MAPPING “WE HAVE DEALS WITH NINE CAR MAKERS WHO SEND US DATA, AND WE HAVE MILLIONS OF CARS EVERY DAY MAPPING THE ENTIRE WORLD AT A VERY LOW COST” Amnon Shashua, CEO, Mobileye The road to autonomous driving is increasingly being led by ADAS specialists such as Intel’s Mobileye, as they add more competence to what started out as simple lane-keep and adaptive cruise control systems. Automatic for the people Mobileye has changed the conversation on what HD mapping means. What started out as a laborious process of running a small fleet of specially adapted vehicles to map a defined area has become an exercise in using millions of regular cars to grab low-data snapshots to build a comprehensive picture of the roadways of entire countries. “The conventional way of creating an HD map is to drive a lidar-and camera-equipped specialized vehicle around a city, collecting cloud points everywhere,” says Amnon Shashua, founder and CEO of Mobileye. “Then someone does some manual work to build it into a map. It costs about US$10m per city, which is huge. And then you need to update them because things change.” Mobileye instead uses its EyeQ camera/sensor, which has become an ADAS standard, to create Road Experience Management (REM) maps for THE AUTO INDUSTRY HAS DEFINED A STANDARD PROTOCOL FOR MAP DATA INFORMATION EXCHANGE, CALLED ADAS INTERFACE SPECIFICATION (ADASIS) what it claims is a much lower cost. “It’s not the raw data that you send to the cloud. It’s processed data – about 10KB per kilometer,” continues Shashua. “We have deals with nine car makers who send us data, and we have millions of cars every day mapping the entire world at a very, very low cost.” This is not high definition because of the lack of lidar collection, but Shashua claims that it has other advantages over HD. “There’s also a lot of information about how humans drive that you don’t find in high-definition maps.” The swarm data collected shows how road users interact with the infrastructure – where they position themselves at junctions, for example. Grabbing 95,000,000km of data daily is not useful solely for informing Level 3 and 4 autonomy, but also for Mobileye’s Level 2+/Level 3 SuperVision technology and, from 2025, the company’s hands-off, eyes-off Chauffeur technology. Built from relatively sparse, anonymized, ultra-lightweight data, Mobileye’s REM HD maps are nevertheless rich in detail Such is Mobileye’s lead in this area that some car companies are looking for partners who can offer something similar but on a more collaborative basis, where the revenue is better shared. For example, Volkswagen and BMW have turned from Mobileye to Qualcomm for certain future vehicle platforms. Currently a leader in providing connectivity and computer power for infotainment systems, Qualcomm is expanding its semi-autonomous capability on its Drive platform, which includes a plan to use sensors from customer cars to map the roads on which they drive. Moving to scenarios where drivers can take their hands off the wheel and then their eyes off the road Revenue share 30 ADAS & Autonomous Vehicle International January 2024