{"id":9663,"date":"2025-08-31T10:56:56","date_gmt":"2025-08-31T07:56:56","guid":{"rendered":"https:\/\/www.itopya.com\/blog\/?p=9663"},"modified":"2025-08-31T10:56:56","modified_gmt":"2025-08-31T07:56:56","slug":"nvidia-genel-robotik-ve-fiziksel-yapay-zeka-icin-gercek-zamanli-akil-yurutmenin-kilidini-aciyor","status":"publish","type":"post","link":"https:\/\/www.itopya.com\/blog\/nvidia-genel-robotik-ve-fiziksel-yapay-zeka-icin-gercek-zamanli-akil-yurutmenin-kilidini-aciyor\/","title":{"rendered":"NVIDIA Genel Robotik ve Fiziksel Yapay Zeka i\u00e7in Ger\u00e7ek Zamanl\u0131 Ak\u0131l Y\u00fcr\u00fctmenin Kilidini A\u00e7\u0131yor"},"content":{"rendered":"

NVIDIA Genel Robotik ve Fiziksel Yapay Zeka i\u00e7in Ger\u00e7ek Zamanl\u0131 Ak\u0131l Y\u00fcr\u00fctmenin Kilidini A\u00e7\u0131yor<\/strong><\/h3>\n

Fiziksel yapay zek\u00e2 geli\u015ftiricileri, NVIDIA Jetson Thor<\/strong> mod\u00fcllerini robotlar\u0131na entegre ettik\u00e7e, d\u00fcnya genelinde robotlar \u00e7ok daha ak\u0131ll\u0131 h\u00e2le gelecek. Jetson Thor, ara\u015ft\u0131rma ve end\u00fcstrideki robotik sistemler i\u00e7in bir \u201cbeyin\u201d g\u00f6revi g\u00f6recek yeni nesil robotik bilgisayar platformu olarak tasarland\u0131.
\nRobotlar\u0131n ba\u015far\u0131s\u0131, zengin sens\u00f6r verilerini h\u0131zl\u0131 i\u015fleyebilme<\/strong> ve d\u00fc\u015f\u00fck gecikmeli yapay zek\u00e2 \u00e7\u0131kar\u0131m\u0131<\/strong> yapabilme kabiliyetine ba\u011fl\u0131. Jetson Thor, tam da bu ihtiyaca yan\u0131t veriyor.<\/p>\n

\u0130nsans\u0131 Robotlarda Yeni Bir \u00c7a\u011f<\/strong><\/h3>\n
    \n
  • Agility Robotics<\/strong>: \u015eirket, Jetson platformunu be\u015finci nesil Digit<\/strong> robotuna entegre etti. Alt\u0131nc\u0131 nesilde ise Jetson Thor\u2019u yerle\u015fik beyin olarak kullanmay\u0131 planl\u0131yor. Bu ge\u00e7i\u015f sayesinde Digit, daha geli\u015fmi\u015f alg\u0131lama, karar verme ve karma\u015f\u0131k g\u00f6revlerde ger\u00e7ek zamanl\u0131 ak\u0131l y\u00fcr\u00fctme kabiliyeti kazanacak.
    \nCEO Peggy Johnson, \u201cJetson Thor\u2019un g\u00fc\u00e7l\u00fc u\u00e7 i\u015flem yetenekleri Digit\u2019i bir \u00fcst seviyeye ta\u015f\u0131yacak\u201d diyerek, lojistik ve fabrika operasyonlar\u0131n\u0131n optimize edilece\u011fini vurgulad\u0131.<\/li>\n
  • Boston Dynamics<\/strong>: 30 y\u0131l\u0131 a\u015fk\u0131n s\u00fcredir en geli\u015fmi\u015f robotlar\u0131 geli\u015ftiren \u015firket, Atlas<\/strong> insans\u0131 robotuna Jetson Thor\u2019u entegre ediyor. B\u00f6ylece Atlas, sunucu d\u00fczeyindeki i\u015flem g\u00fcc\u00fcn\u00fc ve y\u00fcksek bant geni\u015flikli veriyi do\u011frudan cihaz \u00fczerinde kullanabilecek.<\/li>\n<\/ul>\n

    \"Nvidia<\/h3>\n

    Jetson Thor\u2019un G\u00fcc\u00fc: Orin\u2019den 7,5 Kat Daha Fazla AI Performans\u0131<\/strong><\/h3>\n

    Robotik uygulamalarda ger\u00e7ek zamanl\u0131 ak\u0131l y\u00fcr\u00fctme, \u00e7oklu sens\u00f6rlerden gelen e\u015fzamanl\u0131 veri ak\u0131\u015flar\u0131n\u0131 i\u015fleyebilecek y\u00fcksek i\u015flem g\u00fcc\u00fc ve bellek<\/strong> gerektiriyor.
    \nYeni Jetson Thor, \u00f6nc\u00fcl\u00fc Jetson Orin<\/strong>\u2019e g\u00f6re:<\/p>\n

