Long post with digressions on the way to the model AV law

This past February - remember 2020 just before our lives were turned upside down? - I was too busy to write about the formal approval of a model AV statute for US states. The drafting and approval process took a few years and, for disclosure purposes, I participated in committee discussions, although only representatives of states were permitted to vote.  Also participating, but not voting, were staff from the insurance industry, Uber, Waymo, and others. 

If you please, skip past the digressions to the main point below. Actually, I am in a wordy writing mood and I don't feel like editing, so there's some digressions even once we arrive at the main discussion.

Digression #1: Where I was in February

At the time, I was the AV guru, not an official title, at the Community Transportation Association of America (CTAA), which effectively serves all of those places and populations that the large companies ignore; places with a low density of humans, low-income tending populations of older adults and people with disabilities, and those vast stretches of the US with not much, if any, public transportation. 

Note to New Yorkers, who barely enjoy the quality of public transit available in other large international cities: CTAA focuses on communities and regions where catching a taxi, an Uber, or a bus, let alone a subway ride, would be fraught with difficulty, if possible at all. I would not count on much walkability either. You might also be hours from an airport or a hospital. 

Digression #2: A little civics lesson before explaining what is a uniform law and who ever heard of the Uniform Law Commission? 

To understand why uniform state laws are a positive, it's good to have some knowledge here about US history and the US Constitution, so I will give a totally unofficial and incomplete lesson for those unfamiliar with why progress can be so difficult to achieve across the US. Our Constitution is a document that, even with its amendments, sets up a government structure very unlike any centralized national system. 

First, the Constitution, when it was drafted, was a nice piece of political philosophy, but it could not become the law of the land until enough states ratified it, which was by no means a done deal. Both proper advocacy - in the form of wonderful speeches, newspaper articles, and letters - as well as somewhat shady political wheeling and dealing, miraculously combined in favor of ratification in the early states to convene their own conventions, until ratification became a fait accompli. But this did not happen until a general agreement was reached that there would soon be 10 amendments included, known as the Bill of Rights. Most of these rights - to a jury trial, against unreasonable search and seizure, etc. - protect the individual against the government. 

The Founding Fathers were quite paranoid about government overreach, sometimes more than about actually ensuring that the national government could accomplish what it needed to. The Bill of Rights and the limited powers given to Congress and the President are a testament to that skepticism of a strong national government. Many topics were basically left to the states.

On top of the suspicion of centralized power, there was fear of rule by urban majorities, sometimes known as mobs. Democracy was not the Founders' favorite form of government; they instead established a republic. There was also slavery and those seeking to protect that hideous practice. This is the time of compromises where rural states got equal power in the Senate and enslaved persons would be counted as 3/5 of a person, although they had zero votes and the votes of many of their descendants continue to be suppressed to this day.

Digression #3: The whole states vs. national government thing gets even more complicated - and not in a good way.

To add to the mess of a a national government working alongside however many state governments, currently counting in at 50, each state operates differently from its peers. Some states place more power in the governor, some in the legislature, some in the legislative leadership, some in the counties, some in the cities, some in bizarre commissions where some unelected person somehow develops outsized influence and keeps it for a long time. Some states have a government structure that works and there is collaboration and problem solving. Others do not. Passing over a state line can make all of the difference or it might not. Then there is DC and US territories, all adding a bit more variety to the mix. We won't digress further to go down that particular rabbit hole.

In terms of transportation, for the past 70 or so years there has been an understanding that the motor vehicle is the king among modes, that its dangerousness in terms of crashes and public health will be topics mostly avoided (except for the blip of the early NHTSA years), and that, in terms of funding, other modes will receive the crumbs. No national leader of either party has significantly deviated from this understanding because most voters and campaign donors have cars and mainly drive to get around. Or they are connected to the auto industry, which works hard to keep the understanding firmly in place.

So, enough already - get to the point

Arriving at the point - The Uniform Law Commission AV Model State Law

One path to a national law is Congress, but another is passage of separate state laws that all say the same thing, or nearly the same thing. That is where the Uniform Law Commission (ULC) comes in. I have explained its role before. It is quite unlikely that a uniform state law will end up being the way we regulate AVs or make any major changes to this realm of interstate commerce. For one thing, the car companies are comfortable working with Congress and that is the partner they would prefer to keep dancing with.

On the other hand, it is expected that registration, ownership, licensing, insuring, and perhaps setting standards for inspecting and maintaining AVs will be controlled at the state level. If nothing else, even if a uniform law never gets passed in the legislatures of the 50 states, DC, and assorted territories, the background consensus reached by the ULC drafters of the AV model law is a good read. 

