The movement of goods by autonomous vehicle systems is influenced by many factors: from public policy and law enforcement, to staffing the autonomous vehicle fleet of the future, and then outfitting it with an effective technology stack that affords safety. 

The day shaped out with an exhaustive exploration of these and many other topics important in promoting the automated movement of goods in the world around us.
 

Law Enforcement Interaction Plans Need to be Consistent and Collaborative

The Automated Goods Movement Summit kicked off with a panel discussion of "What is a Law Enforcement Interaction Plan? The panel featured one active highway patrol officer (Ohio), and two retired (one from California and one from Washington state) who are now working with organizations engaged in setting law enforcement policy. The overall impetus of the panel discussion centered on the need for robust law enforcement interaction plans (LEIP) to be used as guiding documents for state law enforcement personnel. These plans can then be used as guidance towards establishing policy when encountering a variety of issues associated with the emergence and adoption of autonomous vehicles - in all configurations - as they deploy on the nation's highways and byways. 
 
Plans and policy to implement autonomous vehicles and driving must be consistent and accessible.
Lieutenant Chris Kinn of the Ohio Highway Patrol emphasized the need for consistency: variation of policy, laws, and procedures from one state to another could be awkward and difficult for law enforcement. Lt. Kinn likened it to a scenario where one state requires motorcycle helmets (Michigan) and another does not (Ohio); cross state lines and the laws change. 
 

Some states have established policies and plans, but more than half of all states are not yet "driver ready."  

Retired highway patrolman Brett Fabbri (California) now working for Kodiak Robotics and part of a collaborative task force which includes the American Association of Motor Vehicle Administrators (AAMVA), spoke to the need for collaboration, given the risk of inconsistency, and the differing progress of policy across states. Fabbri showed on a map the status of policy and plans for autonomous vehicles, which indicated that about 20 states are "driverless ready," while another 20 states are in "testing phase.” 
 

Safety is a watchword for the AV industry.  

Autonomous vehicles rely on extremely rigorous standards of safety. Brett Fabbri pointed out that most vehicle accidents are a result of human error; autonomous vehicles are absent of human error. Citing statistics from his home state of California, Golden State residents suffer about 4,000 fatalities per annum due to human error. If autonomous vehicles only reduced casualties from human error, that would result in saving 200 lives annually.  
 

An Agreeable Argument: Aligning Around Safety Cases

In a later panel discussion around automated goods movement, the topic centered on safety and the challenge of assessing risk. Specifically, what constitutes some of the give and take when it comes to safety, a self-driving vehicle, and using a safety case framework to be sure it is safe enough to drive on public roads? 
 
Among the panelists was Nat Beuse of Aurora, whose company is known for their safety case framework to evaluate when vehicles are safe enough to operate on public roads. For Aurora, a safety case framework combines "guidance from government organizations, best practices from safety-critical industries, voluntary industry standards and consortia, academic research, and what an organization has learned in its own work."
 
The panel included representatives from insurance brokers (Insurance Office of America - IOA) and Edge Case Research (a risk-management partner for companies building self-driving vehicles). 
 

When it comes to self-driving vehicles, safety can be complicated. 

Insurers have a legacy of data and experience in insuring human-driven vehicles, however underwriting and risk assessment for autonomous vehicles is new. At present, risk relies on a safety case framework.  It's expected that such a framework will take on new and improved shape over time as more data becomes available and applied to drive meaningful underwriting decisions. 
 
As technology, data volumes, intelligence, and analytics grow, the accuracy of risk assessment will improve and subsequently insurance of autonomous vehicles will move forward.  Such momentum will come as manufacturers improve controls and install additional telemetry that will increase in complexity. In addition, more data will become available, and actuaries will decide which data is meaningful and indicative of risk. In time, a safety case and its accompanying framework will continuously improve and be shaped by new things: technology, systems, engineering, statistics, data, and how well it all helps explain safety to all stakeholders – commercial operators, lawmakers, citizens, underwriters, and technology developers.
 

Building the AV Workforce of Tomorrow

For autonomous vehicles to progress, a right-sized, right-skilled workforce must move and grow in tandem with the industry it supports. A panel discussion ensued with representation from education and vocational schools, as well as industry technology vendors and non-profit providers who promote education for the autonomous workforce of tomorrow.
 
Curricula are slowly, but surely, developing in community colleges and becoming part of workforce development. Workforce development officials at the community college level are working with industry in Arizona at Pima Community College to develop a curriculum to train and fill positions in the AV workforce. Specifically, the Arizona-based community college offers an autonomous vehicle operator certification program it developed with their partner TuSimple.  The community college also offers a class, “Autonomous Vehicles – 101,” to train and educate professionals in AV-friendly Arizona. 
 
The AV workforce will require many skillsets – including an influx of skilled technicians. Mike Pressendo of TechForce Foundation, a nonprofit championing career development for professional technicians, noted that while there is a pronounced shortage of truck drivers, there’s also a shortage of technicians to support trucks and now autonomous vehicles. The gap between supply and demand of skilled technicians is already widening with about five openings for each graduate of a vocational program.  The gap is only expected to grow wider as AV adoption and technology expands.
 
Discussion mentioned the underrepresentation of women in the industry on all fronts (a mere 2% of technicians are women), however that trend could be turning as more women and minorities are added to the payrolls of the overall AV workforce. 
 

Commercializing Automated Freight

Later in the day a panel explored technology and its role in commercializing automated freight. The panel featured technology vendors including those who provide lidar (which stands for Light Detection and Ranging - a remote sensing method that uses light in the form of a pulsed laser to measure variable distances to the Earth) and radar (a legacy technology - which stands for "radio detection and ranging" - that uses radio waves to determine and detect distance), a level 4 self-driving truck company, and a nine-year-old trucking firm in North Texas who hauls for the oil and gas industry. The panel was a diverse representation providing interesting perspectives on technology in the context of moving freight via autonomous vehicle systems.
 
The discussion first focused on the “AV stack” or the types of technology used, now and in the future, for autonomous vehicles with an accent on truck transport. The role of lidar and radar were discussed, with mention of the importance of redundancies in design to produce reliability and safety in AV operations. 
 
A recurrent metaphor was that an AV stack is akin to swiss cheese: a single slice has holes in it. Yet, a stack has overlap which effectively fills the voids. So is the role of redundancy and integration of technologies in the AV technology stack.
 
Michael Clements of PDQ Trucking, whose Texas trucking firm has a specific niche, emphasized the needs of businesses – especially the small- to medium-sized entities – who may or may not be able to invest in technology, depending on their resources. He also emphasized the importance of determining which operations are most suitable for automated trucks.
 
Technology is one among the many issues discussed over the course of the day. As autonomous vehicle technology grows and expands, so will these and many other topics be explored with the same vigor they were approached with on day one of the Automated Goods Movement Summit.