      \n
    • 7,5 kat daha fazla yapay zek\u00e2 i\u015flem g\u00fcc\u00fc<\/strong>,<\/li>\n
    • 3,1 kat daha y\u00fcksek CPU performans\u0131<\/strong>,<\/li>\n
    • 2 kat daha fazla bellek<\/strong> sunuyor.<\/li>\n<\/ul>\n

      Bu sayede robotik\u00e7iler, y\u00fcksek h\u0131zl\u0131 sens\u00f6r verilerini i\u015fleyebilecek, u\u00e7ta g\u00f6rsel ak\u0131l y\u00fcr\u00fctme ger\u00e7ekle\u015ftirebilecek ve daha \u00f6nce \u00e7ok yava\u015f oldu\u011fu i\u00e7in pratikte kullan\u0131lamayan i\u015f ak\u0131\u015flar\u0131n\u0131 hayata ge\u00e7irebilecek. Bu at\u0131l\u0131m, \u00f6zellikle insans\u0131 robotik<\/strong> gibi \u00e7ok modlu yapay zek\u00e2 alanlar\u0131nda yepyeni imk\u00e2nlar do\u011furuyor.<\/p>\n

      Geni\u015f Kullan\u0131m Alanlar\u0131<\/strong><\/h3>\n

      Jetson Thor, yaln\u0131zca insans\u0131 robotlarda de\u011fil, bir\u00e7ok farkl\u0131 alanda d\u00f6n\u00fc\u015f\u00fcm yaratacak:<\/p>\n

        \n
      • Cerrahi Asistanlar<\/strong>: Ameliyat s\u0131ras\u0131nda cerrahlara rehberlik eden yapay zek\u00e2 sistemleri, \u00e7oklu sens\u00f6r giri\u015flerini ger\u00e7ek zamanl\u0131 olarak i\u015fleyebilir.<\/li>\n
      • Ak\u0131ll\u0131 Tar\u0131m<\/strong>: Trakt\u00f6rler, karma\u015f\u0131k tar\u0131msal veriyi u\u00e7ta analiz ederek daha verimli ekim ve hasat s\u00fcre\u00e7leri y\u00fcr\u00fctebilir.<\/li>\n
      • Lojistik ve Teslimat Robotlar\u0131<\/strong>: Y\u00fckleme, istifleme ve teslimat gibi g\u00f6revleri h\u0131zla ve ba\u011f\u0131ms\u0131z \u015fekilde ger\u00e7ekle\u015ftirebilir.<\/li>\n
      • End\u00fcstriyel Manip\u00fclat\u00f6rler<\/strong>: Fabrikalarda montaj ve \u00fcretim hatlar\u0131nda daha karma\u015f\u0131k g\u00f6revleri \u00fcstlenebilir.<\/li>\n
      • G\u00f6rsel Yapay Zek\u00e2 Ajanlar\u0131<\/strong>: \u00c7oklu kamera ak\u0131\u015flar\u0131n\u0131 i\u015fleyerek g\u00fcvenlik, \u00fcretim ve sa\u011fl\u0131k alanlar\u0131nda ger\u00e7ek zamanl\u0131 kararlar alabilir.<\/li>\n<\/ul>\n

        Ger\u00e7ek Zamanl\u0131 Ak\u0131l Y\u00fcr\u00fctme: Yeni Nesil Fiziksel AI<\/strong><\/h3>\n

        Jetson Thor, \u00fcretken ak\u0131l y\u00fcr\u00fctme modelleri<\/strong> i\u00e7in tasarland\u0131.<\/p>\n

          \n
        • B\u00fcy\u00fck dil modelleri (LLM), g\u00f6rme-dil (VLM) ve g\u00f6rme-dil-eylem (VLA) modellerini bulut ba\u011f\u0131ml\u0131l\u0131\u011f\u0131n\u0131 azaltarak u\u00e7ta \u00e7al\u0131\u015ft\u0131rabiliyor.<\/li>\n
        • Cosmos Reason, DeepSeek, Llama, Gemini, Qwen gibi pop\u00fcler modelleri ve Isaac GR00T N1.5<\/strong> gibi robotik odakl\u0131 \u00f6zel modelleri destekliyor.<\/li>\n
        • Jetson yaz\u0131l\u0131m y\u0131\u011f\u0131n\u0131, d\u00fc\u015f\u00fck gecikmeli ve y\u00fcksek performansl\u0131 i\u015fleme i\u00e7in optimize edilmi\u015f durumda.<\/li>\n<\/ul>\n

          B\u00f6ylece geli\u015ftiriciler, insans\u0131 robotlardan ak\u0131ll\u0131 ameliyathanelere kadar \u00e7ok farkl\u0131 senaryolarda yerel olarak h\u0131zl\u0131 denemeler<\/strong> yapabiliyor ve \u00e7\u0131kar\u0131m alabiliyor.<\/p>\n

          Yaz\u0131l\u0131m Ekosistemi ile Tam Entegrasyon<\/strong><\/h3>\n

          Jetson Thor, NVIDIA AI ekosisteminin tamam\u0131yla uyumlu \u00e7al\u0131\u015f\u0131yor:<\/p>\n

