The project, which began in January 2016, looked to develop high performance, energy-efficient systems for vision-based image processing applications. The Tulipp project was funded with almost €4 million by Horizon 2020, a European Union research and innovation programme.
The winding up of the Tulipp project also brings the launch of a reference platform for vision-based embedded system designers, that will allow products to be developed that address the combined challenges of low power, low latency, high performance and real-time image processing. The reference platform includes a development kit, that includes an FPGA-based embedded, multicore computing board, parallel real-time operating system and development toolchain with guidelines, coupled with ‘real world’ Use Cases.
Developed by Sundance Multiprocessor Technology, the Tulipp processing platform measures 40mm x 50mm and is PC/104 compliant. The hardware platform is based around the multicore Xilinx Zynq Ultrascale+ MPSoC which contains, the Xilinx FinFET+ FPGA, an ARM Cortex-A53 quad-core CPU, an ARM Mali-400 MP2 GPU, and a real-time processing unit (RPU) containing a dual-core ARM Cortex-R5 32-bit real-time processor based on the ARM-v7R architecture. An expansion module (VITA57.1 FMC) allows application-specific boards with different flavours of input and output interfaces to be created.
A parallel, low latency embedded real-time operating system developed by Hipperos accompanies the hardware platform. It has been designed to manage complex multi-threaded embedded applications in a predictable manner. Real-time co-ordination ensures a high frame rate without missing any deadlines or data. Additionally, the Tulipp reference platform has been extended with performance analysis and power measurement features developed by Norges Teknisk-Naturvitenskapelige Universitet (NTNU) and Technische Universität Dresden (TUD) and implemented in the Tulipp STHEM toolchain.
Experts from the Tulipp Consortium have also published a set of guidelines, consisting of practical advice, best practice approaches and recommended implementation methods. These guidelines will become a TULIPP book to be published by Springer by the end of 2019. Tulipp has also developed three ‘real-world’ Use Cases in distinctly diverse application domains – medical X-ray imaging, automotive Advanced Driver Assistance Systems (ADAS) and Unmanned Aerial vehicles (UAVs).