Skip to content

Smart infrastructure to provide extra support for autonomous vehicles

Roadside units that leverage 5G can create 3D models of the world around them, and transfer that data to nearby cars in a flash. By Freddie Holmes

There is much debate as to whether 5G is a vital jump in connectivity for future vehicles, or simply a means for automakers to spice up their offering. However, while existing 4G connectivity can make the driving experience more enjoyable, the introduction of 5G—and the services it facilitates—will make the journey safer.

5G is essentially the next step up from 3G and 4G, and allows more data to be processed, and faster. So much faster, in fact, that it has enabled cars to be raced around a track remotely, and is being considered by the healthcare industry to replace physical patient consultations with real-time video conferencing.

Roadside infrastructure will also be outfitted with sensors and connectivity

In the automotive space, 5G will be an enabler of cellular vehicle-to-everything (C-V2X) communication. C-V2X is gradually approaching commercial readiness, and the 3rd Generation Partnership Project (3GPP)—a global initiative that develops standards for mobile connectivity—is closely investigating how the technology can address next generation vehicle use cases. Release 14—better known as ‘Rel-14’— will mark the initial phase of C-V2X, and according to Shailesh Patil, Principal Engineer/Manager at Qualcomm, there is broad industry support for the standard. China should be the first market to see commercial deployment, with the US following in a year or so later.

Much focus is placed on the role of 5G and C-V2X in the vehicle itself, but there are other applications that also show significant promise. In order to achieve ambitions for smart cities, connectivity and sensing capabilities will need to be embedded within the infrastructure itself. So-called smart roadside units (RSUs) could provide wireless communication between vehicles and their surroundings.

Within an RSU will be a communications device, a high level of compute power and a software stack. As Patil explained in a recent Automotive World webinar, this could prove a real boost for road safety in particular. “The RSU will be able to detect humans that are about to cross the road, and advise approaching cars to be aware of that,” he advised.

With connected infrastructure, vehicles have an extra data source to work with

Further out, there is even talk of smart infrastructure being able to create roadside models through sensor fusion. Just as autonomous vehicles are being equipped with various sensors to create 3D models of the world around them, there is no reason why infrastructure cannot perform the same task. That information could then be made available to nearby vehicles—all thanks to 5G and C-V2X. “The idea is that roadside units will be equipped with all these different sensors to generate a view of their environment. This can then be communicated to the vehicle,” he explained. “This way, an autonomous vehicle can obtain an accurate view of the world not only through its on-board sensors.”

In essence, this approach would provide an extra layer of redundancy. While autonomous vehicles may never be able to rely solely on the information provided by smart RSUs, the real-time data they offer is a welcome bonus. If the vehicle is unable to identify whether a human is crossing the road, for example, that RSU can offer an extra point of view to confirm either way.

In some ways, the idea follows a similar concept proposed for smart parking infrastructure. In February 2018, Bosch demonstrated how metal bollards could be outfitted with LiDAR sensors in order to guide vehicles into a parking spot as part of a ‘driverless valet’ service. Speaking at the event in Berlin, Bosch Chief Executive Volkmar Denner explained that it makes sense to “transfer the intelligence” into infrastructure, and not just the vehicle itself.

Welcome back , to continue browsing the site, please click here