Road
Experience
Management™

To facilitate autonomous driving anywhere in the world, we first need to map everywhere in the world. But not just any maps will do. Instead of overemphasizing global accuracy, Mobileye’s crowdsourced, continuously updated map of the world digitizes precisely what AVs need – nothing more, nothing less. We’re making maps for the autonomous future.

Enabling Autonomous Vehicles
Anywhere in the World

The broad, global coverage of the Mobileye Roadbook, powered by REM, enables deployment of AVs in new locations almost instantaneously. Below, you can see videos of our developmental AV driving in different countries, on different continents, in very different environments

Unedited 40-Minute Mobileye AV Ride in Jerusalem

Unedited 1-Hour Mobileye AV Ride in Munich

How it works

Semantic Identification

The algorithms establish the relevant semantics for the car to understand the immediate driving environment.

5

Mobileye Roadbook

This information is compiled into the Mobileye Roadbook – a global bank of driving paths, semantic road information, and aggregate driving behavior.

6

Localization

The global Mobileye Roadbook is continuously sent to production vehicles to be used in autonomous applications.

7

Data Collection

Vehicles equipped with Mobileye technology automatically collect anonymous data from the road as they drive.

1

Identifying & Transmitting

REM™ classifies the relevant information into tagged data points and sends them to the cloud at low bandwidth.

2

Aggregation & Alignment

Data points from individual drives along the same road are aggregated and aligned.

3

Modelling

Post alignment, the algorithms define the precise outline of the road infrastructure.

4

REM maps have no other sources of data. The Mobileye Roadbook is produced independently of any third-party map and relies exclusively on the process you see here.

Unlike conventional static maps, the Mobileye Roadbook encompasses a dynamic history of how drivers drive on any given stretch of road to better inform the AV’s decision-making process.

Privacy is fundamental to Mobileye. Our data-collection processes implement strict methodologies to comply with data protection and privacy laws, foremost among them the European General Data Protection Regulation (GDPR). The data we collect and process is anonymous except where we explicitly state otherwise.

Common Industry Approach vs. The Mobileye Approach

Common Industry Approach – HD Maps

Not Scalable
Reliance on dedicated mapping fleets with expensive sensors (lidar, camera, INS)
Manual/semi-automatic mapping process
Infrequently updated – changes are not reflected in real time
Overly specified geometric accuracy in global coordinates
Relying on raw data on road semantics that have no specifics or insight on human driving patterns

Mobileye’s Approach – AV Maps

Scalable-by-design
Millions of Mobileye-equipped ADAS vehicles sending data to the cloud in small data packets (10kb/km)
Fully automated map generation at the push of a button
Maps can be updated in near-real time because of sophisticated change-detection algorithms on millions of mapping agents
Superior local accuracy:
where it matters
Using the “wisdom of the crowd” to create rich semantic layer of driving culture and traffic rules

Why REM is a unique solution for AV mapping

How AV R&D is impacting ADAS today

While the Mobileye Roadbook was developed for a future in which self-driving cars are ubiquitous, many of the safety benefits can reach roads today by adding our maps to advanced driver-assistance systems.