Algolux Expands Eos Perception Software to Address Critical ADAS and Self-Driving Car Depth Limitations, Canadian Business Journal

MONTREAL, Sept. 12, 2022 (GLOBE NEWSWIRE) — Algolux Inc., a leading provider of robust and scalable depth sensing solutions, has announced Eos Robust Depth Perception software. The expanded AI software product builds on the company’s advanced perception portfolio, providing both industry-leading dense depth estimation and robust perception, enabling passenger and It further enhances the safety of transport vehicles.

This new offering applies the same breakthrough deep learning approach used in Algolux’s award-winning robust recognition software to power today’s automated driver assistance systems (ADAS) and self-driving vehicles ( AV) address platform cost and performance limitations.

Mercedes-Benz has a proven track record as an industry leader in providing ADAS and automated driving systems, so we are working to further improve the perception capabilities of these systems so that the vehicle can be operated more safely in all operating conditions. We fully understand the need to do so,” said Manager Werner Ritter. , Visual Enhancement Technology Environmental Perception, Mercedes-Benz AG. “Algolux is leveraging new AI techniques in its camera-based Eos Robust Depth Perception software to provide high-density depth and detection, delivering next-generation range and robustness, while also enabling advanced technologies for ADAS and autonomous driving.” It addresses the limitations of current lidar and radar-based approaches.

The ability to estimate the distance to an object is a basic function of ADAS and autonomous driving. This requires the vehicle to be aware of its surroundings and know when to change lanes, assist parking, warn the driver or automatically brake in emergencies such as debris on the road. you can know if there is This is currently done through various types of sensors such as lidar, radar, stereo or mono cameras, and edge software that interprets sensor information to determine the distance to important objects, features or obstacles in front of and behind the vehicle. Realized.

Unfortunately, each current approach has limitations that prevent secure operation in all circumstances. LIDAR has a limited effective range (up to 200m) because the point density decreases as the object moves away. This results in poor object detection and less robustness in harsh weather conditions such as fog and rain due to laser backscatter. They are also costly, currently hundreds to thousands of dollars per sensor. Radar has good range and robustness, but low resolution and limited ability to detect and classify objects. Today’s stereo camera approaches are excellent for object detection, but are difficult to maintain calibration, have low robustness, and mono cameras have many problems and poor depth estimation.

Eos Robust Depth Perception software addresses these limitations by robustly providing high-density depth and accurate perception capabilities to identify distances and elevations even at long distances (1 km), objects, walking locating pedestrians, bicycles, even lost cargo and other dangerous road debris. Further improve driving safety. These modular capabilities provide rich 3D scene reconstruction, providing a highly functional and cost-effective alternative to lidar, radar, and today’s stereo approaches.

Eos accomplishes this by:

• A multi-camera approach that supports a wide baseline between cameras (sometimes over 2m) is especially useful for long-haul applications such as trucking.
• Flexible support for automotive camera sensors up to 8MP and any field of view in front, rear and surround configurations
• Real-time adaptive calibration to cope with vibration or motion between cameras or misalignment while driving.
• Efficient Embedded Implementation of Algolux’s New End-to-End Deep Learning Architecture for Both Depth Estimation and Recognition
• Ability to detect and distance typical road objects and even unknown road debris

Algolux has proven Eos depth perception performance in OEM and Tier 1 engagements including both trucking and automotive applications in North America and Europe. To learn more and schedule your evaluation, please visit www.algolux.com.

About Algorx

Algolux is a globally recognized computer vision company tackling critical safety issues in advanced driver assistance systems and autonomous vehicles. Our machine learning tools and embedded AI software products enable existing and new camera designs to achieve industry-leading depth perception performance in all driving conditions. Founded on groundbreaking research at the intersection of deep learning, computer vision and computational imaging, Algolux has been repeatedly recognized by industry and academic conferences, and was named one of the world’s most innovative artificial intelligence startups in 2021 CB Insights AI Selected for the 100 list.

Dave Tokic, VP Marketing and Strategic Partnerships
[email protected]

Related image

Image 1: Algolux Eox Robust Depth Recognition Software

Industry-leading robust detection and high-density depth for ADAS and autonomous vehicles in all driving conditions

This content was published through Newswire.com’s press release distribution service.

• Algolux Eox Robust Depth Recognition Software

CBJ Newsmaker