BMW-backed Mapillary Provides Hyper Mapping Services for Driverless Vehicles

BMW-backed Mapillary Provides Hyper Mapping Services for Driverless Vehicles

Author: Michael Cheng   

Autonomous vehicle networks rely on real-time mapping data during navigation on public roads. Such requirements are stringent, as most maps today are updated on a monthly or yearly basis. When using old maps, driverless cars could be prone to encountering unforeseen structures and signs.

Swedish startup Mapillary wants to enhance mapping services for networks of self-driving cars, connected cities and autonomous fleets. To ensure its maps are always updated, the BMW-backed company is implementing a crowdsourcing model to address bottlenecks in data gathering.

Crowdsourced Mapping Data

Founded in 2013, the company has effective methods for keeping contributors engaged with the mapping platform. Mapping challenges that encourage winners with prizes and a leaderboard-tracking feature for contributors have been successful in keeping the startup's database of images fresh. Images of faces and license plates are intentionally blurred out to maintain privacy.

"Driverless cars need the latest view of the road," said Jan Erik Solem, CEO of the startup.

"They require a higher and higher update frequency for maps, from quarterly to monthly to weekly to daily. The only scalable way to do that is using technology."

To date, the startup has acquired more than 430 million images from 190 countries, resulting in over 186 digital mapping features. Streamlining this aspect of the platform is Mapillary's robust computer vision algorithms, which are capable of detecting 42 different classifications, including utility poles and barriers. Interestingly, the platform accepts panoramic images and data from photo spheres.

The company's crowdsourced mapping services have attracted several tech and automotive giants in the sector. Mapillary's partners include the following: Toyota, Audi's AID, HERE, Mapbox and the World Bank.

Catering to Autonomous Vehicles and Cities

The startup's services are currently being leveraged by cities and automotive companies across the globe. Mapillary's clients range from the City of Amsterdam, as well as transport groups located in Vermont and Arizona. Additionally, at the City of Clovis in New Mexico, local organizations are relying on the startup's mapping services to boost large-scale maintenance services of roadway signs, potholes and more.

Before Mapillary's platform, the city paid a contractor to meticulously log the location's surrounding environment. To ensure accurate data, the contractor was required to record mailboxes, fire hydrants and road signs, which took some time to process.

"If you look at any object on a street, someone is responsible for it. It all has to be catalogued and checked," explained Steven Hewett, a representative from the City of Clovis.

"Without this software, we'd be walking around collecting all the data by hand, and I can't even guess how long that'd take for a 23-square-mile city."

For the World Bank, Mapillary's services supported the financial organization's infrastructure development projects. During the projects, the group worked with local contractors based in China to capture local images for evaluation using the Mapillary app.

At the moment, the startup's offerings are free to use for charities, educational programs and personal projects. Commercial applications require a licensing fee. Mapping features are also available under a subscription plan (pricing is based on road specifications).

Michael Cheng
Michael Cheng
Michael Cheng is a legal editor and technical writer with publications for Blackberry ISHN Magazine Houzz and Payment Week. He specializes in technology business and digesting hard data. Outside of work Michael likes to train for marathons spend time with his daughter and explore new places.
Prev:Porsche Taycan Owners to Get 3 Years of Free Charging from Electrify America Next:Volvo, Veoneer’s Autonomous Joint Venture to Start Testing in Sweden
    view more