{"id":3518,"date":"2025-09-15T10:15:46","date_gmt":"2025-09-15T08:15:46","guid":{"rendered":"https:\/\/irtnanoelec.fr\/?page_id=3518"},"modified":"2026-04-03T06:52:04","modified_gmt":"2026-04-03T04:52:04","slug":"increasingly-involved-in-embedded-artificial-intelligence","status":"publish","type":"page","link":"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/","title":{"rendered":"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE"},"content":{"rendered":"
\n
\n
\n
\n \"\" <\/div>\n
\n

L\u2019IRT Nanoelec aborde le sujet de l\u2019intelligence artificielle (IA) sous l\u2019angle de l\u2019\u00e9lectronique embarqu\u00e9e. Le d\u00e9veloppement et la g\u00e9n\u00e9ralisation d\u2019applications d\u2019IA embarqu\u00e9es s\u2019imposent pour la gestion de grandes quantit\u00e9s de donn\u00e9es qui n\u00e9cessitent un traitement local au niveau des composants et du syst\u00e8me. L\u2019objectif est de donner \u00e0 ces composants la capacit\u00e9 de prendre des d\u00e9cisions de mani\u00e8re plus d\u00e9centralis\u00e9e, autonome et fiable.<\/p>\n

 <\/p>\n<\/div> <\/div>\n \n \n <\/div>\n <\/div>\n <\/section>\n

\n
\n
\n
\n

<\/span>Imageurs intelligents<\/h2>

L\u2019imagerie est l\u2019un des trois principaux domaines qui utilisent l\u2019IA. L\u2019objectif n\u2019est pas seulement d\u2019obtenir une meilleure qualit\u00e9 d\u2019image, mais aussi d\u2019extraire des donn\u00e9es pertinentes de l\u2019image, en tenant compte de l\u2019environnement, de l\u2019objet et de la sc\u00e8ne (potentiellement dans diverses conditions d\u2019\u00e9clairage), et de la connaissance du contexte. L\u2019objectif principal du\u00a0programme Nanoelec\/Smart Imager<\/a>\u00a0est d\u2019\u00e9valuer les avantages de l\u2019utilisation de la technologie 3D-stack pour int\u00e9grer le traitement de l\u2019intelligence artificielle dans la troisi\u00e8me couche d\u2019un capteur d\u2019image. Les \u00e9quipes de recherche se concentrent sur le d\u00e9veloppement de blocs de construction g\u00e9n\u00e9riques pour l\u2019IA et le traitement associ\u00e9 ainsi que sur l\u2019exploration de l\u2019impact de ces blocs sur des architectures d\u2019imageurs \u00e0 basse \u00e9nergie.<\/p>\n<\/div> <\/div>\n \n \n <\/div>\n <\/div>\n <\/section>\n

\n
\n
\n
\n

<\/span>Cybers\u00e9curit\u00e9 des microcontroleurs<\/h2>

Les futurs objets connect\u00e9s auront besoin d\u2019impl\u00e9mentations fiables d\u2019algorithmes d\u2019IA embarqu\u00e9s et de m\u00e9canismes de s\u00e9curit\u00e9 pour les prot\u00e9ger des menaces logicielles et mat\u00e9rielles potentielles. Le d\u00e9ploiement et l\u2019utilisation s\u00fbre de cette technologie dans les syst\u00e8mes embarqu\u00e9s requiert de s\u00e9curiser la mise en \u0153uvre d\u2019algorithmes d\u2019IA au sein de ces objets ainsi que leurs protocoles de communications avec le monde ext\u00e9rieur. Des chercheurs de laboratoires acad\u00e9miques et des \u00e9quipes industrielles travaillent ensemble au sein du\u00a0programme Nanoelec\/Pulse<\/a>\u00a0pour r\u00e9aliser la mise en \u0153uvre et le d\u00e9ploiement d\u2019algorithmes d\u2019IA, en particulier d\u2019algorithmes d\u2019apprentissage automatique embarqu\u00e9s dans des composants IoT. Concernant la s\u00e9curit\u00e9 des v\u00e9hicules autonomes, les \u00e9quipes du programme s\u2019int\u00e9ressent \u00e0 la validation de syst\u00e8mes d\u2019intelligence artificielle (IA) pour la mobilit\u00e9\/les v\u00e9hicules. La question de l\u2019\u00e9valuation et de la validation de l\u2019efficacit\u00e9 de ces technologies est l\u2019un des derniers enjeux centraux retardant l\u2019adoption des technologies embarqu\u00e9es de capture de contexte. C\u2019est donc un enjeu majeur pour l\u2019acc\u00e8s au march\u00e9 des v\u00e9hicules autonomes.<\/p>\n<\/div> <\/div>\n \n \n <\/div>\n <\/div>\n <\/section>\n

