Kontron is presenting boards and DIN rail controllers based on i.MX8M Plus at SPS 2024
Artificial intelligence is making its way into devices and control cabinets.
www.kontron.com
Kontron, one of the world's leading suppliers of IoT/Embedded Computer Technology (ECT), is expanding its product range based on the powerful i.MX8M Plus processor to include CPU boards and rugged DIN rail computers. The integrated NPU offers up to 2.3 TOPS for demanding applications, making the Arm®-based board perfect for machine learning and vision systems.
The BL i.MX8M Plus features a high-performance Arm® Cortex® A53 microprocessor with a clock speed of 1.8 GHz and its memory can be expanded to 8 GB LPDDR4-RAM. The integrated Arm® Cortex® M7 coprocessor ensures real-time control of machine-related processes.
The board comes with the latest multi-media functions: a 4K HDMI connection, an 2D/3D GPU, and a video processing unit (VPU). The VPU supports H.265 (High Efficiency Video Coding) 1080p60 video encoding and decoding. The BL i.MX8M Plus is ideal for demanding applications in HMIs, medical devices and control solutions thanks to its improved use of memory and bandwidth combined with higher video quality. The board’s dual display solution provides an LVDS interface for a local display and a HDMI interface for an external monitor. It also comes with a serial display MIPI CSI interface that provides four lanes. Alternatively, camera systems can be connected via the external USB 3.0 type C or Gigabit-Ethernet interfaces.
The integrated Cortex® M7 processor core enables real-time control via the CAN FD interface as well as the dual GbE connection. The LAN port supports the Time Sensitive Networking (TSN) standard for secure and quick real-time data transmission. The low latency is suitable for time-critical applications that place high demands on data transmission and synchronization. The powerful NXP i.MX8M Plus SoC features an integrated neural processing unit (NPU) with a 2.3 TOPS AI processor. This supports the processing of large amounts of data in real time, making it ideally suitable for applications in the areas of machine learning and vision systems.
The first prototype for customers will be available starting in Q1/2025.
www.kontron.com