Autonomous driving technology developer Plus and infrastructure owner-operator Transurban recently announced a unique partnership in which Transurban’s intelligent roadway will provide look-ahead data to trucks equipped with Plus’s Automated Vehicle (AV) software, as well as providing proactive support from Transurban’s traffic operations center. Compared to past connected-vehicle plays, there are some new and interesting aspects to their approach that I’ll explore here.
Transurban is an Australian-owned company that builds and operates toll roads. With over 10 million customers globally and 22 roads under operation, Transurban operates assets in Australia’s three largest cities – Sydney, Melbourne and Brisbane – plus the A25 Motorway in Quebec and the 495 Express Lanes, 95 Express Lanes and 395 Express lanes in Virginia near Washington, DC. The company notes that they are “a global leader in sustainability performance, which has been evaluated and recognized by two leading investor sustainability ratings, the Dow Jones Sustainability Index (DJSI) and the Global Real Estate Sustainability Benchmark (GRESB) Infrastructure. Beyond building and operating roads, Transurban has a strong technology focus, researching and developing innovative tolling and transport technology.”
Plus is an SAE Level 4 AV software company that has developed an SAE Level 2++ highly automated driving product (PlusDrive), which is deployed with fleets in the U.S. and being tested in Europe. In parallel, the company is developing an SAE Level 4 driverless truck product (SuperDrive), based in part on real-world data provided by the deployed PlusDrive trucks. Plus is headquartered in Silicon Valley, California and has operations in the U.S. and Germany.
The partnership between Transurban and Plus will leverage Plus autonomous driving software and their global commercial deployment experience. This includes the Plus partnership with IVECO, the global commercial vehicle brand of Iveco Group, which maintains a presence in Australia, to co-develop self-driving trucks for Europe.
In working together, Transurban and Plus aim to create awareness across the broader freight and technology ecosystems, while demonstrating that there is a pathway to commercial operation of AV trucks leveraging smart motorways.
The Evolution Of Smart Infrastructure And Smart Vehicles: A Winding Road
The underlying idea driving this partnership – road/vehicle communications – is far from new. When I got into this sector in 1991 working for the U.S. Federal Highway Administration, the buzzword was Intelligent Vehicle Highway Systems (later to become Intelligent Transportation Systems, or ITS, the term in use today). The vision was to “share the load” of sensing and perception between vehicles and the roadway to avoid crashes and improve traffic efficiency. The end goal was to make cars autonomous. Trucks weren’t even in the discussion back then. What actually happened was that roads got smarter for purposes of traffic management and vehicles got smarter for purposes of crash avoidance. The vehicle-highway linkages were great on paper but didn’t translate to broadly scaled products and deployment at the time.
Then, about ten years later, a new vision of connected vehicles gained traction worldwide: ubiquitous deployment of communications transponders in cars to continuously share vehicle data relevant to safety and efficiency, while transponders along highways and streets provided additional data, such as traffic signal status. In the U.S. this was given momentum by a U.S. Federal Communications Commission spectrum allocation for this purpose. The idea of low-latency communications to avoid a fast-developing crash situation was powerful – if we could only get all the road actors equipped with standardized comms.
The mass deployment of these transponders hasn’t happened yet, even though the technology plays a role in niche applications. Meanwhile, with multiple generations of cellular system upgrades since the 1990’s, information can now travel quite fast through the ubiquitous cloud. Not fast enough for crash avoidance, but adequate for important safety and efficiency related information to be disseminated.
In the late 1990’s, another trend came onto the scene which has proved to be vitally important. The rise of sensor-laden vehicles (Advanced Driver Assistance Systems) over the last 25 years has enabled highly effective crash avoidance capability on millions and millions of vehicles on the road today. Thus, the need to rely on super-fast comms to avoid crashes has faded – which is great, since an effort to get all vehicles equipped (a classic chicken-and-egg conundrum) would take decades.
Thus, while “instantaneous data exchange across all road actors all the time” is not needed thanks to on-board sensors, there is still value in having detailed, fresh, and reliable knowledge of what’s happening on the road and in the upcoming traffic stream. This is especially true for a self-driving vehicle: it provides an additional layer of situational awareness beyond the sensing capability of the on-board systems.
(I’ve over-simplified this discussion of road-vehicle communications to make some basic points relevant to this article. There’s a lot more going on in this space!)
Autonomous Vehicles Do Not Need Infrastructure Support to Operate Safely
Autonomous vehicles being developed and deployed today are fully capable of fulfilling their safety case with their own sensing and computing capability. Within their operational design domain, they are designed to operate safely. This is a fundamental premise of bringing autonomous driving to the people- and freight-carrying industry.
That said, infrastructure-sourced information still has a valuable role to play.
Extending The Information Horizon Makes A Big Difference
This Plus-Transurban partnership builds on the results of Transurban’s first self-driving truck trial on CityLink and the Monash Freeway in Melbourne last year, marking the first time in Australia that a highly automated self-driving truck was tested in live traffic conditions on public motorways.
With road freight projected to grow steadily in the coming decades, the partners see self-driving trucks having the potential to transform the freight industry by moving more goods, more quickly, and more sustainably, while helping to address the shortage of truck drivers around the world. They emphasize that this will also deliver positive economic and social outcomes for consumers, motorists, and governments by reducing congestion and improving road safety.
