Skip to main content
Aller à la page d’accueil de la Commission européenne (s’ouvre dans une nouvelle fenêtre)
français fr
CORDIS - Résultats de la recherche de l’UE
CORDIS

Brain-inspired Light Sensory AI computing using nanophotonic sensory spiking 3D neuron chips

Objectif

It is known that even a simple organism like the Jellyfish, with only a thousand spiking neurons, can learn and adapt to unpredictable environments. Despite attempts in electronics, photonics, 2D materials and other approaches to emulate spiking neurons, a single biological neuron is a thousand times more complex than any artificial neuron. Thus, increasing the complexity of a single neuron leads to a substantial increase in both the size and energy cost of the neuron chip. This raises a fundamental question: can nanoscale physical systems embody the diversity and efficiency of biological nonlinear spiking?

In LightSensorAI, I will push the frontiers of brain-inspired science and will address this challenge by exploring a new paradigm of III-V semiconductor nanophotonic neurons down to 100 nanometer in size, while retaining their advanced nonlinear spiking properties with ultralow energy expenditure. First, I will create light-emitting and detecting nanopillar neurons, featuring i) quantum resonant tunneling to enable negative differential resistance for neuron functions, and ii) charge storage effect for memory–central to provide both programmable spiking and synaptic heterogeneous dynamic functions within a single element. Then, I will scale nanopillar neurons to form compact 2D pixel nanoarrays and connect in 3D these arrays with a light field enabled by on-chip integrated meta-optics, essential to emulate sensing and spike computing in networks. I will validate the 3D neural nanosystem for spatio-temporal spike-based learning of dynamic complex visual scenes (videos). This platform is uniquely suited for power-efficient nanosensors and brain-like processors to enable in-sensor edge computation in unpredictable environments, just as biological systems already do. It will revolutionize nanoscale hardware using energy-efficient and programmable dynamical sensory spiking neurons and will contribute to our understanding of visual sensory and spiking neural activity.

Champ scientifique (EuroSciVoc)

Ce projet n'a pas encore été classé par EuroSciVoc.
Proposez les domaines scientifiques qui vous semblent les plus pertinents et aidez-nous à améliorer notre service de classification.

Mots‑clés

Programme(s)

Thème(s)

Régime de financement

HORIZON-ERC - HORIZON ERC Grants

Voir tous les projets financés dans le cadre de ce programme de financement

Appel à propositions

(s’ouvre dans une nouvelle fenêtre) ERC-2025-ADG

Voir tous les projets financés au titre de cet appel

Institution d’accueil

INTERNATIONAL IBERIAN NANOTECHNOLOGY LABORATORY
Contribution nette de l'UE
€ 2 493 487,00
Coût total
Aucune donnée

Bénéficiaires (1)