Cognitive Vehicles

TomTom is Leveraging Deep Learning in its Map-Making Processes

Frans de Rooij
Regional Director Product Management, Autonomous Driving

we.CONECT had the chance to meet with Frans de Rooij, Director Product Management at TomTom to talk about the role AI and cognitive computing play in self-driving car technologies and how TomTom is applying deep learning in its map-making processes.

Frans De Rooij is regional director product management for TomTom Autonomous Driving. He is based on TomTom’s Michigan office and responsible for building collaborations with OEMs, Tier 1s and other partners to deploy TomTom HD map products in self-driving vehicles, and for taking the market requirements back into TomTom product development roadmaps. In previous roles at TomTom, Frans was director product marketing for TomTom Automotive and product manager for TomTom Traffic. Prior to TomTom, he worked as strategy consultant at Arthur D. Little and in several product management and sales & marketing roles. He holds an MSc from TU Eindhoven (NL), a PhD from Cambridge University (UK) and an MBA from the University of Warwick (UK).

we.CONECT: What are your main responsibilities in your current role?

Frans de Rooij: I bring the TomTom HD map products for Autonomous Driving to the U.S. market, and bring the needs from the market to the TomTom product development team.

we.CONECT: What fascinates you most about autonomous driving?

Frans de Rooij: On the short term, that it will gradually make driving more comfortable, and especially, safer. On the longer term, autonomous driving will take transportation to an entirely new level: offering individual mobility to everyone, reducing the parking space needed in cities, and thus freeing up time and space.

we.CONECT: What role does AI and cognitive computing play in self-driving car technologies?

Frans de Rooij: In the cloud, AI and cognitive computing play a big role in making sense of the vast amounts of data that can be crowdsourced, such as in the HD map creation and maintenance processes that TomTom uses. In the car, these technologies will also power the real-time processing of sensor observations to build an environmental model and construct a safe, driveable path.

we.CONECT: There is a lot of confusion around autonomous driving. What are the different routes to level 5 and what are the biggest challenges to get there in the short term (in the next 5 years)?

Frans de Rooij: For privately owned vehicles, we are already seeing an increase of ADAS driver warnings and assistance, building up to automation that allows the driver to disengage for a short amount of time under specific circumstances, such as highway driving. The challenge is to bring these features at a cost level (for sensors, maps, software, etc.) that is in line with the value they bring to private individuals. For specific routes (like on-campus transport, or airport-to-parking) and specific areas, where the environment can be to some extent simplified, vehicles are already driving in an increasingly autonomous way. I expect these “robotaxi”-like vehicles to gradually expand the environments that they can handle. The cost of sensors and other system components play a smaller role, as the vehicles are not owned by individuals, and the cost comparison is versus having a full-time on-board driver.

we.CONECT: How is your company developing deep learning capabilities? What are the challenges?

Frans de Rooij: TomTom is leveraging deep learning in its map-making processes. We process huge amounts of data, both from our own sources such as our survey fleet, as well as from the crowd, such as the GPS traces and Sensor-Derived Observations that we receive. The challenge, and our mission, is to create HD Maps with an increasing amount of features in a globally scalable, maintainable and affordable way. Deep learning makes a high level of automated feature extraction possible.

we.CONECT: Thank you very much for taking part in our interview.

About TomTom Automotive:
TomTom is the leading independent location technology specialist, shaping mobility with highly accurate maps, navigation software, real-time traffic information and services. To achieve their vision of a safer world, free of congestion and emissions, TomTom creates innovative technologies that keep the world moving. By combining their extensive experience with leading business and technology partners, the company powers connected vehicles, smart mobility and, ultimately, autonomous driving. Headquartered in Amsterdam with offices in 30 countries, TomTom’s technologies are trusted by hundreds of millions of people worldwide.

Car HMI USA is happy to present TomTom as event partner. Car HMI USA is an award-winning international knowledge exchange platform bringing together all stakeholders leading the technical and design future of automotive HMIs and UX. Over 2.5 days you will get the unique chance to engage with your OEM, T1 & Supplier peers to discuss innovative opportunities as well as challenges in HMI and UX development that lie ahead in the industry.

29 - 30 April | The Henry, Detroit, USA