ADAS Revenues Pave the Way to a Driverless Future

While the near-term prospects for Level 4 and above driverless technology seem to be receding, automakers are cashing in on the revenue streams offered by current ADAS technology.

That’s the view of Ziv Binyamini, CEO of the simulated testing specialist Foretellix. He suggests this shows a healthy approach that will pave the way to full commercially viable autonomous driving technology in the future, albeit one further off that some Silicon Valley zealots have predicted.

Speaking to TU-Automotive, Binyamini admits the driverless dreams may have taken a knock but they have not gone up in smoke. He said: “In Level 4 robo-taxi autonomy there is definitely reduced investment. However, some of the big well-funded companies, like Waymo and Cruise, are not slowing down whatsoever which means, in the end, they will probably have an advantage. In the other areas, I see no slow down in investment and of these the biggest interest is in highway advanced ADAS.”

Binyamini believes this return by automakers to what they do best is a positive move. “This is core business and we are talking with many OEMs on this. Any OEM that will slow down in their advanced ADAS development will lose business because it’s already a current technology today and not a futuristic thing. It already brings in revenue and there is, and will be, a fierce competition in this area. We see several companies shifting funding from development of Level 4 into ADAS technology for use today.”

He suggests, however, that automotive industry is guilty of an amount of naivety in how it approaches the testing of autonomous driving technology. “Today the metrics the industry is using is the number of miles traveled in the real world,” said Binyamini . “However, we don’t know what is being tested – what if they are just going in circles, so to speak? Have they actually tested all the 10 million or hundred million scenario variations that they should be testing? We are not being told. All we hear is the number or miles tested when what we need to measure is the coverage of all the possible scenarios and not the naive and meaningless metrics of counting miles.”

He suggests much more work has to be completed in the virtual world to ensure automated technologies are robust enough to cope with a vast range of potential situations it will face in the real world. He points to the recent recall by Volvo which filed documents with the National Highway Traffic Safety Administration saying it will recall 121,605 vehicles for possible automatic emergency brake (AEB) failure.

Binyamini said: “This is, supposedly, a very simple function, yet, in a certain situation, it failed to work. The failure was put down to some unique temperature situation where the AEB did not kick in at the right time. So, they have had to recall many 2019 and 2020 models because it’s very hard to test for this exact situation in any physical location.”

Virtual testing could have covered the exact situation and allowed the technology to cope in the real world weather conditions. He explained: “In general terms, we need, what we call, measurable safety. You can collect the metrics to demonstrate that you actually exercise all of the, say, 10 million scenarios, including all of the variations. If it’s a stationary object in the way, for example, I want it as a small car, a big car at different angles or else something that has fallen from another car in front.”

Binyamini concluded that the lessons learned in the semi-conductor industry should now be more widely applied to automotive. “With this sort of data, you can compile all of the possible scenarios, systematically measured, and show that you have effectively dealt with all of the possible different situations. We believe this approach can be deployed in the automotive industry. I’m optimistic that the industry is on the right path and that path is going to lead the industry into an approach of measurable safety that will allow people to do the thorough work in the lab before the car is put on the road.”

— Paul Myles is a seasoned automotive journalist based in London. Follow him on Twitter @Paulmyles_