The goal of ULPEC is to demonstrate a microsystem that is natively brain-inspired, connecting an event-based camera to a dedicated spiking neural network enabled by memristive synapses. The memristive synapses of the neural network are based on ferroelectric tunnel junctions (FTJ).
As TSST, we develop and deliver functional memristor devices on silicon wafers based on these ferroelectric tunnel junctions (FTJs), ready for flip chip integration with CMOS. The targeted user case for ULPEC technologies is the vision and recognition of traffic events (signs, obstacles like other cars, persons, etc.) which are part of the major disruption of autonomous and computer assisted driving in the transport and car manufacturing industry.
More information about ULPEC can be found on:
Website ULPEC Project
Currently TSST is involved in the Horizon 2020 “Ultra-Low Power Event-Based Camera” (ULPEC) project (Project ID: 732642), running since January 2017.
Other partners involved are:
• Chronocam S.A.S.
• IBM Research GMBH
• Robert Bosch GMBH
• University of Twente
• Université Pierre et Marie Curie
• University of Bordeaux