Implications of Connected and Automated Driving Systems, Vol. 2: State Legal and Regulatory Audit (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

NCHRP 20-102(07) Interim Report 1 Introduction Connected and Automated Driving Systems (C/ADSs)—a term that in this report encompasses both vehicle connectivity and Automated Driving System (ADS) features—present both opportunities and challenges for state agencies and lawmakers. Among the highest priority challenges is charting how C/ADSs, particularly for high and full driving automation (SAE levels 4–5)3-equipped vehicles, fit within existing legal frameworks at the state level. Over the last 50 years, states have developed elaborate statutes, often consisting of hundreds of sections of state code, dedicated to regulating all facets of vehicle use and transportation. In some states, DMVs and DOTs have promulgated even more detailed standards and rules to implement these statutes. These have also included the setting of standards for law enforcement, including when/where they can intercede depending upon the level of infraction. Most of these codes and regulations, however, were written without any anticipation of C/ADSs. When vehicles are driven by computers rather than humans (or a mix of both), the types of legal statutes required may fundamentally change in some instances, or may at least require modest adjustments to existing laws. This study—part of a larger project for the NCHRP—endeavors to provide assistance to state officials as they work to adapt legal programs to reflect the realities of C/ADSs. While most states are already responding to the challenges of automated technology, some may find themselves forced to operate more reactively, responding to problems as they arise. Given the rapid pace of change associated with these technologies, what is often missing from state efforts is a systematic assessment of whether existing state laws can be modified to account for C/ADSs. This study addresses this need by providing a front-to-back audit of a number of state codes (more than 25% of U.S. states) and then uses this information to identify the types of legal impediments embedded in existing codes that are likely to surface in the future. The results of this audit can be used to support ongoing efforts to prepare state programs for the inevitable and swiftly moving introduction of C/ADSs onto public roadways. In particular, the inventory method presented here offers states a template for conducting their own internal state audits. Such a legal audit offers a number of benefits that are worth highlighting. First and foremost, a legal inventory of existing provisions that may impede or conflict with the safe use of C/ADSs will assist front-line government officials (e.g., state DMVs and DOTs) in identifying and addressing legal barriers and impediments. Not only will this avert the need to take emergency action, but early identification of issues will give states more time to deliberate over what course of action is best and to revise laws accordingly. Second, an audit and analysis of the existing state codes by its nature identifies and extracts specific statutory provisions that ultimately may require attention from the legislature. While 3 Readers should refer to Figure A1 in the Appendix for information on SAE driving levels.

