Impacts of connected and automated driving on national road authorities’ policy targets

The main purpose of this article is to summarize state of the art on the impacts of connected and automated driving (CAD) on travel demand, travel behavior, traffic flow, safety and energy (reported in MANTRA deliverable D3.1). The review is based on ongoing and recently completed EU and national projects, and a comprehensive literature review of key publications and articles on the topic.

Most of the CAD impact estimates in the current literature are based on either expert evaluation or traffic simulations. The other source for current estimates are available field studies on driver assistance systems. Moreover, even the models to estimate impacts, e.g. traffic microsimulation models, still need adjustment and parameters designed specifically for automated vehicles. The current vehicle behavior models are based on the behavior of human drivers. In addition, the behavior of human drivers might also change when interacting with automated vehicles. The development of the technology also have great impact on the area and conditions where automation can be used (operational design domain, ODD), and hence can have impacts in. The variety of impact mechanisms need to be kept in mind when considering the potential impacts of connected and automated driving on not only traffic safety, but also other impact areas.

Connected and automated driving

Highway cutting through forest in Finland. Photo: Julius Jansson/Mostphotos


Expected impacts of connected and automated driving

The studies reviewed give an overview of the expected impacts of the deployment of connected and automated driving. As the reader can see when going through the various impact areas, the expectations of the magnitude of the impacts vary a lot. Where someone is expecting traffic safety to be improved by 90%, others are much more conservative and present only one-digit estimates. The same applies for other impact areas; even fully contradicting estimates exist.

One can anticipate that AVs in mixed traffic will tend to drive at longer following distances so as to permit less rapid deceleration when the preceding vehicles slow down, to negotiate sharp curves at lower speeds and to be less prone to undertake rapid manoeuvres such as abrupt lane changes compared to human drivers in the same situations. Such smoother driving may reduce incidents, but there might also be an impact in traffic throughput, i.e. decreased capacity.

Another issue closely related to capacity is that automated vehicles even when shared can compete with public transport and active transport modes (walking and bicycling) leading to better individual mobility but less transport system efficiency. In addition, when assessing safety, the additional risks the technology could introduce need to be taken into account. Moreover, as long as there is mixed traffic, i.e. not all vehicles are fully automated, conventional safety strategies are still needed.

Waiting for more studies

It is important to keep in mind that even if the search for the literature and studies is extensive, it only gives an overview of the situation as of today. Most of the impact estimates are based on either expert evaluation or traffic simulation or make their assumptions based on the available studies on driver assistance systems. It is of utmost important to continue following the studies in this area, to complement the results when on-road testing of automation and data received from those tests are available in large scale. In addition, one area, which remains open, and which will definitely develop over time, is the interaction between automated vehicles and other road users.

As many of the studies summarized in this deliverable also remind: automation is not the only megatrend that affects road transport in the oncoming years. Shared mobility is one issue, which may have great impact on how people select to move around. In addition, electrification will certainly have an impact on CO2 emissions, and maybe even on travelling patterns.

MANTRA is an acronym for “Making full use of Automation for National Transport and Road Authorities – NRA Core Business”.  MANTRA is a CEDR-funded two-year project responding to the CEDR Automation Call 2017 question: How will automation change the core business of national road authorities.

More information about MANTRA-project.

Merja Penttinen
VTT Technical Research Centre of Finland, Finland





This entry was posted in Intelligent Transport Systems, Safety, Simulation, Traffic and Traffic Control and tagged , , , , . Bookmark the permalink.

Comments are closed.