\n
\n
\n
\n

<\/span>Cytom\u00e9trie<\/h2>

Les principaux partenaires du\u00a0programme Nanoelec\/Photonic Sensors<\/a>\u00a0participent au projet europ\u00e9en Neoteric, qui vise \u00e0 concevoir et construire un d\u00e9monstrateur de cytom\u00e9trie utilisant des circuits photoniques. La cytom\u00e9trie est une technique de comptage et de caract\u00e9risation de particules, mol\u00e9cules ou cellules biologiques dans un fluide. Neoteric consiste \u00e0 d\u00e9montrer qu\u2019un circuit photonique avec des fonctions d\u2019apprentissage (IA) permet de mieux r\u00e9aliser la reconnaissance d\u2019images de particules compt\u00e9es. Par rapport \u00e0 l\u2019analyse d\u2019images conventionnelle utilisant un logiciel d\u2019apprentissage en profondeur, la technologie photonique sur silicium promet une augmentation de la fr\u00e9quence d\u2019images, des performances de classification am\u00e9lior\u00e9es et une consommation d\u2019\u00e9nergie r\u00e9duite de plusieurs ordres de grandeur.<\/p>\n<\/div> <\/div>\n \n \n <\/div>\n <\/div>\n <\/section>\n

\n
\n
\n
\n

<\/span>Traitements d\u2019images pour la caract\u00e9risation<\/h2>

Le d\u00e9veloppement d\u2019algorithmes d\u2019apprentissage profond et l\u2019utilisation de l\u2019intelligence artificielle pr\u00e9sente int\u00e9resse beaucoup les \u00e9quipes sp\u00e9cialis\u00e9es dans la caract\u00e9risation des composants \u00e9lectroniques et syst\u00e8mes. En particulier, ces technologies permettront d\u2019am\u00e9liorer l\u2019efficacit\u00e9 des exp\u00e9riences men\u00e9es avec les grands instruments disponibles dans le cadre de la\u00a0plateforme Nanoelec\/PAC-G<\/a>. Ces exp\u00e9riences g\u00e9n\u00e8rent de tr\u00e8s grands ensembles de donn\u00e9es. Le traitement mutualis\u00e9 des donn\u00e9es, m\u00e9tadonn\u00e9es, simulations et images utilisant l\u2019IA dans les images associ\u00e9es peut acc\u00e9l\u00e9rer la convergence vers le d\u00e9veloppement de mat\u00e9riaux pour r\u00e9duire l\u2019empreinte environnementale, assurer la souverainet\u00e9 de l\u2019approvisionnement et la fiabilit\u00e9 op\u00e9rationnelle.<\/p>\n<\/div> <\/div>\n \n \n <\/div>\n <\/div>\n <\/section>\n

\n
\n
\n
\n

<\/span>D\u00e9monstrateurs applicatifs pour l\u2019imagerie<\/h2>

Les \u00e9quipes impliqu\u00e9es dans\u00a0l\u2019initiative Nanoelec\/SystemLab<\/a>\u00a0d\u00e9finissent, d\u00e9veloppent et testent des cas d\u2019usage en utilisant des technologies d\u2019IA embarqu\u00e9es int\u00e9gr\u00e9es ou associ\u00e9es \u00e0 des capteurs d\u2019images. Le projet vise \u00e0 d\u00e9velopper des d\u00e9monstrateurs fonctionnels et \u00e0 explorer de nouveaux sc\u00e9narios d\u2019applications. Un de ces cas d\u2019utilisation consiste \u00e0 positionner des capteurs d\u2019images sans fil dans divers milieux naturels pour surveiller les incendies ou mesurer la biodiversit\u00e9, par exemple. Un autre sujet porte sur l\u2019\u00e9tude des capteurs d\u2019images capables d\u2019analyser les postures et les expressions humaines, dans le but de d\u00e9tecter les \u00e9motions, pour pr\u00e9venir des situations dangereuses ou des comportements inappropri\u00e9s. Et un troisi\u00e8me cas d\u2019utilisation implique la d\u00e9tection d\u2019obstacles, tels que des nids-de-poule et des plaques de glace pour les v\u00e9hicules.<\/p>\n<\/div> <\/div>\n \n \n <\/div>\n <\/div>\n <\/section>\n

\n
\n
\n
\n

<\/span>Enseigner, \u00e9tudier l\u2019intelligence artificiel gr\u00e2ce \u00e0 des composants Risc-V<\/h2>

Le\u00a0programme Nanoelec\/Capital Humain et ing\u00e9nierie de formation<\/a>\u00a0a d\u00e9bouch\u00e9, en 2020, sur une nouvelle formation bas\u00e9e sur la co-conception logiciel\/mat\u00e9riel (hardware\/software<\/em>), gr\u00e2ce \u00e0 l\u2019utilisation d\u2019un processeur bas\u00e9 sur le jeu d\u2019instructions Risc-V. Ce jeu d\u2019instructions et les processeurs qui l\u2019impl\u00e9mentent sont libres de droits et sont de plus en plus utilis\u00e9s par l\u2019industrie dans le monde entier. En 2020, l\u2019\u00e9quipe p\u00e9dagogique de Grenoble-INP, dans le cadre de Nanoelec, a d\u00e9velopp\u00e9 une plateforme de r\u00e9f\u00e9rence semi-g\u00e9n\u00e9rique bas\u00e9e sur Risc-V, con\u00e7u un syst\u00e8me de vision embarqu\u00e9 sur FPGA et un FPGA pour le traitement mat\u00e9riel d\u2019algorithmes d\u2019intelligence artificielle. La co-conception logiciel\/mat\u00e9riel, jusqu\u2019au prototype proprement dit, permet aux \u00e9tudiants d\u2019exp\u00e9rimenter les avantages respectifs de ces technologies (flexibilit\u00e9 du logiciel, efficacit\u00e9 du mat\u00e9riel) et de trouver un compromis entre les deux, au sein d\u2019un m\u00eame syst\u00e8me.<\/p>\n

