GT1I-302ED: MACHINE LEARNING APPLICATION BENCHMARKING ON COTS INFERENCE PROCESSORS
17, July 2020

ESA Open Invitation to Tender AO10370
Open Date: 16/07/2020
Closing Date: 24/09/2020 13:00:00

Status: ISSUED
Reference Nr.: 20.1ED.06
Prog. Ref.: GSTP Element 1 Dev
Budget Ref.: E/0904-611 – GSTP Element 1 Dev
Special Prov.: DE
Tender Type: C
Price Range: > 500 KEURO
Products: Satellites & Probes / On-board Data Management / On Board Data Management / Other
Technology Domains: Onboard Data Systems / Onboard Data Management / Onboard Networks and Control/Monitoring
Establishment: ESTEC
Directorate: Directorate of Tech, Eng. & Quality
Department: Electrical Department
Division: Data Syst & Microelectronics Division
Contract Officer: Singer, Anze
Industrial Policy Measure: N/A – Not apply
Last Update Date: 16/07/2020
Update Reason: Tender issue

The performance of the algorithms assessed during this activity will enable practical usage of the selected processors on board of satellites for Earth Observation (EO) data processing and Vision Based Navigation (VBN) tasks, as well as typical generic platform control applications.The activity target is to build fair and useful benchmarks for measuring training and inference performance of Machine Learning (ML) hardware, software, and services. A widely accepted benchmark suite will benefit the entire community, including researchers, developers, hardware manufacturers, builders of machine learning frameworks, cloud service providers, application providers, and end users.Among targeted platforms there are: – COTS FPGA from XILINX, using machine learning HLS tools – GPUs and Tensor processors (NVIDIA, Intel) – Intel Myriad2 and X processors. – Space grade processors and FPGA (Xilinx, BRAVE, GR740)These algorithms are selected to be representative for tasks that are needed for future space missions (e.g. EO image processing, VBN, Super-resolution, Neural Network-based image classification for payload application, but also non-linear control and time series analysis for platform applications).

If you wish to access the documents related to the Invitation to Tender, you have to log in to the ESA Portal.