To understand what Mobileye does, it is important to first understand the technology that our engineers use to design our products.

Artificial Intelligence

Artificial Intelligence is when a computer system exhibits behavior that is thought of as requiring intelligence and solves complex problems. With AI, a computer is smart enough to take actions to achieve its goals. It has the ability to learn from experience over time. Artificial Intelligence has been around since the 1940s, but in the 1990s, AI research accelerated and focused on real-world problems. Since 2010, big data has been applied to enhance machine learning and create smarter algorithms. As a result, more advanced AI has emerged. Image recognition, such as that which is used in Mobileye’s sensing technology, is an area where there has been significant progress and investment.

AI has the potential to become a major contributor to economic growth and have a positive impact on our society by making education more impactful, allowing the elderly to stay independent, providing mobility to people with disabilities, and making buildings smarter. Many of us already experience how AI can make our lives easier when we use Apple’s Siri and IBM’s Watson. Other familiar applications are Google Assistant, Amazon Alexa, and Microsoft Cortana, just to name a few.

How Mobileye Uses AI

Autonomous driving is a multi-agent situation, in which the host vehicle must use sophisticated negotiation skills to drive in harmony with other vehicles on the road. The autonomous vehicle has to consider the movement of other vehicles when overtaking, giving way, merging, taking left and right turns and while pushing ahead in unstructured urban roadways. There are so many possible scenarios, as well as unexpected behavior from other drivers, pedestrians, and other road users, that Mobileye uses AI to train its vehicles to make the types of decisions needed to negotiate the road. We create environmental models and maps from the data that our cameras collect. Our technology is one of the first embedded versions of AI, which means the technology does not live in the cloud but instead lives in the vehicle, on the chip.

Deep Learning 

Deep learning is a branch of machine learning in which a machine uses algorithms to learn from data. With machine learning, a machine learns to do better in the future based on what was experienced in the past. In the past engineers used code to program all activities that they wanted a computer to complete. With machine learning, a computer is trained with large amounts of data to figure out how to do the task by itself. In other words, the goal is to devise learning algorithms that do the learning automatically without human intervention or assistance. The machine learning paradigm can be viewed as “programming by example.” Often we have a specific task in mind. But rather than program the computer to solve the task directly, in machine learning, we seek methods by which the computer will come up with its own program based on examples that we provide. Deep learning uses structures that are similar to the way the human brain thinks and allows a machine to be given layers of information. The computer can recognize a complex pattern in the data and reacts accordingly.

How Mobileye uses Deep Learning

Deep learning is what powers Mobileye’s computer vision system. Sensing algorithms use a technique called Supervised Learning, while our Driving Policy algorithms use Reinforcement Learning, which is a process of using rewards and punishments to help the machine learn how to negotiate the road with other drivers. We are training the vehicle to take the information that the sensors provide it, and the data that the computer chip is processing, to make informed decisions based both on past learning and informed predictions on how other drivers will react to our decisions based on simulations.

Internet of Things (IoT)

The Internet of Things is when various devices are connected to each other via the Internet so they can collect data, interact and share information. More specifically, it refers to things (like thermostats, cars and phones) that did not used to be connected to the Internet but now are. If you can use the Internet to control a machine, or if you are getting data from that machine via the Internet, that machine is part of the Internet of Things.

How Mobileye is Taking Part in the IoT

Mobileye uses the Internet to share the location-based map data that it gathers on its vehicles with other cars. Then, other cars can learn from that data and that data can be used to crowdsource a map. The IoT is what will ultimately allow one driverless vehicle to communicate with the other so they can safely negotiate the road together.


Crowdsourcing is when you gather information from a large variety of sources, with the sources supplying data either in an active or passive way. Wikipedia is the best-known example of crowdsourcing. The website gives the crowd a platform for sharing information via the web. Another example is when a company crowdsources ideas for a new advertising campaign or an app is created to allow drivers to crowdsource information on traffic jams in their immediate area.

How Mobileye uses Crowdsourcing

Mobileye’s mapping technology, called Road Experience Management (REM™), crowdsources real-time data for precise localization and high-definition lane data. This data is crowdsourced from the millions of vehicles on the road with cameras and Mobileye chips on-board. The images that these vehicles accumulate as they drive are processed by Mobileye algorithms, with the goal to create a Roadbook that contains a highly-accurate (within 3cm-10cm) map of the average drivable path on every lane of every road in the world. These Mobileye-equipped vehicles will also sense when the environment changes, and update the Roadbook in real-time.