A new technical paper titled “Split Additive Manufacturing for Printed Neuromorphic Circuits” was published by researchers at Karlsruher Institut für Technologie (KIT).
“Printed and flexible electronics promises smart devices for application domains, such as smart fast moving consumer goods and medical wearables, which are generally untouchable by conventional rigid silicon technologies. These devices are characterized by their unique properties like flexibility, non-toxic material, and low-cost per square inch. Combining printed neuromorphics circuits with neuromorphic computing is a solution that can be attractive for these application domains. Additive printing technologies, in particular, can reduce a large amount of fabrication complexity and costs. One hand, additive printing processes that have a high-throughput, such as roll to roll printing, reduce the fabrication cost and time for each device. Jet-printing, on the other hand can offer point-of use customization but at the cost of a lower fabrication throughput. In this paper, we present a machine-learning based design framework that takes into account the physical and objective constraints of split additive manufacture for printed neuromorphics circuits. The proposed framework allows multiple printed neural networks to be trained together with the goal of sensibly combining multiple fabrication techniques, such as roll-to-roll printing and jet-printing. This should lead to a cost-effective fabrication of multiple different printed neuromorphic circuits and achieve high fabrication throughput, lower cost, and point-of-use customization.”
You can find the technical document here. Preprint published April 2023.
Zhao, Haibin. Michael Hefenbrock. Michael Beigl. Mehdi Tahoori. “Split Additive Manufacturing for Printed Neuromorphic Circuits.” In 2023 Design, Automation & Test in Europe Conference & Exhibition (DATE). 2023.