NCHRP 20-102(07) Interim Report 2 DMVs and DOTs often enjoy considerable discretionary authority to promulgate clarifying rules, this authority does not allow the agencies to craft interpretations that conflict with or undermine the clear terms of state statutes. Rather, legislation sets the frame and constraints on vehicle regulation and oversight; thus, in some cases, state legislators will need to make legal modifications. By developing a rigorous accounting of all the state codes (i.e., legislation), decision makers can gain a bird’s-eye view of those legal provisions that can be amended by agencies versus those that will require legislative action. Of course, the best process for modifying existing laws will also depend on whether the issues are of sufficient public import to warrant public deliberations. Highlighting these issues in advance will help to ensure that the right government agencies are weighing in on the difficult issues presented by C/ADSs. Third, a comprehensive audit of existing laws is necessary to ensure that any changes that are made are crafted in ways that maintain legal consistency throughout the larger legislative code. State codes are integrated legal programs that use terms and concepts consistently from one section or program (e.g., licensing and rules of the road) to another. Altering the mode by which vehicles will move (i.e., computers rather than humans) thus intersects in fundamental ways with the entire structure of the law. Piecemeal changes made to the code—such as providing branding options for titles and new license plate requirements for C/ADSs, changing vehicle inspections to account for different equipment in C/ADSs, or enlarging the scope of licensed drivers to include currently excluded groups—without a systemic understanding of the whole code may create internal inconsistencies that risk conflicting with and confusing the larger body of law. It is thus vital to gain an understanding of the entire code and its integrated features before making individual changes to small sections of the code.4 States, especially DMVs and their Counsels, are very aware of these possible conflicts and deal with them in practically every legislative session or almost every time a law is changed. However, most of those changes have not fundamentally addressed key foundational definitions such as “driver,” “vehicle,” “occupant” or “operator”. Fourth and finally, by inventorying existing laws using a relatively objective method that can be applied consistently between states, the legal audit tool developed here offers a replicable method that state regulators and legislators can employ to conduct their own state-specific legal assessments in the future. This effort can also be used in model law development and in an effort to ensure state to state harmonization. At the NCHRP project panel’s behest, rather than reading codes with a much more subjective, open-ended, and unwieldy approach regarding the ways state laws should or could be drafted to integrate C/ADSs,5 this study examines existing codes and asks whether each existing legal requirement may need modification or pose direct 4 Surveying the legal terrain from such a high altitude, however, does have drawbacks. Some details within state codes will be missed, and new laws that are being created will change the legal landscape. In addition, our findings and recommendations have not been tested through judicial opinions, regulatory interpretations, and the public policy lens, all of which are needed to make legal determinations, which normally occur as new technologies are introduced into society. This study will require follow-up with much more detailed, state-by-state examinations, with legal counsel in each state providing assistance. It is our hope that by offering this larger landscape view, we also offer a preliminary template to guide the states as they conduct future, more detailed legal analyses. 5 Beyond the lack of objective benchmarks to guide such an open-ended inquiry, the decision about what could or should be done is inevitably filled with policy choices. Government officials, not academic lawyers, should make these policy decisions. Readers should also note that in Task 1, we do trace current state activity on this open-ended question without weighing in on whether these ongoing enlargements of state codes is desirable.

NCHRP 20-102(07) Interim Report 3 challenges or conflicts with C/ADSs. While we note in passing several places in existing legal systems where the wholesale adoption of new standards or programs may be beneficial, our core recommendations derive from the identification of existing provisions that may require modification due to being rendered obsolete or obsolescent as well as those that may impede the integration of C/ADSs onto state roadways. This approach will also help state officials ground their assessments in ways that help protect audits from variations introduced by a more subjective line of inquiry. The review of statutes and regulations in 15 states as well as the UVC, a model law that is partly adopted by a number of states, identified a number of provisions that may need to be modified if states are to provide comprehensive legal oversight for the deployment of C/ADSs on public roadways. Most urgent in our view is the identification of fundamental terms that are in need of clarification or revision to provide predictability and consistency in the legal treatment of C/ADSs in the states. Terms like “drive,” “driver,” “due care,” and “operator” occur hundreds, sometimes thousands, of times in a single state legislative motor vehicle code and form the underpinnings of much of the code’s legal applicability and jurisdictional reach. Yet these terms are susceptible to substantial ambiguities in C/ADSs where the driver is not necessarily controlling the vehicle. Additionally, as currently drafted in state codes, various rules of the road, render aid provisions, following distance requirements, and attentive driver requirements, to name a few, may impede the legality of C/ADSs and undercut their public benefits. Recommendations for features of these existing provisions that need to be modified are included throughout the findings and analysis. An equally important finding, however, is not the prevalence of laws that need to be modified, but the opposite—namely, the significant gaps in existing legal oversight of C/ADSs, including, for example, definitions of “drive” or “operate” that are tied to a licensed human driver. A front-to-back audit of state codes, in other words, not only reveals the way codes may ultimately benefit from modification, but, by logical extension, highlights the fact that codes often stop far short of where may be desirable to engage with this new technology. At numerous points in the report, we discuss how some C/ADSs, particularly those that are self-driving, may slip through the large sets of laws that currently apply to conventional vehicles, such as rules of the road. We also identify a number of other potential regulatory and related safety concerns— remote operators, mixed human and computer operations that could lead to crashes, the use of software over human judgment—that are sometimes ignored by existing state laws. These gaps indicate areas where modification to existing laws, and in some instances the creation of new laws, may be necessary. The report is organized as follows: • Chapter 1 provides an overview of this legal and regulatory inventory audit, including its objective and scope.

