• en
  • sv
  • e20xx/e21xx Compute Modules

    Unibap’s e20x0 and e21x0 heterogeneous computing modules for critical embedded applications are uniquely tailored and extended Qseven(R) modules featuring multi-core x86 CPU, GPU cores, and Microsemi FPGA with advanced error correction. The e20/e21 product families are using x86 architecture and are compatible with practially all available x86 linux software, including well known space frameworks such as  NASA’s Operational Simulator for Small Satellites (NO3).

    This product is distributed in the U.S.A. by Moog Inc through Moog Broad Reach.


    Heterogeneous computing combining real-time processing and x86 flexiblity

    Intelligent industrial automation
    Robot and drone guidance

    Product information

    Unibap e20x0/e21x0 family of Heterogeneous Processing Modules for advanced mission critical embedded solutions featuring 100 GFLOPs using AMD x86 CPU/GPU and Microsemi FPGA.

    The Unibap e20x0 and e21x0 products are the perfect choice for smaller autonomous systems with high demands of visual perception, sensor fusion, control and decision support using optimized the hyper parallel architecture.

    The modules are able to run artificial intelligence in real-time on high resolution video or sensor fusion with lidar, radar or other sensor inputs.

    A wealth of input/output (IO) is provided through the Qseven MXM connector and an additional board to board connector on the back for a total of 350 IO, including:

    • Gigabit LAN
    • Controller Area Network (CAN) v2.0b
    • USB 3.0
    • USB 2.0
    • SATA v3.0 (6 Gbps)
    • PCIexpress v2.0
    • SPI
    • I2C
    • LVDS (695 Mbps)
    • Serdes (5 Gbps)
    • Digital IO
    • Analog input

    Unibap support a full open source software stack including:

    • Linux kernel 4.9 (AMD64, x86-64)
    • (Optionally VxWorks, Windows Realtime)
    • UNiLINK driver
    • OpenGL 4.5
    • OpenCL 1.1/1.2
    • Hardware accelerated H.264 encoding
    • OpenCV
    • Caffe
    • Robotic Operating System, ROS
    • TensorFlow
    • FreeRTOS (ARM, in the FPGA)
    • (Optionally RTEMS, µcLinux)
    linkedin facebook twitter