 <\/p>\n

 <\/p>\n

 <\/p>\n

Afin d\u2019acc\u00e9l\u00e9rer le transfert de comp\u00e9tences entre la recherche acad\u00e9mique et l\u2019industrie autour de l\u2019IA, les huit IRT ont conduit, en 2019, l\u2019initiative commune\u00a0EngageAI<\/a>.<\/p>\n<\/section>\n<\/div> <\/div>\n \n \n <\/div>\n <\/div>\n <\/section>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"edito.php","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":""},"class_list":["post-3518","page","type-page","status-publish","hentry"],"acf":[],"yoast_head":"\nINCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE - IRT Nanoelec<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE - IRT Nanoelec\" \/>\n<meta property=\"og:url\" content=\"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/\" \/>\n<meta property=\"og:site_name\" content=\"IRT Nanoelec\" \/>\n<meta property=\"article:modified_time\" content=\"2026-04-03T04:52:04+00:00\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"4 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/\",\"url\":\"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/\",\"name\":\"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE - IRT Nanoelec\",\"isPartOf\":{\"@id\":\"https:\/\/irtnanoelec.fr\/en\/#website\"},\"datePublished\":\"2025-09-15T08:15:46+00:00\",\"dateModified\":\"2026-04-03T04:52:04+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/irtnanoelec.fr\/en\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/irtnanoelec.fr\/en\/#website\",\"url\":\"https:\/\/irtnanoelec.fr\/en\/\",\"name\":\"IRT Nanoelec\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\/\/irtnanoelec.fr\/en\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/irtnanoelec.fr\/en\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/irtnanoelec.fr\/en\/#organization\",\"name\":\"IRT Nanoelec\",\"url\":\"https:\/\/irtnanoelec.fr\/en\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/irtnanoelec.fr\/en\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/irtnanoelec.fr\/wp-content\/uploads\/2025\/07\/irt-nanoelec.svg\",\"contentUrl\":\"https:\/\/irtnanoelec.fr\/wp-content\/uploads\/2025\/07\/irt-nanoelec.svg\",\"width\":181,\"height\":89,\"caption\":\"IRT Nanoelec\"},\"image\":{\"@id\":\"https:\/\/irtnanoelec.fr\/en\/#\/schema\/logo\/image\/\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE - IRT Nanoelec","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/","og_locale":"en_US","og_type":"article","og_title":"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE - IRT Nanoelec","og_url":"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/","og_site_name":"IRT Nanoelec","article_modified_time":"2026-04-03T04:52:04+00:00","twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"4 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/","url":"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/","name":"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE - IRT Nanoelec","isPartOf":{"@id":"https:\/\/irtnanoelec.fr\/en\/#website"},"datePublished":"2025-09-15T08:15:46+00:00","dateModified":"2026-04-03T04:52:04+00:00","breadcrumb":{"@id":"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/irtnanoelec.fr\/en\/increasingly-involved-in-embedded-artificial-intelligence\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/irtnanoelec.fr\/en\/"},{"@type":"ListItem","position":2,"name":"INCREASINGLY INVOLVED IN EMBEDDED ARTIFICIAL INTELLIGENCE"}]},{"@type":"WebSite","@id":"https:\/\/irtnanoelec.fr\/en\/#website","url":"https:\/\/irtnanoelec.fr\/en\/","name":"IRT Nanoelec","description":"","publisher":{"@id":"https:\/\/irtnanoelec.fr\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/irtnanoelec.fr\/en\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/irtnanoelec.fr\/en\/#organization","name":"IRT Nanoelec","url":"https:\/\/irtnanoelec.fr\/en\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/irtnanoelec.fr\/en\/#\/schema\/logo\/image\/","url":"https:\/\/irtnanoelec.fr\/wp-content\/uploads\/2025\/07\/irt-nanoelec.svg","contentUrl":"https:\/\/irtnanoelec.fr\/wp-content\/uploads\/2025\/07\/irt-nanoelec.svg","width":181,"height":89,"caption":"IRT Nanoelec"},"image":{"@id":"https:\/\/irtnanoelec.fr\/en\/#\/schema\/logo\/image\/"}}]}},"_links":{"self":[{"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/pages\/3518","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/comments?post=3518"}],"version-history":[{"count":5,"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/pages\/3518\/revisions"}],"predecessor-version":[{"id":5155,"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/pages\/3518\/revisions\/5155"}],"wp:attachment":[{"href":"https:\/\/irtnanoelec.fr\/en\/wp-json\/wp\/v2\/media?parent=3518"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}