NCHRP 20-102(07) Interim Report 4 • Chapter 2 sets the backdrop for the audit and places it within the context of the larger NCHRP 20-102(07) project. • Chapter 3 presents the methodological considerations guiding the audit, including not only the methods for reviewing state codes, but also the choices and assumptions built into the investigation. • Chapter 4 provides a report of our findings and analysis, broken down into several categories of legal provisions. These categories of provisions track the same responsibilities that the National Highway Traffic Safety Administration (NHTSA) suggests lie primarily within the state, as well as the categories utilized in the questionnaire administered in Task 2: Issue Identification and Stakeholder Outreach. Given that states are not uniform in how their transportation agencies are constructed (some states group their DOTs and DMVs into one agency, while most states separate these agencies, and at least one state combines these agencies with public safety), recommendations may apply to one or all of these agencies. Areas within codes that require modification are indicated with the symbol . • Chapter 5 collects the analyses and recommendations identified in the Findings and Analysis section and provides a summary of key considerations for agencies and a final aggregate analysis that was conducted through the 15-state audit. These are the starting point for the Prioritization Assessment and Harmonization Analysis. • Chapter 6 summarizes the major findings. Since there is considerable detail in the body of the report, particularly Chapter 4: Findings and Analysis, busy readers are directed to the Executive Summary and to the final recommendations section (Chapter 5: Opportunities for Anticipatory Legal Planning and Recommendations). Readers can refer back to the more detailed findings in Chapter 4 as needed.

NCHRP 20-102(07) Interim Report 5 ROADMAP FOR READING THE REPORT Various readers of this report will approach the report with differing goals and interests. Because of the scope and detail, we offer a roadmap tailored to these diverse readers. The roadmap allows a reader to navigate the report more quickly to extract the topics of interest to them without extraneous detail or analyses. • Executives from Departments of Motor Vehicles and Transportation interested in understanding the key legal and regulatory issues should focus on the Executive Summary and the Connected and Automated Driving Systems Legal and Regulatory Prioritization Assessment and Harmonization Analysis. • Readers interested only in the overarching types of changes recommended for state legal codes should skip to the final Chapter 5: Opportunities for Anticipatory Legal Planning and Recommendations. • Staff from Departments of Motor Vehicles interested in conducting their own state audits should read through the full report, beginning to end. Specific subtopics include the following: o The methods for the state audit are explained in Chapter 3: Methods and at Appendix 2: Triggers Used to Identify Problematic State Provisions o A detailed analysis from the audit of 15 states and the UVC is provided in Chapter 4: Findings and Analysis. o The general findings are summarized and listed in Chapter 5: Opportunities for Anticipatory Legal Planning and Recommendations. • Staff from the Departments of Transportation should read Chapter 2: Overview of Automated Technology and its implications for transportation for orientation and then skip to Appendix 4: DOT Traffic Control and Design Element Modifications. • Lawmakers and others interested in particular recommendations for the modification of state codes should begin with an orientation provided in Chapter 2: Overview of Automated Technology and then skip to the general conclusions regarding the types of changes needed, found in Chapter 5: Opportunities for Anticipatory Legal Planning and Recommendations. For particular types of provisions of interest, these readers can then refer back to the relevant sections in Chapter 4 to consider the more detailed analyses that support the recommendations. • Staff from legislative or state agencies interested in privacy and cybersecurity issues that may potentially arise in the states should read Chapter 4: Cybersecurity/Data Recording and Sharing. • Legislative or regulatory staff interested in general recommendations for truck platooning should read Chapter 5: Platoons and Following Distance; those readers interested in more detailed analysis regarding truck platooning should read that section in Chapter 4 regarding Platoons and Following Distance.