LANGUAGE
  • en
  • sv
  • 13 July, 2016

    Unibap announces TRL 9 heterogeneous processing and imaging solutions for space

    Unibap successfully demonstrates mission critical heterogeneous computing and high resolution imaging in space. Unibap’s heterogeneous computing and imaging solutions has been successfully deployed in a Low Earth Orbit (LEO) on May 30th 2016 with successful commissioning by our undisclosed customer.

    After more than one month of operations on multiple spacecraft we declare the Unibap e2055 product at NASA definition Technology Readiness Level (TRL) 9.

    – It is always a pleasure to return to space with flight hardware. This particular flight is of big importance to our customer and us, and generally for the aerospace community, since we have successfully demonstrated TRL-9 information processing capability above 100 GFLOP. This will enable many future important applications such as intelligent vision processing and control for autonomous Lunar and Mars vehicles as humans start to colonize other interplanetary bodies. It also holds great promise for our 3rd generation processing modules which will reach over 1000 GFLOP when they are released later this year, says Dr. Fredrik Bruhn, CEO of Unibap AB.

    The supplied on-board processing solution comes with a tailored carrier board with 4 TB of storage and provides redundant interfaces and connectivity, of which Ethernet, Controller Area Network (CAN), LVDS, SERDES, SPI, and RS232 are the major ones. However, it is also possible to include an optional SpaceWire/SpaceFibre offering.

    Included on the same flight is a Unibap tailored high resolution optical imaging board for Earth Observation. The imaging solution provides an industrial CMOS sensor capable of tracking up to 32 individual objects in real time at a data output speed of 10 Gbps.

    Unibap provides a complete open source software solution for the heterogeneous computing product including commonly used applications like:

    • Open Computer Vision library, OpenCV 3.1 GPU accelerated and optimized for the AMD SOC
    • Caffe, Deep learning framework, GPU accelerated and optimized for the AMD SOC
    • Robot Operating System, ROS 1.11.16 Jade
    • Point Cloud Library
    • Unibap IVS SDK core infrastucture
    linkedin facebook twitter