At the 2017 Frankfurt IAA just-auto’s Calum MacRae met with Guillaume Devauchelle, vice president of innovation and scientific development at Valeo. During the show Valeo unveiled that its future strategic direction would be determined by a focus on three industry megatrends: electric drive; autonomous drive and mobility. 

In all three aspects Valeo is a company playing to its strengths, but this seems to be particularly the case with autonomous driving where Valeo can deliver nearly all of the required systems. As Devauchelle observes, “I will not say 100%, but we can deliver nearly all of the required systems. With SCALA Lidar we are first in the market with Audi. Additionally, we do cameras, radar and ultrasonic sensors. Already we deliver 700,000 ultrasonic sensors a day so it’s a huge business for us.” 

Valeo also has high hopes for its sensor fusion units, the box of tricks that aggregates all the inputs from the various sensors, “They manage the triple redundancy we believe you need for autonomous driving. On board we’ll know exactly what is happening even in bright sunlight or if it’s foggy and the camera doesn’t see that much. So we can always guarantee that the image we give to the OEM will be accurate. And we can also provide the PCU that decides what the car should be doing depending on the inputs.”  

Autonomous provides wide-ranging possibilities for Valeo’s business portfolio of comfort & driving systems, powertrain, thermal and visibility. In the latter area Devauchelle observes there’s even scope for that most prosaic of components, the windscreen wiper, to play a role in autonomous driving. Additionally, lighting, also within Valeo’s vision portfolio, has a significant role in autonomous driving. “We need a different lighting system for autonomous cars because just imagine if you are pedestrian and you have to cross the road. At the moment we, as pedestrians, make eye contact with the car driver to determine whether it’s safe to cross the road or not. But with autonomous the car will have to express itself to give the pedestrian the all clear, and that’s where our lighting expertise will have a role to play.”

One of the main debates in autonomous driving is how many layers of sensing and cameras are going to be required to deliver Level 3, 4 and 5 autonomy. Here Valeo remains unequivocal in its belief that triple redundancy is the way forward. As Devauchelle says, “Cameras. Lidar. Radar. We don’t believe that the stereo camera is a good solution because both cameras rely on the same physics – so if one camera is struggling the other will struggle too. Lidar does not operate in the same bandwidth so it adds more security.” What of Mobileye’s view that its monocular cameras are sufficient?  “We have an agreement with Mobileye. So our cameras include some Mobileye hardware and software. We do consider for the time being that we need Lidar on top of that. Camera is very good for images but it’s not so good at computing distances.” 

It’s unsurprising that Valeo sees Lidar as a necessity for Level 3-5 autonomous driving as it’s just in the stages of developing its third generation – concurrently with its second generation – Lidar system. Its first generation Lidar debuted on the Audi A8 earlier in 2017 and is manufactured at Valeo’s Wemding plant near Munich. The second generation system will be a rotating mirror arrangement, but with more than the four layers of the first generation. So does the much-vaunted solid state Lidar exist in Valeo’s development plans? Yes, and it’s the third generation system being worked on concurrently with the second generation Lidar. They were first to market with their first generation Lidar so it’s very probable they’ll want to be first to market with solid-state too. What about solid-state being the one way to make Lidar affordable and below US$1,000 per unit? According to Devauchelle, “Our first generation Lidar is not US$1,000 – it’s much less, much less.” 

For its part Valeo is certain that there’s a consumer driven market for autonomous vehicles. In some quarters there are dissenting voices that it’s an answer to a question that nobody has asked and that it, along with the new mobility eco-system, is the OEMs’ response to the fear that in the future people won’t buy cars in the same volumes. As Devauchelle states, “We see two trends. The first one, and we have checked this in many countries, is that every driver expects some degree of automatic – not autonomous – driving and this is especially true of automatic parking manoeuvres. The global need is also true for managing traffic jam situations. However, for highway it’s not everywhere. In Germany, one dreams to drive oneself at 200kmh for the thrill of speed. But in the US highways are straighter, longer and more monotonous. Here autonomous would be a real boon.” For Valeo the second key determinant is increasing congestion in cities and urban pollution. Here Valeo sees the city authorities as the main players, “One of the solutions is to offer autonomous shuttles or robotaxis to cut the cost of semi-public, semi-private transportation like taxis or Uber. This is a main driver for autonomous driving because those in charge of our cities are pushing for these solutions to achieve a big reduction in pollution and accidents at a low cost.”