Hyundai CRADLE Investing in AI-based Fleet Software Company Netradyne
Before a self-driving vehicle can navigate, it requires a high definition (HD) map of the route, similar to how a human driver might get directions from Google Maps. However, these HD maps are much more detailed than the maps human drivers use for simple GPS navigation. HD maps offer centimeter level precision, so a self-driving vehicle knows precisely where it is in the world.
However, collecting data to built these maps is a huge undertaking, requiring vehicles to be manually driven on the route to be mapped while onboard cameras and sensors collect data to construct the map. This data includes the position of traffic signals, road lanes, trees, mailboxes and other infrastructure.
To address its future mapping needs, Hyundai Cradle, the venture arm and advanced mobility business of automaker Hyundai, is investing in Netradyne, a technology company founded in 2015 that specializing in fleet safety management software.
Netradyne uses artificial intelligence (AI) vision-based dashcam devices to monitor safety performance of fleet vehicles. The same data used to monitoring fleets can also be used to build maps. Netradyne's system can crowdsource road and driving behavior metadata from the fleet of vehicles while they are out driving around. This metadata can be used to build HD maps for autonomous vehicles and also keep them updated.
The partnership between Hyundai and Netradyne will support further development of Level 3 ADAS and autonomous driving features that may launch over the next few years in Hyundai vehicles.
"Netradyne has advanced vision-based and AI driven technology for managing the safety of drivers and fleets," said John Suh, Vice President of Hyundai CRADLE. "These systems can be leveraged as a reliable, high-quality crowdsourcing platform to support updating HD maps at a frequency high enough to support L3 autonomous vehicles."
HD maps for autonomous vehicles offer lane level accuracy. (PHOTO: HERE Technologies)
Hyundai and Netradyne will collaborate to utilize the road and driving behavior data collected by Netradyne to support HD mapping and map updates for Hyundai's future autonomous vehicles and advanced driver assist systems (ADAS).
Building HD maps that autonomous vehicles use for driving is a long process, requiring a human driver to drive a specific route multiple times to collect the data. Netradyne said it has analyzed more than 350 million miles of road data using AI. The data was collected by professional drivers.
Cameras and senors on the vehicles are used capture data to make the maps. Features such as the position of travel lanes, traffic signals, and even elevation are used to map detailed maps for autonomous vehicles.
The costs associated with outfitting traditional mobile mapping system (MMS) vehicles, which are currently employed for HD mapping, make it challenging for many OEMs to place enough mapping vehicles on the road to keep the maps constantly updated.
For example, if a road lane is closed due to construction, the maps require updating to reflect the new traffic pattern. The updated maps also need to be pushed to all nearby self-driving vehicles that will travel the area with a software or over-the-air update.
Netradyne's devices are more cost-efficient for collecting map data, since they are already equipped in numerous vehicle fleets throughout the country.
Netradyne said it has has already captured and analyzed over 1 million miles of the 2.7 million total miles of paved roads in the United States. This data includes numerous passes over the same roads to provide deeper insights into how driving and road different conditions may change throughout the year.
"We're very pleased to announce this partnership with Hyundai CRADLE," said Avneesh Agrawal,Founder & CEO of Netradyne. "At Netradyne, our goal has always been to
make the world a safer place to drive with our advanced AI platform. We're glad that Hyundai shares this vision and are excited to support its development in advanced ADAS and Autonomous Driving innovation."
This announcement follows last month's announcement that Hyundai and mobility solutions technology company Aptiv formed a $4 billion autonomous driving joint venture to advance the development and commercialization of autonomous technologies.
The joint venture will begin testing fully driverless systems in 2020 and have a production-ready autonomous driving platform available for robotaxi providers, fleet operators and automotive manufacturers by 2022.
- Tesla Model 3's Built in China Will Be Eligible for New Energy Vehicle Subsidies
- Italian Truck Maker Iveco Unveils the ‘Nikola TRE’, an Electric Truck Built in a Joint Venture with Nikola Motor
- A Look at the Asian Companies Looking to Dominate the Global EV Battery Market
- VW’s Audi to Cut 10% of its Workforce to Fund Shift Towards Electric Vehicles
- Hyundai Shows Off a Redesigned Electric IONIQ at the Los Angeles Auto Show
- Lidar Startup Ouster Introduces a More Affordable 32-Channel Lidar Sensor for Self-driving Cars
- China’s BYD Receives Order for 130 Electric Buses, its Biggest Order to Date in the U.S.
- BMW Announces its Next-Gen Digital Key Technology