The Uniform Automated Operation of Vehicles Act (UAOVA)

The UAOVA only focuses on full automation where there is zero human operation of the vehicle. However, the UAOVA or its comments acknowledge vehicles that have all of the following capabilities: 

1. Complete automation;
2. Driver assist capability;
3. Remote operation; and 
4. On-board switch to select or toggle between complete automation or human driver operation.

As for a vehicle that allows for either human or fully automated operation, the UAOVA would only govern full automation capability or time periods when the vehicle is actually in fully automated operation. The UAOVA does conceive that a vehicle without automation could become an AV and that the AV would then be required to be registered as such.

The What the ULC AV model legislation does not cover:

• AV testing or pilots
• Remote operation of AVs
• Automated driver assistance systems
• AV planning
• Any human driver equipment or operation on board the vehicle

The UAOVA model legislation is more of an addition to a state motor vehicle code. By being agnostic as to modes and roadway design and speeds, it basically avoids and, therefore, perpetuates our current modal split and the inconvenient conundrum of road safety. One step at a time.

Slight digression: Inconsistent definitions

The UAOVA uses the term "automated vehicle" instead of "autonomous vehicle," which is unfortunate because the USDOT refers to "automated vehicle" as one that has any level of automation, even a low level, whereas the UAOVA uses the same term to mean a vehicle fully capable of driving itself without human assistance. 

USDOT's definitions regarding AVs can be found in AV 3.0. AV 4.0 does not provide a glossary. 

Automation: Use of electronic or mechanical devices to operate one or more functions of a vehicle without direct human input. Generally applies to all modes. 

Automated Driving System (ADS): The hardware and software that are collectively capable of performing the entire Dynamic Driving Task on a sustained basis, regardless of whether it is limited to a specific operational design domain. This term is used specifically to describe a Level 3, 4, or 5 driving automation system. (SAE J3016) 

Automated Vehicle: Any vehicle equipped with driving automation technologies (as defined in SAE J3016). This term can refer to a vehicle fitted with any form of driving automation. (SAE Level 1–5)

It is critical to be clear on the definitions because people who are well versed in this whole AV rabbit hole just may be using the same terms to mean entirely different things. 

Who is driving?

What the UAOVA (sounds like a gynecological procedure, no?) answers the question of who/what is the responsible party when the software and hardware combination is driving rather than the human. 

Under the act, a qualified entity declares to the state that it will be the legal driver for certain automated vehicles. Provided that it meets certain qualifications, this “automated driving provider” might be an automated driving system developer, a vehicle manufacturer, a component supplier, a data provider, a fleet operator, an insurer, an affiliated firm, or another kind of market participant that has yet to emerge. The automated driving provider is primarily defined not by a specific role in the stream of commerce but, rather, by a willingness to self-identify and an ability to meet the technical and legal requirements specified in the act.

    ...            ...            ...         

The act uses the motor vehicle registration framework that already exists in states—and that already applies to both conventional and automated vehicles—to encourage automated driving providers to self-identify. Existing law generally requires the registration of a motor vehicle that is operated on a public road, and the vehicle’s owner or lessee typically obtains this registration. Under this act, however, an owner or lessee may register an automated vehicle only if an automated driving provider has designated that vehicle as an associated automated vehicle. If the automated vehicle is not “associated” in this way with an automated driving provider, then it may not be registered and therefore may not be operated on public roads. [Emphasis added.]

Associated?

To be clear, the AV registrant might not be the automated driving provider. The AV might be owned by an individual, but it's software updates and operation will be controlled by the entity that is recognized as the automated driving provider, an entity that must be "has designated that vehicle as an associated automated vehicle. If the automated vehicle is not “associated” in this way with an automated driving provider, then it may not be registered and therefore may not be operated on public roads." 

Is this workable? 💁 It's confusing. I need a break.

Musical Intermission: A performance of Don Juan from Smokey Joe's Cafe. The song provides a very weak analogy about the availability of a technology tied to wealth. It's a good intro to the next section of this entirely too long blog post.

Humans - No license necessary

People with disabilities for whom driving is difficult or impossible, irresponsible teenagers out and about, older adults who don't want to or feel uncomfortable driving, parents who prefer to spend quality time with kids rather than focusing on the road, and people who would prefer to nap rather than drive will no longer be inconvenienced by the requirement to have a license or to have a licensed human behind the wheel - if there is even a steering wheel. The UAOVA does away with the requirement that a licensed human driver be present and ready to operate the vehicle. An AV would be permitted to be operated without human intervention. 

Warning: Let's not confuse legal permission to ride in an AV without a licensed driver, which the UAOVA grants, and actual affordable, accessible, or available AVs. You now have the right to purchase or rent that pretty private jet and and the legal authority to hire a pilot, but that does not mean that you have either the money to do so or an airport or airfield nearby.

Not so familiar way to get to Delaware?

