Editorial for the special issue – Autonomous vehicle policy

Introduction, purpose, context of special issue Fully autonomous vehicles (AVs) have long been a staple of science fiction and Hollywood.In the coming years, they may become part of our everyday lives.The SAE International (2016) levels of vehicle automation range from 0 to 5, with Level 0 being no automation at all and Level 5 being fully autonomous in all conditions.Currently, the most advanced models available for use by the general public are Level 2 vehicles, with the exception of a few automated vehicle pilots at higher levels.At Level 2, one or more automated driver assistance systems (ADAS) controls both steering and acceleration/deceleration using information about the driving environment and with the expectation that the human driver will perform all the other aspects of a driving situation.And yet, almost every day we learn of another automaker or startup forecasting its introduction of highly automated vehicles.These vehicle technologies have enormous potential to allow for more productive use of time spent in a vehicle and to reduce crashes, costs of congestion, energy consumption, and pollution.They may also alter models of vehicle ownership and patterns of land use, and may create new markets and economic opportunities.Yet they also pose risks and challenges related to safety, cybersecurity, privacy, liability, and much more.Managing their risks and maximizing their benefits requires carefully designed policies that are based on objective research and evidence from the science, technology, and policy research communities.This special issue seeks to draw together recent research in a rapidly evolving field that highlights the specific challenges of developing sound policies to govern AVs, and answers important questions around their use and impacts that would inform such policy.Transport policymakers in various countries are increasingly interested in the large-scale implementation of AVs.However, policy development regarding AVs is hindered by large uncertainties related to their development and deployment, the contribution of such deployments to general transport policy goals, and societal constraints and conditions for AV deployment.There is a need for AV policy that recognizes these uncertainties and provides strategies to address them.In this introduction to the special issue we provide an analytical lens for considering relevant uncertainties within the context of AV policymaking.In Section 2, a framework for structuring uncertainty is presented.In Section 3, relevant uncertainties within the context of AV policymaking are identified according to this framework and mapped to the different contributing papers in this special issue.Conclusions are discussed in Section 4. Framing AV policymaking under uncertaintyPolicymaking on transport requires an integrated view of the various alternative options, their possible consequences for the transport system performance, and societal conditions for implementation.The basis for such a view has been provided by Walker (2000).According to this view, policymaking, in essence, concerns making choices regarding a system in order to change the system outcomes in a desired way.In our case, the system is the transport system (see Fig. 1).The following elements are distinguished in this view:• A transport system can be defined by distinguishing its physical components (e.g., loads, vehicles, and infrastructure) and their mutual interactions.• The results of these interactions are the outcomes of interest, and determine the characteristics of the system that are considered relevant criteria (e.g., lives saved, level of congestion) for the evaluation of policies.• The valuation of outcomes refers to the (relative) importance given to the outcomes by crucial stakeholders, including the vehicle industry, traffic operators, general public, and system users.• Two types of forces act on the system: external forces and policies.Both types of forces are developments outside the system that