Engineering Volvo’s Next Big Leap

April 1, 2019

Volvo Cars has set one of the industry’s most ambitious engineering goals: to have zero traffic fatalities in any of its new models beginning in 2020. It also aims to have one-third of its new-vehicle range be capable of SAE Level 4 autonomous operation by 2025. First to arrive with Level 4 capability will be the next-generation XC90 due in 2021.

Henrik Green (below), the senior VP of research & development, is pleased with his company’s progress in both the electrification and automated-driving areas. As a computer-engineering graduate, Green joined Volvo in 1996. He’s since led engine development, powertrain-controls, and the 40-series vehicle program, and has headed global product strategy. He spoke recently with editor-in-chief Lindsay Brooke on Volvo’s technology trajectory.

Some OEMs say LiDAR is necessary while others say it’s not. Does Volvo’s relationship with Luminar validate LiDAR’s future with your company?

Yes, it does. We’re coming from the safety perspective so we pay close attention to the algorithms that will ultimately control an autonomous vehicle. In that sense we need to understand the objects around us — what they are and what they’re going to do in order to predict their path. And we need that to determine our path forward and to avoid any false positives. So, need super perception. And we believe long-range, very high-resolution LiDAR is the key to that.

When do you expect LiDAR sensors to be commercially available with a form factor that Volvo’s vehicle designers will love for their compact size?

Our next-generation ‘SPA-2’ [scalable product architecture] vehicles that we’re launching at the beginning of the next decade are being engineered from scratch with the necessary electronics and software to have full autonomy. So, the form factor of the sensors is a key thing here. It’s a challenge because we want to be able to access the absolutely-latest technology. At the same time, we need to make the car look amazing in the consumer’s eyes. It’s a matter of physically integrating sensor hardware into the vehicle so it looks great—but doing it as late as possible in the vehicle-development stage.

With up to 80 companies now claiming to be ‘LiDAR suppliers,’ how do you as Volvo R&D boss sort them out?

It boils down to three things: Range, resolution at maximum range, and scalability. With that, I think we have a really good partnership going with Luminar.

Do you see unit costs for LiDAR coming down?

In general, I would say ‘yes’ to that for all the parts of the technology. We will have to deal with the cost as we get close to launch, but at that point we will always prioritize safety. We want a capable system — and that will not come cheaply in the beginning. But over time the cost will come down from both the technology evolution and from scale.

How are you managing thermal challenges in development?

The energy consumption of these vehicles — for the compute platform and sensor platform — is significant. Power supply and thermal management are among the issues we’re dealing with. These have been less important in past vehicles.

Electrified vehicle platforms offer sufficient power for this. Absolutely. The programs we’re running today, current initiatives, use our plug-in hybrid platform. It’s good to have a plug-in battery of that size because we need the power to run the processor. But when we launch the next generation SPA-2 we will have full EV capabilities and a great deal of energy capacity on board.

How difficult has it been for Volvo to transition from ICE to pure-EV platforms?

The scaling challenge is, I think, interesting. Anybody can build one good EV, and now nearly everybody is doing that. We’ll see a number of new EVs coming out within a few years and we’ll be among them. Our SPA-2 architecture is engineered for EVs from the ground up. We’ll be able to scale EVs across our vehicle ‘top hats’ as market demand grows.