AImotive's latest aiWare3P delivers superior NN acceleration for production L2-L3 automotive AI

News & insights

Check our latest stories on automated driving

[#if Thumbnail??]
    [#if Thumbnail?is_hash]
        [#if Thumbnail.alt??]
            ${Thumbnail.alt}
        [/#if]
    [/#if]
[/#if]

Written by AImotive / Posted at 12/19/19

AImotive's latest aiWare3P delivers superior NN acceleration for production L2-L3 automotive AI

BUDAPEST, Hungary, Dec. 19, 2019 – AImotive, one of the world's leading suppliers of modular automated driving technologies, announced that it has begun shipment of the latest release of its acclaimed aiWare3 NN (Neural Network) hardware inference engine IP. The aiWare3P IP core incorporates new features that result in significantly improved performance, lower power consumption, greater host CPU offload and simpler layout for larger chip designs.

"Our production-ready aiWare3P release brings together everything we know about accelerating neural networks for vision-based automotive AI inference applications;" said Marton Feher, senior vice president of hardware engineering for AImotive. "We now have one of the automotive industry's most efficient and compelling NN acceleration solutions for volume production L2/L2+/L3 AI."

Each aiWare3P hardware IP core offers up to 16 TMAC/s (>32 TOPS) at 2GHz, with multi-core and multi-chip implementations capable of delivering up to 50+ TMAC/s (>100 INT8 TOPS). The core is designed for AEC-Q100 extended temperature operation and includes a range of features to enable users to achieve ASIL-B and above certification. Key upgrades include:

  • Enhanced on-chip data reuse and movement, scheduling algorithms and external memory bandwidth management
  • Improvements ensure that 100% of most NNs execute within the aiWare3P core without host CPU intervention
  • Range of upgrades reducing external memory bandwidth requirements
  • Advanced cross-coupling between C-LAM convolution engines and F-LAM function engines
  • Physical tile-based microarchitecture, enabling easier physical implementation of large aiWare cores
  • Logical tile-based data management, enabling efficient workload scalability up to the maximum 16 TMAC/s per core
  • Significantly upgraded SDK, including improved compiler and new performance analysis tools

The aiWare3P hardware IP is being deployed in L2/L2+ production solutions, as well as studies of advanced heterogeneous sensor applications.  Customers include Nextchip for their forthcoming Apache5 Imaging Edge Processor, and ON Semiconductor for their collaborative project with AImotive to demonstrate advanced heterogeneous sensor fusion capabilities.

As part of their commitment to open benchmarking using well-controlled benchmarks reflecting real applications, AImotive will be releasing a full update to their public benchmark results in Q1 2020 based on the aiWare3P IP core.

The aiWare3P RTL will be shipping from January 2020.