The UAOVA seeks to build upon a familiar legal framework of registration. The question arises then of which state would a company prefer to be registered by? Will we have a Delaware equivalent for AVs, where the state with the most corporate-friendly standards registers the vast majority of AVs? 

The UAOVA does suggest in its comments that "[i]ndividually or in concert, states may also wish to develop a system to track and sanction automated driving providers that is comparable to the one for human drivers." (Please note that in the US commercial vehicles are a whole different ball of wax and are regulated differently than car, taxi, or ridehailing vehicles.) 

Why wouldn't a company choose one state with lax standards rather than registering in up to 50 separate state regimes and standards, not to mention DC and US territories, all with different standards? I am pretty sure that at least one state would volunteer to be the lax state; perhaps Nevada, Arizona, Texas, or Florida.  Maybe Michigan. 

Registration as an Enforcement Mechanism

Vehicle registration and inspection laws require that vehicles be in some minimal operable condition. There are also emission regulations that mandate performance standards. This could be a potent weapon for safety. Under the UAOVA, a state would have the authority to:

decline, suspend, revoke, or decline to renew the registration of an automated vehicle that is not: (1) an associated automated vehicle; (2) associated with an automated-driving provider recognized by [the relevant state agency]; (3) properly maintained; (4) lawfully insured; (5) compliant with a registration requirement; or (6) fit to be operated.

Who can be an automated-driving provider?

Three types of businesses are permitted to register vehicles as automated-driving providers: 

1. Traditional car companies; 
2. Companies that "have submitted to the United States National Highway Traffic Safety Administration a safety self-assessment or equivalent report for the automated-driving system as required or permitted by the United States National Highway Traffic Safety Administration;" or
3. Companies that have "participated in a substantial manner in the development of an automated driving system."

No. 2 could become a gaping loophole, especially under the current NHTSA hands-off regime where voluntariness is the order of the day and companies submit glossy, ad-like self-assessments. Don't get me wrong, the provisions relating to automated-driving providers could have real teeth, but the type of tech-informed serious enforcement we are talking about is generally done at the federal level, with few exceptions. One has to ask whether a state such as California or New York would step up? Another question is whether the type of enforcement possible here to regulate automated-driving providers is better accomplished at the federal level?

I am not answering these questions because we have seen the last few years that Attorneys General of states can do a fabulous job of protecting ordinary citizens around the country. The idea that the responsibility and authority should only be vested in the executive branch of the federal government is putting all of one's eggs into one basket. 

Associated vehicle of the automated-driving provider

Now we have an automated-driving provider, who effectively claims the "associated vehicle." "For the owner of an automated vehicle to register the vehicle, an automated driving provider must have designated that vehicle as an associated automated vehicle." In effect, the legal entity that must be approved to "drive" is the automated-driving provider. A comment within the UAOVA explains:

In other words: A human driver must obtain a license, whereas an automated driving provider must make a declaration. A human becomes a driver by driving a vehicle, whereas an automated driving provider becomes a driver by designating an associated automated vehicle that is then used under automated operation. Both conventional and automated vehicles are typically registered by their owners. The owner (or lessee) of a conventional vehicle may or may not be its driver, and the owner (or lessee) of an automated vehicle may or may not be its automated driving provider.

When the equipment is the driver and why that relates at all to the Citizens United Supreme Court decision

The UAOVA mandates the usual insofar as requiring equipment to be in operating condition, but for a conventional vehicle the driver is separate from the equipment - there is a human operator and a machine that he or she operates. Now the mandate of proper maintenance means that the elf inside the AV (like the elves inside my laptop) will not possibly fall asleep as the AV rides down the road. With current technology, that could mean a computerized "inspection" that occurs whenever the AV is turned on rather than an annual physical inspection.  Could be does not mean will be.

Indeed the last section of the UAOVA provides that the automated driving provider "is responsible for a violation of [the state’s rules of the road] during automated operation of an associated automated vehicle." The state would have the express authority to fine an automated driving provider for violations, but the real enforcement mechanism - presuming that contracts retain these rights for passengers and even individual AV owners - could be lawsuits filed against automated driving providers following any crashes. 

Don't count on that. 

Minor digression (not an apology): Corporations in our current lobbying and campaign contribution system have outsized power. Individuals do not read through long contracts that divest them of their legal rights. You can be sure that automated driving providers will draft contracts that screw the rider, the owner, the lessee, or whatever other individual pays for or uses the AV. Unless we pass a constitutional amendment that reverses the Supreme Court decision in Citizens United and gets money either out of or reduces its role in politics, AV contracts will not protect the regular person.

This is not that corporations or their leaders are evil, it is that they are rational beings acting in a legal manner to reduce their liability and legal costs. In my opinion, as is pretty obvious, corporations are not people and they should not have the right to control the political system by donating tons of cash. The video is Senator Jon Tester's one-minute explanation of the issue.

