Aeva®, a leader in next-generation sensing and perception systems, today announced Aeva Atlas™ Ultra, its newest 4D LiDAR sensor built to meet the performance demands of SAE Level 3 and 4 automated driving systems in passenger and commercial vehicles. To enable safe travel at highway speeds, Atlas Ultra provides up to three times the resolution of Atlas, and configurable field of views with up to 150 degrees of vision across the horizon. On-sensor perception software enables unique detection capabilities at a maximum detection range of up to 500 meters. A 35% slimmer design makes Atlas Ultra ideal for passenger cars in roofline and behind windshield integrations with minimal impact to vehicle styling and aerodynamics.

“Atlas Ultra is our most powerful LiDAR sensor yet, designed to meet the performance needs of L3 and L4 highway driving,” said Mina Rezk, Co-Founder and CTO at Aeva. “Over the past few years, Aeva’s pioneering industrialization of 4D LiDAR coupled with our production programs and ongoing engagements with multiple top global automotive OEMs has converged on the highly versatile Atlas Ultra product. We believe Atlas Ultra, along with its critical on-sensor perception and localization offering, provides a competitive edge for passenger and commercial vehicle OEMs looking to introduce highway speed functionality without compromise.”

Delivering Industry-leading Performance
Aeva’s unique Frequency Modulated Continuous Wave (FMCW) 4D LiDAR technology detects velocity and position simultaneously to enable new levels of safety and automation. Using instant velocity data, automated driving systems can detect important objects faster, with higher confidence, and from greater distances.

Atlas Ultra delivers on critical requirements for highway-speed autonomous driving with a 250 meter detection range for low-reflectivity targets and a maximum detection range of up to 500 meters. Importantly, Atlas Ultra enables clear perception across a wide variety of everyday driving scenarios because it is immune to interference from direct sunlight, signals from other LiDAR sensors, and to blooming and ghosting around retroreflective objects like lane markers and street signs.

On-Sensor Perception Capabilities
Leveraging per point velocity information, Atlas Ultra provides perception and localization outputs directly from the sensor to enable unique capabilities, while helping to reduce compute costs. These capabilities include:

• Small Object in Roadway Detection: Detect small objects on the roadway and accurately discriminate retroreflective lane markers from non-drivable road hazards, such as a small brick or tire fragments, at distances of up to 150 meters.
• Dynamic Object Detection and Tracking: Detect and track the position and velocity of all dynamic objects with centimeter accuracy at up to twice the distance of conventional 3D time-of-flight LiDAR sensors.
• Late Reveal Dynamic Object Detection: Instantaneously detect dynamic objects that suddenly appear even while remaining partially obstructed, such as a pedestrian or animal emerging onto the road from behind a parked car or thick vegetation.
• Lane Lines & Drivable Region Detection: Detect lane lines and determine drivable regions.
• Aeva Ultra Resolution™: A real-time camera-like image that provides a high-density point cloud resolution.
• Advanced Vehicle Motion Estimation: Estimate vehicle motion in real-time with six degrees of freedom for accurate positioning and navigation without the need for additional inertial or positioning sensors.
   

Advanced AI Perception Capabilities
Additional advanced perception functions harness the latest AI-based detection, tracking and classification algorithms. These advanced perception algorithms utilize per point velocity information to deliver advantages for automated vehicle systems.

• Pedestrian & Vehicle Detection: Detect, classify and track static and dynamic objects including vulnerable road users (VRUs) such as pedestrians, bicyclists, and motorcyclists, as well as a large range of vehicle classifications including passenger cars, trucks, buses, emergency vehicles and more.
• Semantic Segmentation: Per-point segmentation of the entire point cloud classifies points as: pedestrians, vehicles, lane markings, drivable regions, road signs, infrastructure, vegetation, and more.
   

Powered by Aeva Custom Silicon
The advancements in Atlas Ultra are powered by custom silicon including the Aeva CoreVision™ Lidar-on-Chip module, and Aeva X1™ System-on-Chip processor.

• Aeva CoreVision: Designed to strict automotive standards, Aeva’s fourth-generation LiDAR-on-Chip module incorporates all key LiDAR elements including transmitter, detector and an optical processing interface chip. Built on Aeva’s proprietary silicon photonics technology, CoreVision replaces complex optical fiber components with silicon photonics, ensuring quality and enabling mass production at affordable costs.
• Aeva X1: Aeva’s powerful FMCW LiDAR System-on-Chip processor seamlessly integrates data acquisition, point cloud processing, scanning system and application software into a single mixed-signal processing chip. Designed for dependability with automotive-grade functional safety and cybersecurity built in.
   

Meets Automotive Requirements
Designed to meet challenging OEM requirements across passenger and commercial vehicle applications, Atlas Ultra meets automotive grade requirements for functional safety ISO 26262 ASIL-B (D), cybersecurity ISO 21434, and is IP69K compliant for robust dust and water protection, and strong shock and vibration endurance.

(Photo credit: Aeva)

TrendForce 2025 Infrared Sensing Application Market and Branding Strategies
Release: 01 January 2025
Format: PDF / EXCEL
Language: Traditional Chinese / English
Page: 196