FutureFive New Zealand - Consumer technology news & reviews from the future
Story image
WeRide demonstrates pre-commercial level 4 autonomous driving solutions
Wed, 20th Mar 2019
FYI, this story is more than a year old

WeRide announced it will demonstrate its L4 sensor suite and pre-commercial solutions during the NVIDIA GPU Technology Conference 2019 (GTC), and display its latest L4 autonomous vehicle model, the Nissan LEAF 2.

The company also announced it will use the NVIDIA DRIVE AGX Pegasus platform for its next-generation autonomous vehicle fleet.

WeRide CEO Tony Han says, “The pre-commercial L4 solutions that we are debuting at GTC are paving the way for WeRide's rollout of RoboTaxi services in the future.

“Our demonstration of the Nissan LEAF 2 is a significant step forward in showing that WeRide can help bring reliable, safe autonomous vehicles to market. Currently, WeRide is establishing partnerships with OEMs in China and around the world.

Technology Updates

The new sensor suite is combined with WeRide's latest hardware and software solutions and is designed to increase accuracy and safety:

  • Wavelength detection on roof rack covers full 360-degree FoV (Field of View). The multiple wavelength cross-verified sensor design makes the detection more reliable.
     
  • The performance of LiDAR sensors has been significantly improved with up to 250 meters detection range facing front and extra coverage on both sides.
     
  • A front nose LiDAR for early detection has been designed to adapt to road and traffic conditions in China.
     
  • Data volume collected from all sensors is increased significantly, with real-time synchronisation leveraging the NVIDIA DRIVE AGX Pegasus platform.

About NVIDIA DRIVE AGX Pegasus

The NVIDIA DRIVE AGX Pegasus AI supercomputer achieves an unprecedented 320 trillion operations per second (TOPS) with an architecture built on two NVIDIA Xavier systems-on-a-chip (SoCs) and two next-generation Tensor Core GPUs.

This energy-efficient high-performance platform runs an array of deep neural networks simultaneously and is designed to safely handle highly automated and fully autonomous driving.