We don't know

No one knows how AV regulation will play out or what will happen along the way to influence legislation at either the federal or state levels. 

In the UAOVA comments, the drafters acknowledge that the Act could be used as a sharp enforcement mechanism or not.

[T]he automated driving provider is the legal driver ... and is therefore subject to corresponding sanctions under the state’s vehicle code. In other words, the automated driving provider should receive the speeding ticket when an associated automated vehicle under automated operation is caught speeding. At the same time, this section does not address the appropriate level of enforcement. It is expected that federal, state, and local authorities will continue to evaluate the role of various forms of automated enforcement (including self-reporting obligations) in improving road traffic safety.

Since the Uniform Law Commission's approval of the UAOVA in February no state has adopted it. IF you have read the entirety of this enormously long blog post, you either have insomnia or you remained interested. Congratulations and good luck with the sleeping thing.

Hawaii Dipping Toes, Not Surfing

For the first time in 2020, a US state has passed an AV-related law, but the newly enacted Hawaii law does not exactly grant permission for wide scale AV operations on public roads. Indeed, although more than half of US states have passed legislation that in some way touches upon autonomous or partially automated vehicles, only California has anywhere near a comprehensive regulatory system in place. California can get away with that because a major portion of the AV engineering workforce wants to live in and around San Francisco/Silicon Valley. That is a unique situation that many companies have chosen to accept.

Toes in the Water 

Hawaii joins the ranks of those states that wish to appear as if they are doing something, the equivalent of dipping ones toes in the water, without really accomplishing anything. That is fine; the legislators had a conversation. Some state representatives probably examined the issues closely. Nothing to be ashamed of.

Designed by Freepik.

I Digress to an Unrelated and False Impression of Hawaii

Now I've never been to Hawaii and my first introduction to that state was from watching reruns on afternoon television of Gidget Goes Hawaiian. No doubt it was a completely accurate depiction of the surfer life. Carl Reiner plays her dad. According to Wikipedia, the movie was filmed on location at the Royal Hawaiian Hotel, but a stand-in for the actress who played Gidget did the actual surfing. 

I've known people who have lived in Hawaii or spent considerable time there, but Gidget and her boyfriend Moondoggie (played by James Darren), alas, created a lasting false impression. This Gidget actress was not as good in the role as either Sandra Dee, the first Gidget, or Sally Field, who played Gidget in the TV show, before she went on to Flying Nun fame. Enough for 60s television and my elementary school years doing homework while sitting in front of the TV.

Back to Hawaii's Legislation

Hawaii's law creates an "autonomous vehicle testing program" within the state's department of transportation, without mention of funding or other details. The one pertinent point that the state legislators made when writing the law was to make sure that the citizens of Hawaii and its visitors would not feel or be threatened by a Batman-like AV. (I could not help myself after delving into ancient TV references.) Although AVs are expressly permitted to be tested on public roads, "a conventional human driver shall remain physically present in the vehicle at all times" just in case human intervention is needed.

Now before your mind goes where mine did, straight to humorous definitions of who or who might not be considered conventional, the law defines "conventional human driver" as a person who manually operates a vehicle. Fun fact: The law does not, I repeat NOT, require that the "conventional human driver" be licensed to drive in Hawaii or anywhere else. So, to appear as a conventional human, wear a polo shirt with a collar, look like you are headed to a golf course, but don't worry about passing the driving test.

Like many state laws before it, the Hawaii statute requires that a report be written. The deadline is about three weeks before the legislature convenes in 2023. This year, they convened in mid-January.

I'm back

I am back to blogging again after a very busy couple of years of work. I haven't even changed my LinkedIn profile or anything else that needs editing. Indeed, I should be sending my new phone number out to friends and relatives or else I will soon be receiving emails along the spectrum from concerned to panicked. I am writing and, for the moment, it just feels calming not to proceed from morning until evening, and sometimes when wakening in the wee hours of the night, with my triage list churning in my brain.

This election matters to the survival of democracy; it matters very much to public health. But there is not much in the way of conversation about transportation per se, how we fund it, what we fund, or what exactly are our goals on issues related to transportation. Why? I believe that most people view transportation as local rather than national, that they have no idea how transportation funding decisions are made, and that the inertia and legal-favoring of personal vehicle use is simply a presumption in most US communities. 

Imagine you redo the bathroom

Transportation is like grout. When you redo the kitchen or the bathroom, you spend lots of time and focus on the tile and two minutes on the grout. But tile is nowhere without the grout. People do not usually move to or remain in a place because of the transportation choices, but education, work, healthcare, gathering with friends and family, or simply a walk in the neighborhood or to the grocery store, are nowhere without the transportation connection to them, whether that's a bike lane, a sidewalk, a bus, or a car - and the roadway and traffic signals maintained to support them.

