Recogni aims to offer high-performance, low-power AI processing for AVs

0

USA/German startup Recogni Inc. has emerged from ‘stealth mode’ with US$25m in Series A funding and information about its vision-oriented artificial intelligence (AI) platform for autonomous vehicles (AVs).

As the automotive industry is transitioning to automated transport, a network of computers is needed to drive these new vehicles efficiently on a limited energy budget. While current AI systems are trained offline, they need to process the sensor data in real-time within the vehicle, but many systems have hit the processing efficiency ‘wall’ and are unable to transition to SAE Level 3+ autonomy and beyond. Recogni says it is focused on creating high-performance and low-power AI processing to help make self-driving vehicles a reality.

Founded in 2017 and headquartered in San Jose, California, the company has operations in Munich, Germany and is positioning itself to revolutionize perception processing for Level 2+ autonomous vehicles. Using a Vision Cognition Processor, Recogni aims to solve the endpoint inferencing problem with AVs efficiently and change the trajectory of full level 3, 4, and 5 autonomy. The company’s founders possess deep industry experience in system design, AI, vision, and custom silicon design.

The company has just announced US$25m of new financing led by GreatPoint Ventures, with participation from Toyota AI Ventures, BMW i Ventures, Faurecia (one of the world’s leading automotive technology companies), Fluxunit (venture capital arm of lighting and photonics company OSRAM), and DNS Capital. Recogni plans to use the funds to deliver the most capable inferencing system to enable state of the art sensor fusion of visual and depth sensor data while continuing to grow its top-tier engineering team. Recogni is currently in discussions with multiple auto manufacturers, to provide them with the full suite of enabling technology from modules to the software.

Recogni’s system identifies small objects, such as traffic lights, from over 656 feet (200m) away in real-time. Unlike lidar and radar systems, Recogni can tell if the lights are red, yellow or green because it works on imaging data. The Recogni Vision Cognition System uses a diverse set of image sensors to identify significantly smaller objects at a much larger distance compared to competitors, while consuming a fraction of power. AVs need vast amounts of computation power at very high efficiency, and even with current accelerator architectures, they are still consuming tens of kilowatts of energy, which is not practical for mainstream usage. Recogni says its technology will change that entirely and get the work done for the entire car in less than 100W, while processing the mass of data in real-time.

“The issues within the Level 2+, 3, 4 and 5 autonomy ecosystem range from capturing/generating training data to inferring in real-time. These vehicles need datacenter class performance while consuming minuscule amounts of power,” said RK Anand, CEO of Recogni. “Leveraging our background in machine learning, computer vision, silicon, and system design, we are engineering a fundamentally new system that benefits the auto industry with very high efficiency at the lowest power consumption. This round, one of the largest initial venture rounds raised by any AI silicon company in the space, is testament to our experience and responsible approach.”

Share.

About Author

mm

Adam joined the company in 1994, and has been News Editor of TTT since 2009. In his other role as Circulation Manager, he helped create the original Traffic Technology International distribution list 23 years ago, and has been working on it ever since. Outside of work, he is a keen fisherman, runs a drumming band, and plays an ancient version of cricket.

Comments are closed.