Transportation

Embark And Texas DPS Partner To Develop Best Practices For Officers To Pull Over Automated Trucks


Truck automation developer Embark has taken a strong step into a space that every truck ADS developer must navigate: how to detect, respond to, and communicate with law enforcement while traveling public roads.

Founded in San Francisco in 2016, Embark was one of the earliest players to tackle the automation of heavy trucks. Now public, Embark’s stated go-to-market strategy is to build a manufacturer-agnostic Automated Driving System (ADS) that can be used in multiple truck brands. Carriers and shippers would pick the trucks brands they prefer for their fleets and would own and operate the vehicles, using Embark’s technology for driving. Embark is aiming to have a network of 100 highway-adjacent transfer hubs operating over the next five years, with their initial freight markets in the southern regions of the U.S., i.e. the Sunbelt.

Stopping For A Cop

When it comes to dealing with “official” vehicles, law enforcement vehicles are the key. When they want you to stop, you stop. When they ask questions, you answer. This contrasts with ambulances and fire equipment who just want you to get out of their way.

The first industry-wide document on this topic was published by the Automated Vehicles Safety Consortium in 2020. Titled “Best Practice for First Responder Interactions with Fleet-Managed ADS Driverless Vehicles,” it provides “a framework of recommended procedures and protocols system developers can follow to facilitate first responder interactions in multiple use cases.”

The AVSC Best Practice is generic to all types of automated vehicles. The automated truck world is distinct in many ways. Most prominently, when a truck gets pulled over officers can demand detailed information for both the driver and the load.

For years there has been talk among truck ADS players to standardize this process across all automated trucks. This is a noble goal but given the many layers of law enforcement (federal, state, local, turnpike) who have their own ways of doing things, many saw such a process as taking years and slowing deployment.

Embark’s view is to make the interface between the AV truck and an officer on the roadside so simple and intuitive that it is learn-able on the spot. The implication is that other ADS developers can do the same.

An Embark spokesperson told me that their approach “is based on a broad understanding of law enforcement requirements sourced from direct conversations with multiple law enforcement and regulatory agencies, as well as review of industry best practices, such as those published by AVSC on first responder interactions.”

As Embark puts it, their trucks will “identify, stop for, and interface with law enforcement vehicles.” They are working closely with the Texas Department of Public Safety (“Texas DPS”) to train Embark’s automated trucks to “identify law enforcement vehicles in situations such as traffic stops, and to develop communication protocols and standard operating procedures between autonomous trucks and law enforcement officers.”

Step By Step

Doing this job right requires at least three steps.

First, detection: the Embark automated truck must identify emergency vehicle lights and other cues to slow down. The trucks must discern the difference between blue lights on the opposing lane or on a parallel road. They must understand which blue lights nearby apply to them.

Second, when it is determined “they’re it,” a safe maneuver must be implemented to pull over expeditiously at a proper location on the highway shoulder.

Third, once both vehicles are stopped on the side of the road, the robot and human must communicate somehow. Embark says law enforcement must “receive this information from an autonomous truck intuitively and without any additional equipment” on the law enforcement side. As the law enforcement officer approaches the truck, he or she must be confident that the vehicle has come to a safe stop with no risk of restarting unexpectedly. Embark says this “may include outfitting Embark trucks with clear visual cues and information to signal to law enforcement personnel that an Embark-powered truck is an autonomous vehicle.” Embark also plans to outfit trucks with a lockbox accessible to law enforcement containing vehicle and load information such as registration and bills of lading, as well as contact information so that law enforcement officers can reach an Embark Guardian operator to verify documentation (presumably by phone).

“Making sure law enforcement can safely and intuitively interact with autonomous trucks is a ‘must’ for deployment,” said Emily Warren, Head of Public Policy at Embark. “Our work with Texas DPS prioritizes safety as we achieve this key technical milestone, and enables us to create a scalable emergency vehicle interaction model that can work across Texas and the U.S. Sunbelt.”

These types of roadside interactions are one piece of a larger puzzle. Truck ADS developers are working with the state law enforcement officials to define the optimum approach to truck inspection as well.

Being A Solid Road Citizen

Embark’s activities announced here are the eleventh and latest accomplishment in its 16-step technical capabilities roadmap. Embark says they plan to publicly demonstrate this law enforcement interaction capability later this summer.

Other truck ADS developers are active in this space as well. Aurora has discussed their remote support tool that allows remote support specialists to provide high-level guidance when their automated vehicles encounter a situation where the appropriate actions are unclear, such as interactions with law enforcement officials. And, as one would expect, Texas DPS is in discussions with other truck ADS developers as well.

There was a time when “hey the truck can drive itself!” was big news. No longer. While the core automated driving functions continue to mature, AV developers are intensely focusing on every aspect of how their vehicles should be interacting in the environment.



READ NEWS SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.