To change modal and systemic priorities, such that transit and other low-cost shared-use modes - including sidewalks and safe pedestrian crossings - get a bigger piece of the pie requires lots of advocacy, public relations, persistence, and luck that the times are ripe for the message. With Covid-19, with relatively new modal offerings, with incredible possibilities that technology brings, with a greater share of society finding it more and more difficult to pay the price of owning and maintaining vehicles, and with a seismic shift in transportation patterns, perhaps the time is almost ripe.  

[From https://vocal.media/theSwamp/buses-were-always-the-answer]

My thoughts on autonomous vehicles assume that we have an opportunity with this upcoming major change to make other changes. That doesn't mean that we will take advantage of that opportunity.  

No answers today. Just a rant.

3 New AV State Laws - 3 Approaches

While we wait for Congress to act on automated vehicles (AVs), and AVs are probably not high on the list right now, three new state laws have recently passed and become law. These are from Florida, Iowa. and Oklahoma. Louisiana has legislation pending that is likely to pass.

Short memory and low expectations

Perhaps we are sufficiently far in time from two realities that state legislatures are again moving quickly on AVs.

1. Mar. 18, 2018 - the date of the Uber not-really-an-AV crash in Arizona; and
2. Congress failed to pass a national AV law in the last session.

At the moment, states perceive themselves to be in the driver's seat in pushing AVs forward. The  new laws uniformly assert state supremacy, preempting local regulation or restrictions on AV operations.

One trend is a change in legislative definitions is that the term "automated driving systems" has become common and has generally replaced the term "autonomous vehicles." It has also become commonplace to require that AVs be able, among other things, to achieve a "minimal risk condition." The Florida, Iowa. and Oklahoma laws include these.

Florida

Florida HB 311

Whole hog - "human operator is not required to operate a fully autonomous vehicle; authorizing a fully autonomous vehicle to operate in this state regardless of whether a human operator is physically present in the vehicle"
No human need be present when AV is in operation
Allows teleoperation of AV

Allowing operation of an AV with video or television entertainment
Allowing operation of an AV with wireless devices in use

Definitions edited from earlier AV law
Preempts local governments from in any way regulating AV ridehailing (with some exception for airport authorities)

A provision that governs partial automation technology looks like fodder for litigation:

If the autonomous vehicle is not fully autonomous, the vehicle must: (a) have a system to safely alert a licensed human operator physically present in the vehicle if an automated driving system failure is detected while the automated driving system is engaged. When an alert is given, the system must: require the licensed human operator to take control of the autonomous vehicle or must achieve a minimal risk condition. 

Vague language allowing the Florida Turnpike Enterprise to "fund, construct, and operate facilities" that "advance" AV transportation for the purpose of reducing congestion AND improving safety. Note to Florida Turnpike Enterprise: The safest modes to integrate into your AV infrastructure are walking, biking, and transit.

Addresses long-range MPO transportation plans and mandates that "Each M.P.O. is encouraged to consider strategies that integrate transportation and land use planning to provide for sustainable development and reduce greenhouse gas emissions."

Oklahoma

Oklahoma SB 365

1. Preempts municipalities
2. Defines terms
3. Creates new section of vehicle law

Nothing in the Oklahoma law specifically discusses AV operation on public roads.

Iowa

Iowa SF 302

Allows for AVs to operate on public roads, but a human driver must be present in the vehicle. In fact, in the Iowa framework, it is still the human driver who is responsible, meaning the party who is potentially liable. Nothing about the possible liability of software providers or of hardware or vehicle manufacturers.

Local governments are preempted from regulating or taxing AVs, AV operations, or the business of AV ridehailing.

The Iowa law provides for AV ridehailing, but it does not explicitly permit an AV to operate without a human driver present. This leads one to believe that the law means that a human driver must be present when a ridehailing passenger is picked up or when a delivery is made with an AV.

Louisiana legislation pending

Louisiana HB 455 - This legislation is pending in the Louisiana Senate. It only concerns AV commercial motor vehicles and probably is geared toward attracting the trucking industry testing going on down South. [Update on July 9, 2019: The Louisiana bill passed and has become law.

Defines AV terms
Autonomous commercial motor vehicles - meaning trucks and buses - may operate in Louisiana
"Prior to commencing the operation of an autonomous commercial motor vehicle without a conventional driver present in the cab, a person or entity shall submit a written statement to the Department of Transportation and Development certifying that the vehicle meets the requirements of this Section."
Allows for teleoperation of automated commercial vehicle, but a human remote operator is considered the operator during such periods of operation.
If an AV commercial vehicle is equipped with teleoperation capability, it may operate on public roads without a human present, but only if "if a remote driver is operating the vehicle."

