Micron Technology and Continental plan to enter into a collaboration agreement to explore and adapt Micron’s deep learning accelerator for next-generation machine learning automotive applications.
This agreement addresses the need to meet the extreme memory requirements of modern vehicles as automobile infotainment, communications, advanced driver-assistance systems (ADAS) and powertrain control systems are becoming increasingly sophisticated.
Continental and Micron will work together to develop an application-specific version of Micron’s deep learning accelerator (DLA) technology designed to be flexible and scalable while delivering the low power and high performance needed to support industry-standard programming models.
“Working together with Micron to build a scalable and flexible solution for edge inference that supports multiple networks and frameworks, will enable us to efficiently deploy machine learning across our platforms and deliver intelligent mobility technologies for our customers ,” said Dirk Remde, VP of Innovation Center Silicon Valley at Continental.
Continental envisions deploying deep learning solutions in areas such as ADAS advanced development, in-cabin monitoring and other connected and autonomous mobility solutions.
“Machine learning is a critical component to ADAS, making it imperative to understand the requirements and use cases for this market,” said Steve Pawlowski, corporate VP of Advanced Computing Solutions at Micron.
“One of our key goals in collaborating with Continental is to create an agile edge-inference solution that uses machine learning and delivers the ease of use, scalability, low power and high performance the automotive sector requires,” said Pawlowski.
Micron integrates compute, memory, tools and software into a comprehensive artificial intelligence (AI) development platform with its DLA technology. Micron and Continental will leverage this platform to explore innovative memory optimised for AI workloads, including deep learning solutions for data analytics, particularly targeted toward the internet of things (IoT) and edge computing.
