thompson


    Average User Rating

    0 ( ratings)

    Rate this Lesson

    (click stars to rate)

    Lesson ID: 2020-00929