The KBox is ready-to-use and can be integrated directly into existing X86 landscapes in order to quickly and efficiently implement your own application scenarios for machine learning and artificial intelligence at the intelligent edge. This includes, for example, image and video data processing for visual inspection, quality assurance, predictive maintenance, sorting or object recognition, as a stand-alone device or gateway. Kontron says that it will, in the future, also be making its AI platform available as a KBox PC based on an Intel Atom x7-E3950 processor.

With the integrated Google Coral Edge TPU (Tensor Processing Acceleration Unit) with up to 4 TOPS (trillion operations per second), the speed is increased many times over compared to applications with simple USB cameras without TPU. In addition, the TPU supports small and low power applications and only requires one watt for two TOPS.

Unlike the Kontron AI platform based on the NXP i.MX8M, which is available as a board product, the KBox A-203-AI-GC can be placed directly in the control cabinet or hung on a machine without any further preparation. The gateway solution is based on the tried and tested KBox A-203 system, which is used in practice for collecting, processing and distributing sensor data. It networks different end devices and ensures smooth and secure transmission to a higher-level server infrastructure. With the AI ​​accelerator, visual information can now also be processed. Two front USB 3.0

In the context of deep learning, the close conceptual connection between hardware and software is particularly important in order to optimally solve computationally intensive tasks. The AI ​​platform runs on Debian Linux, is compatible with TensorFlow Lite and is already equipped with application examples. This allows users to benefit from the wide range of applications offered by Google AI right from the start. On this basis, you can quickly and intuitively develop your own software solutions with a short time-to-market, which are based on pre-trained models for the respective application.

Kontron