The interior of the Volvo 360c concept functions like a first-class private cabin on wheels. The challenge is to make it safe. (Volvo)

Volvo safety guru Lotta Jakobsson on AV crash tech

One of Volvo’s top safety engineers discusses how the company’s safety strategies will be applied to increasingly automated vehicles.

It’s been 60 years since Nils Bohlin, the legendary Volvo safety engineer, invented the three-point safety belt, which remains the single most important safety feature of today’s vehicles. “We haven’t managed to figure out something superior to that,” said Lotta Jakobsson, a safety engineer who this year celebrated her 30th year at Volvo. Jakobsson hadn’t yet been born in 1959 when Volvo introduced the three-point safety belt, but the Ph.D. in biomechanics has contributed to an impressive list of safety features.

Her handiwork is evident in Volvo’s Side Impact Protection System (SIPS) and its inflatable curtain that prevents occupants from striking their heads on the inside of the car during a collision, along with the innovative seats used in Volvo’s scalable architecture vehicles since MY2015. For those seats, metal webbing in strategic locations works like tiny crumple zones, absorbing forces that would otherwise impact the human spine and other bodily structures. We spoke with Jakobsson in Banff, Canada, during a mid-2019 test drive of Volvo’s updated 2020 models, to discuss how Volvo’s safety strategies will be applied to increasingly automated vehicles.

Automotive Engineering: You say that we’re in the middle of a paradigm shift in safety. Please explain.
Lotta Jakobsson: Safety today is very much based on standardized testing and standardized crash-test dummies. Now we need to create new knowledge beyond those crash-test dummies. We need to understand a lot more about human properties when interacting with seatbelts, not just while sitting in an upright posture.

AE: Do you get concerned when people talk about ADAS and automation as a panacea for safety?
LJ: Many people hope that these technologies will solve everything. I don’t think so. We don’t rely on crashes suddenly disappearing. We believe that future cars will still be exposed to crashes. AVs will crash because they’re in mixed traffic and somebody can run into them. So we better focus our work on crashworthiness. That’s our mindset.

AE: Will we remain belted down in an autonomous car?
LJ: I don’t know the technical solutions. But I know for sure that we need to interact with the strong parts of the body. That’s what the safety belt does. The pelvis and hip bone are very strong. And across the shoulder is also very strong. You don’t want to be strapped over the belly or neck.

AE: What about sensor technology to make self-driving cars safer?
LJ: I have colleagues who are experts in that. But I’m an expert in how bones are broken. Forward collision warning and auto-brake systems will assist the driver. But when you can’t altogether avoid a collision, the key questions are about biomechanics.

AE: What role do external sensors and automatic braking play?
LJ: Automatic braking is part of it. If the speed isn’t too high, you can avoid the crash. That’s brilliant. But in many cases, the speed might be higher than what is possible to avoid a crash, so you have to mitigate the collision. It makes a huge difference for protection to crash at a lower speed.

AE: People expect self-driving cars to be much safer. But you’re saying that they’ll create new safety challenges?
LJ: The challenges are not new. It’s just that we need new tools and methods. The whole way of addressing the protection of people in cars needs to be on a deeper level of understanding of injury mechanisms. Passengers today are not that much different from the future drivers in an autonomous car except if you start using a rotating seat or change the seating position. Already today, we are learning a lot about how different passengers behave in the car or sit in ways they’re maybe not supposed to. In research studies, we expose people to controlled evasive braking and steering-maneuvers instruments. And we instrument them with stick-on EMG sensors to measure muscle activation, which provides input to the development of the virtual active human-body models.

AE: Can other internal sensors and passenger-facing cameras also help?
LJ: We don’t start with sensors and ask, “What can we do with this one?” It’s the other way around. We go deep into the real-world data and then develop a methodology that we use in the laboratory to evaluate solutions. It’s probably not the most exciting test, but for our run-off-road solution, we simply dropped the seat on a damper to get the right characteristics. We then we tune the seat for the characteristics we want in the seat.

AE: How will you respond to new form factors in vehicle interiors?
LJ: Just reclining the seat already poses a huge challenge to the lap belt, because the lap belt today grabs the edgy points of your hip bone. With a conventional seatbelt design, you will slide under the belt. So we have to adjust that conventional design of the seat belt to something that will grasp that hip bone or stop you another way. Another thing is we have dramatically reduced injuries compared to years ago. That means crashes are more unique.

AE: More edge cases?
LJ: Yes, almost all of them are edge case nowadays because the big steps have been taken. Also, our acceptance of injuries will be very low in an assisted-driving car. You might accept certain things from a human driver but have other expectations if a machine is driving. So we need to work in greater depth. People are very different sizes and shapes. Crashes can occur from all directions. And if you take all that together, there is an extremely high number of combinations. One part of the learning is to see the synergies between them so you can reduce the number.

AE: How does your safety blanket concept work?
LJ: It’s a very early idea based on the Volvo 360C, a concept car (top) that in the future you might take during the night instead of an airplane. You can have a good meal sitting upright working, and then you lay flat and sleep. To make this possible to launch, we need to make sure that we can protect passengers when lying completely flat. So you could get a nice blanket for comfort, but it would be activated in a crash.

AE: What would happen in a crash event?
LJ: The blanket might have two built-in two transversal straps. They’re quite loose when you are sleeping because you want to sleep in different positions. But when there’s a crash, the sensors will locate these straps over the strong parts of the body, which is over the upper part of the chest and the hip bones.

AE: So the blanket is for the most part just a blanket, but the straps do the work?
LJ: Yes. It’s the same mindset as seat belts. First, crashes will occur. So we need to consider the safety of occupants. Second, we integrate the protection in something that you use for convenience, for some desirable purpose other than safety. And third, we interact with the strong parts of the body. The blanket is the idea we have at the moment. But we might figure out something else. But the same principles are still valid.

Continue reading »