This new initiative aims to demonstrate that Level 4 autonomous trucks using data from smart road systems will further enhance traffic flow and operational efficiency. The companies see this partnership as a model to more quickly enable scaling of autonomous trucking for operation on complex freeways in and around major cities.
As the road operator, Transurban has a comprehensive understanding on what is happening on each lane for every inch of their motorways. While this may also be true for some sections of roads in Australia and North America built and managed by the public sector, the data intensity is at another level for a privately financed and operated motorway.
Sharing information with an autonomous truck serves to extend the vehicle’s “information horizon.” This offers two key benefits – improved traffic flow and further gains in fuel efficiency. On-board sensors collect the basic info key to safe operation, but having real-time information beyond the sensor view can, for instance, enable the truck to gracefully slow down or change lanes and then cruise at optimal speed due to a traffic disturbance ahead. With on-board sensors alone, the vehicle would not react until the sensors are within range to detect the disturbance.
The operational benefits that will result from a long look-ahead beyond the range of what sensors on the autonomous trucks can detect will result in much smoother driving, which translates to fuel savings, as well as improved traffic flow for all road users.
Transurban emphasizes yet another benefit that can be provided to autonomous vehicles by infrastructure owners. Sophisticated control rooms, which manage the motorways in real time, can provide additional support for autonomous vehicles by detecting and responding quickly to incidents and even closing lanes if needed.
Consider what this means to initial deployment of autonomous trucks. Given the density of passenger cars during the daytime, AV trucks might predominantly run at night. In its first self-driving truck trial last year, Transurban’s managed motorway technology created a “rolling dedicated lane” for the trucks i.e. overhead signage indicating the lane is closed to regular traffic. The system can create a buffer zone ahead and behind the truck, which moves dynamically with the truck. Yet another way to improve traffic flow and add a layer of risk reduction.
A Transurban spokesperson noted that “autonomous vehicles are only one side of the coin – they will benefit from smart infrastructure to travel on. As the operator of some of the most technologically sophisticated roads in the world, it’s a natural fit for Transurban to take the lead on testing and preparing for a future with autonomous transport technology. Crucially, industry needs to demonstrate the benefits of autonomous road transport to governments and the community to drive this innovation forward.”
Cavnue’s Play in Michigan and Beyond
Another AV-infrastructure play is continuing its development here in the U.S. Since last year, Cavnue has made headlines with their vision to equip highways with physical and digital infrastructure to support autonomous vehicles. They partnered with Michigan DOT last year to create “the country’s first corridor of tech-enabled infrastructure for connected and autonomous vehicles.” This is being implemented initially on a 25-mile segment of Interstate 94 between Ann Arbor and Detroit. The Cavnue website depicts newly constructed dedicated lanes for highly automated vehicles.
A 2022 Michigan law authorized MDOT to collaborate with technology partners to modernize roads for automated driving.
Silvio Memme, Cavnue’s Director for Automotive and AV Partnerships, told me they are making small civil modifications to the roadway and equipping this section of I-94 with sensor pods placed every ~200 meters. “We leverage proprietary machine learning algorithms to generate insights that are shared directly with the connected and automated vehicles on the road to create a more predictable and reliable driving experience by extending vehicle planning horizon,” he added. The aim is to share insights with automotive OEMs and automated trucking developers, MDOT, and road operators to improve safety and operations along the corridor.
Cavnue believes this corridor will be replicated further in Michigan and across the rest of the country, particularly in the Southeast as the autonomous trucking industry continues to accelerate in these states.
The company’s CEO and Co-Founder, Tyler Duvall, said, “At Cavnue, we believe strongly that an integrated approach to deploying hardware and software solutions is the key to developing truly smart roads that enable the automated, safe, and efficient movement of people and goods. To be effective, smart roads must deliver a complete view of the operating environment, delivering real-time and predictive insights that can inform key decisions and interventions by all users.”
Road-Vehicle Cooperation Coming Next Year
Plus and Transurban tell me that the project is ramping up quickly. During the first half of next year, planning is underway to be running two pre-commercial autonomous trucks on a freight route up to five nights a week. Transurban also noted that they are actively engaging with governments, communities, and freight stakeholders to advance the project and identify the challenges required to be solved in bringing forward autonomous trucking.
Driverless trucks on a lightly traveled highway in the boonies is one thing. Autonomous trucks on metropolitan freeways will get more scrutiny. As commercial deployment of autonomous trucks starts, cities and the public could see more of these types of smart-road and smart-vehicle measures Plus and Transurban plan to implement. Data from this program that shows how they make good on the promises of autonomous trucks – improved safety, enhanced fuel savings and reduced traffic congestion among them – will be play an important role to get key stakeholders on board.
CAVNUE’s approach as a startup contrasts with Transurban’s heft as an owner and operator of a significant number of roadways. Transurban can take the long view and bring significant resources to the table to scale globally. CAVNUE can bring an agile approach and leverage investment funding.
Both will energize the road operator sector to consider what the future of roads will look like in an autonomous driving world.
Disclosure: I advise and/or hold equity in companies named in this article, including Plus and Transurban.
Read the full article here