I'm confused, is the intention that the vehicle be automated or only partially automated. Is this really a commercial vehicle teleoperation bill? Is this a platooning bill, though that term or a description of such activity is left unmentioned?

AV State Laws Passed From 2018 to Present

I was wondering recently what has been the trajectory in terms of state laws passed in the post-2016-17 heyday of perfectly safe AVs will soon be here! The killing - yes, killing - of Elaine Herzberg on Mar. 18, 2018, with a combination of unsafe pedestrian infrastructure and Uber's hubris was a major dump of cold water on a free pass for lenient AV legislation. Ms Herzberg did not die in vain; state legislatures slowed down considerably.

This slowdown did not mean inaction. In the last year, two types of state laws have become popular: those mandating AV studies and those allowing for truck platooning. Please note that the source for most the provisions discussed below is the set of links from the National Conference of State Legislatures (NCSL) webpage that houses a list of enacted state laws dealing with autonomous vehicles, and some independent research.

I have no idea how much trucking trade associations and companies are paying for lobbying at the federal level, but they have quietly infiltrated state legislatures and, without fanfare, accomplished the passage of platooning bills in many states.

There are a few exceptions in terms of topics among the 2018 and 2019 statutes, which are explained below.

Another reason for a slowdown and look around among state legislators is the anticipation that the US Congress will act and that it needs to act. There is limited authority to among states to regulate vehicles anyway. What I find most interesting in the passage of recent state laws is the diversity among the "Let's study this" laws as to what is actually being pondered and examined.

Study and report

Maine requires state government agency participation related to aging and people with disabilities,  and participation of a non-profit transit provider.

New York requires that its second annual AV report, in 2019, be from the Commissioner of Motor Vehicles.

Oregon's task force is mandated to include representatives from transit, the taxi industry, and transportation unions, as well as the usual AV, cybersecurity, and insurance industry representation, among others. The study topics go beyond those routinely mentioned in such legislation, with land use, transit, and infrastructure design among the specified long-term topics.

Pennsylvania requires transit participation and either pedestrian or biking participation on its advisory committee.

Washington State has created a work group that is tasked with reporting annually and which is set to expire in 2023. The work group is made up solely of state officials and legislators. The net that the work group is required to cast is broad in that it includes examination of AV social impacts, among other topics, and the task force is legislatively mandated to engage stakeholders and the public.

Washington, District of Columbia (DC) has an impossible legislative search system, so I did a Google search for the name of the legislation. The legislative text (link gives you a Word document) authorizes an expansive AV study, but it does not restrict or discuss who specifically (or their designees) will serve on any committee to research and consider AV laws, regulations, and possible impacts. The District Department of Transportation (DDOT) is required to produce a study that will be made publicly available by July 1, 2019. The DDOT study must consider many of the usual AV study topics, as well as public space and public health, safety of pedestrians and bicyclists, various transportation modes - "including mass transit, shared-use vehicles, and public and private vehicles-for-hire" - and the "impact on the District's disability community."


Platooning

Alabama allows for truck platooning, offers a definition, and authorizes its state Department of Transportation to regulate. Alabama does not appear on the studies list because it passed a "thou shalt study and prepare a report" law in 2016.

Indiana platooning law is not limited to trucks.

Kentucky requires that a proposed plan be submitted to the state Department of Vehicle Regulation, which must approve before platooning is permitted; notification is required to be made to the state police.

Louisiana platooning is not permitted on two-lane roads.

Mississippi does not allow platooning on two-lane roads. Platooning must be also expressly approved by both the state department of transportation and the department of public safety after a "plan for approval of general platoon operations" is submitted.

Oregon does not use the term "platooning," instead calling it "connected automated braking system" and this term conceivably applies to any type of vehicle, not merely commercial vehicles or trucks, that is equipped with the appropriate technology.

Pennsylvania allows for platooning with military, bus, or motor carrier vehicles. Platooning vehicles must bear a visual mark. Platoons are limited to a maximum of three vehicles and each must have a driver on board. There's more, which means that platooning takes a considerable amount of lead time.

Utah has passed a platooning law. It is not limited to any particular type of vehicle. This is in addition to Utah's general AV law discussed below.

Wisconsin passed a simple platooning law. It is not limited to specific classes of vehicles.


Other

California statute allows law enforcement officers to remove an AV from a road if the vehicle does not possess a permit to operate as an AV on public roads within the state.

Another California law allows the City of San Francisco to impose a fee for every AV ridehailing or shared ride provided for a fare.

Nebraska general AV law: Requires that an AV be able to achieve on its own a minimal risk condition, but does not require proof or testing of such capability. Also explicitly allows for ridehailing, other shared-use AV commercial passenger transportation, and public transit. Preempts local regulation or taxes related to AVs.

