The two companies merge advanced cameras, radars, sensors, software and computing power into a turnkey automated driving system ready for production in 2019.
Everyday challenges on the roads are hard enough for humans, so imagine a computer trying to manage highway merges, congested city streets with pedestrians and cyclists and a tunnel.
Through a 6.3-mile drive around Las Vegas this week (Jan. 2-7), the Delphi and Mobileye “Centralized Sensing Localization and Planning” (CSLP) automated driving system will prove it can do all of these things without the help of a human holding the steering wheel and pushing the gas pedal. CSLP is the first turnkey, fully integrated automated driving solution with an industry-leading perception system and computing platform. It will be ready for production in 2019.
“Three factors will separate the leader from the pack in the race to offer driverless vehicles by 2019–best-in-class perception sensors such as cameras, radar and LiDAR, automotive experience and computer processing speed,” said Glen De Vos, vice president of services for Delphi. “We will demonstrate the capability of the CSLP solution in our intensive drive at CES 2017 in Las Vegas.”
The 2019 CLSP system will feature several advanced technologies including:
- Localization capability: Ensures the vehicle knows its location within 10 cm; even without GPS connectivity
- Free space detection: Helps the car navigate complex lane splits or areas lacking lane markings
- 360-degree pedestrian sensing
- 3D vehicle detection: detects vehicles at any angle; can detect partial cars by identifying vehicles by overall shape and detect by wheel movement if a car is stationary or parked; critical for urban situations with unusually angled intersections; enables lateral turning vehicle detection critical for intersections
- Path and motion planning: Allows the car to behave more human-like in its driving behavior and determine the best path forward
In addition, teams from both companies continue to develop next generation of sensor fusion technology as well as the next generation human-like “driving policy.” This module combines Ottomatika’s driving behavior modeling with Mobileye’s deep reinforcement learning in order to yield driving capabilities necessary for negotiating with other human drivers and pedestrians in complex urban scenes.
Watch more videos:
Delphi Drive Vegas B-Roll Package
Delphi’s Multi-Layer Display (MLD®)