New York specifies the coordination with the state police required prior to AV testing demonstrations in the Empire State. This is an update to a pretty restrictive AV statute passed in 2017. The 2018 law requires that a "law enforcement interaction plan shall be included as part of the demonstration and test application that includes information for law enforcement and first responders regarding how to interact with such a vehicle in emergency and traffic enforcement situations." The law also calls for a report to be written; see above for details about that.

Pennsylvania allows for automated work zone vehicles as part of its Turnpike Commission's road projects.

Utah's legislature just passed an AV law, awaiting the governor's signature, that:

• Governs and allows for AV ridehailing 
• AV registration requirement
• Fully allows, for level 3 automation, whether with driver on board or a remote driver 
• No license required for AV systems
• Preemption of local government regulation of AVs 
• Low speed vehicles have different rules. Defined as four passengers, including the driver or fallback operation, or less. BUT that human driver is permitted to be a remote operator. These vehicles have a maximum speed permitted of 25 mph.

It should be noted that Utah was an early state that studied AVs. Perhaps the state is a bellwether for others that are or have studied AVs and will then consider AV legislation.

Will We Have an AV Delaware?

Delaware is the First State, meaning the first state to adopt the United States Constitution, and, I'll add, at a time when that document's seminal place in our history was not assured. Delaware is also known as the state of incorporation for many large corporations that have no connection with the state - no factories, no headquarters, and no staff - due to the corporate friendly laws that have created an onshore equivalent of an offshore tax haven. This saves corporations millions of dollars and starves those states where these companies actually operate from collecting the tax revenues that any reasonable person would consider their due.

Think business models

A little context: Right now, we have a transportation system in which most people own the vehicles in which they ride each day. Ownership and vehicle operation are thus located in the same state.

But this may change IF we transition to a shared-use fleet model of transportation, whether it is today's car companies, ridehailing companies, tech companies, rental car companies, transit agencies, or who knows what kind of public or private entities actually own the AVs on our roads. Will there be one national company with a virtual monopoly, a few companies that pretty much divide major markets (similar to air travel), many different kinds of business models, or a public-private mix that leaves transportation-vulnerable riders to poor service of inadequately-subsidized agencies? And what will the role of public sector and non-profit transportation be?

No one knows.

No one knows if the average person will be better or worse off than today in terms of their transportation options. No one knows whether older adults and people with disabilities will be better served than they are today. And almost no one has even thought about what to do about rural areas, where it is unlikely that anyone will make a profit from transit or shared-use transportation or what we can do to avoid leaving rural communities behind when the AV transition happens in big cities and their suburbs.

So what the heck does this have to do with AVs?

I'm writing this post imagining the assumption, albeit an assumption that could prove false, that we will have a passenger transportation system of primarily shared-use AV fleets.

What is to prevent one state - an AV Delaware, if you will - from being THE main state for vehicle registrations?

Presumably, this would be a state with lenient requirements for registration, maybe technology licensing (if that becomes the norm or required), and insurance. We would still have 50 state standards, but, for large companies, there would, for all intents and purposes, there would be that one sweet state. What about a state inspection system with lax standards?

And where would this leave transit? Looking at the history of car companies lobbying Congress and having cushy relationships with key federal agencies, it doesn't take much to imagine transit retaining its lowly, poor service, except in certain cities.

What can prevent this scenario from emerging?



I hate to go to the source, but it would really help to limit campaign contributions - meaning get rid of Citizens United v. Federal Election Commission - which would take a constitutional amendment. Just a monumental feat, but it would limit the influence of big money.

What else could prevent this? Federal leadership. This is certainly doable and Congress is likely, in my opinion, to take the lead at some point. This aspect of AV regulation might not happen, however, until we see AVs en masse on our roads. After all, we did not see vehicle safety regulation at all for decades after automobiles appeared on our streets and roadways. We have a different mindset now, so I do not expect decades to pass.

The question is whether such legislation and regulation would favor the companies providing the transportation or whether it would be oriented toward the actual people using the shared AV fleets and transit services. I don't mean to be a downer, but the latter is not what we in the US have seen when it comes to either accessibility for people with disabilities or transportation equity for low-income communities, whether in urban or in rural areas.

State-by-State Conversation by Report

State-level automated vehicle (AV) task forces and their reports vary a great deal, with some examining scenarios of a wholesale transition of the roadway space or the transportation system, while others appear to presume a static, mostly auto-based transportation system, similar to what we have now in most of the US.

Most of these reports, such as Wisconsin's, serve as AV primers for state legislators, government agencies, and others who work on transportation-related issues within a given state. They explain basic AV concepts, levels of automation, and how AVs differ from connected vehicles. Some explain well the traditional federal-state division of responsibilities over regulation and oversight of vehicle manufacturing, driver licensing, insurance, and infrastructure design. The Utah report does a nice job of describing the federal-state division of responsibilities, as well as which federal concerns are monitored by which federal agencies.

Progressive, inclusive, and geographically correct

These reports reflect the culture of each state, especially in terms of the scope of topics explored. Minnesota's report is a good example. This report emphasized equity and accessibility concerns, even describing the inclusiveness of the meetings that the Governor's advisory council in charge of the report conducted to discuss AV issues with people around the state.

The fear of an inequitable lack of balance between urban and rural areas was evident throughout the Minnesota report and, reflecting the concern with public perception, public engagement was a major priority for the advisory council.

Each meeting included remote participation. In addition to the public meetings, individuals could participate online, by survey, or share their feedback directly with the CAV-X office. To ensure transparency for the process, MnDOT placed all materials on its public website, including dates and times for each meeting. MnDOT conducted additional outreach activities for those unable to attend meetings, including individual meetings and calls, public events, presentations at various conferences and events, and a demonstration at the Minnesota State Fair. MnDOT also participated in intergovernmental consultation with tribal governments through the Advocacy Council on Tribal Transportation, the Minnesota Indian Affairs Council, and individual meetings with tribal executives. The final recommendations considered input from all of these outreach efforts.

Unlike most such bodies, the Minnesota advisory council did not include any AV companies or car manufacturers. Instead the advisory council included staff from government, non-profit organizations, labor, the insurance industry, and the energy sector.

Transportation modes besides cars and trucks? What are those?

On the other end of the spectrum is the report from Idaho, which mostly ignores the possibility of shifting transportation modal choices, and explores the usual concerns of cybersecurity, privacy, licensing and insurance. This might be due to the fact that members of the Autonomous and Connected Vehicle Testing and Deployment Committee - that's a mouthful - were auto dealers, representatives of various vehicle associations, and lots of law enforcement staff. Transit was not represented.

Indeed, though none of the recommendations mention transit, walking, or biking, a subcommittee that included transit, biking, and walking advocates managed to include some interesting text in the report - just not in the executive summary or in the recommendations.

The predictable routes, limited number of vehicles, fixed infrastructure in the public right-of-way, and public oversight place public transit in a unique position to pilot AV/CV technology. By piloting AV/CV for transit, the public could become more familiar and comfortable with the technology. Enabling and deploying AV/CV technology for public transit would also provide operational benefits to transit agencies by providing more consistent operations at potentially lower costs.

...      ...      ...

In the future, most vehicles using Idaho’s transportation infrastructure may not be individually owned, as they are today. The Mobility as a Service (MaaS) model predicts that most vehicles would be owned by corporations or collectives, and dispatched to users on demand. This model already exists with services like UBER and Lyft that currently use human drivers. These services have pilot projects testing the use of AVs. 

AV technology is accelerating faster in urban areas than rural areas. If state policies fail to address the needs of rural and local jurisdictions, the state could develop a disconnected network for AV/CV operations. 

Garden State: Not exactly innovative

Photo from Hyundai. https://www.motor1.com/news/300142/hyundai-elevate-concept-revealed/

The New Jersey joint resolution to establish a task force requires the task force to evaluate existing state and federal law, consider whether to pass AV legislation, whether to allow AVs to operate on public roads, whether to enact AV safety standards. Also looking at the other usual topics of licensing, registration, and liability. This is a completely typical state AV task force. It basically allows a state to appear to be doing something while hedging for time, and it also allows various transportation players in a state to come together, learn, and have an AV-focused conversation that will result in a public report.

*[In New Jersey, a joint resolution is defined as "a formal action adopted by both Houses (of the legislature) and approved by the Governor. A joint resolution has the effect of a law and is often used instead of a bill when the purpose is of a temporary nature, or to establish a commission or express an opinion."] As of Feb. 20, 2019, the Governor has not yet approved of this joint resolution.

While across the pond, if it's required for a 16 year old ...

Meanwhile, Britain surges ahead in a much more organized fashion than we are seeing from the US federal government. The UK automotive minister and - love this title - future of mobility minister have announced that passing a stringent test, using the term "rigorous safety assessment," is required before an AV is permitted to be tested on UK roads.

The UK also has an AV report out of its Law Commission (jointly with the Scottish Law Commission) that provides a much more complete assessment of the legal issues surrounding AV laws and regulations. First, it examines the legal rules in other countries; second, it makes a distinction between partially automated vehicles (such as the Tesla autopilot system) and highly automated vehicles; third, it looks at local versus national control and whether new government agencies should be created or not. The report also raises the issue, in tort liability terms, of what a reasonable AV manufacturer or software updating company would do in terms of an obligation to update sensors, cameras, software, and other technology.