            \n
          • Isaac<\/strong> \u2192 Robotik geli\u015ftirme ve sim\u00fclasyon,<\/li>\n
          • Metropolis<\/strong> \u2192 G\u00f6rsel AI ve video analiti\u011fi,<\/li>\n
          • Holoscan<\/strong> \u2192 Sens\u00f6r i\u015fleme ve veri ak\u0131\u015f\u0131.<\/li>\n<\/ul>\n

            Bu entegrasyon sayesinde geli\u015ftiriciler, \u00e7al\u0131\u015fan g\u00fcvenli\u011fini izleyen kamera sistemlerinden, yap\u0131land\u0131r\u0131lmam\u0131\u015f ortamlarda g\u00f6rev yapabilen insans\u0131 robotlara kadar \u00e7ok \u00e7e\u015fitli \u00e7\u00f6z\u00fcmler \u00fcretebiliyor.<\/p>\n

            Ara\u015ft\u0131rma D\u00fcnyas\u0131nda Jetson Thor<\/strong><\/h3>\n
              \n
            • Stanford<\/strong>, Carnegie Mellon<\/strong> ve Z\u00fcrih \u00dcniversitesi<\/strong> gibi \u00f6nde gelen ara\u015ft\u0131rma merkezleri Jetson Thor\u2019u kullanmaya ba\u015flad\u0131.<\/li>\n
            • Carnegie Mellon\u2019daki AirLab ekibi, arama-kurtarma ve t\u0131bbi triyaj g\u00f6revlerinde otonom robotlar\u0131 g\u00fc\u00e7lendirmek i\u00e7in Jetson tabanl\u0131 \u00e7\u00f6z\u00fcmler geli\u015ftiriyor.<\/li>\n
            • Prof. Sebastian Scherer, Jetson Thor\u2019un bilgisayarl\u0131 g\u00f6rme i\u015f y\u00fcklerinde devrim yaratt\u0131\u011f\u0131n\u0131, art\u0131k \u00e7ok daha h\u0131zl\u0131 ve incelikli g\u00f6revlerin m\u00fcmk\u00fcn h\u00e2le geldi\u011fini belirtiyor.<\/li>\n<\/ul>\n

              Jetson Thor Geli\u015ftirici Kiti<\/strong><\/h3>\n

              Jetson Thor ailesi, bir geli\u015ftirici kiti ve \u00fcretim mod\u00fcllerinden olu\u015fuyor.<\/p>\n

                \n
              • Geli\u015ftirici Kiti<\/strong>: Jetson T5000 mod\u00fcl\u00fc, ba\u011flant\u0131 noktalar\u0131 zengin bir ta\u015f\u0131y\u0131c\u0131 kart, fanl\u0131 aktif so\u011futucu ve g\u00fc\u00e7 kayna\u011f\u0131n\u0131 i\u00e7eriyor.<\/li>\n
              • \u00dcretim Mod\u00fclleri<\/strong>: End\u00fcstriyel \u00f6l\u00e7ekte da\u011f\u0131t\u0131mlar i\u00e7in optimize edilmi\u015f.<\/li>\n<\/ul>\n

                Bu yap\u0131, hem ara\u015ft\u0131rmac\u0131lar hem de end\u00fcstriyel geli\u015ftiriciler i\u00e7in h\u0131zl\u0131 prototipleme ve geni\u015f \u00f6l\u00e7ekli uygulamalara kap\u0131 a\u00e7\u0131yor.<\/p>\n

                Sonu\u00e7: <\/strong>NVIDIA Jetson Thor<\/strong>, robotik d\u00fcnyas\u0131nda yeni bir d\u00f6nemin ba\u015flang\u0131c\u0131n\u0131 simgeliyor. Art\u0131k robotlar, bulut ba\u011f\u0131ml\u0131l\u0131\u011f\u0131ndan kurtularak u\u00e7ta, ger\u00e7ek zamanl\u0131 ak\u0131l y\u00fcr\u00fctme<\/strong> yapabilecek. \u0130nsans\u0131 robotlardan t\u0131bbi asistanlara, ak\u0131ll\u0131 tar\u0131mdan lojisti\u011fe kadar geni\u015f bir alanda Jetson Thor, fiziksel yapay zek\u00e2n\u0131n s\u0131n\u0131rlar\u0131n\u0131 ileri ta\u015f\u0131yor.<\/p>\n

                Robotlar\u0131n gelece\u011fi art\u0131k \u00e7ok daha zeki, h\u0131zl\u0131 ve \u00f6zerk.<\/p>\n","protected":false},"excerpt":{"rendered":"

                NVIDIA Genel Robotik ve Fiziksel Yapay Zeka i\u00e7in Ger\u00e7ek Zamanl\u0131 Ak\u0131l Y\u00fcr\u00fctmenin Kilidini A\u00e7\u0131yor Fiziksel yapay zek\u00e2 geli\u015ftiricileri, NVIDIA Jetson Thor mod\u00fcllerini robotlar\u0131na entegre ettik\u00e7e, d\u00fcnya genelinde robotlar \u00e7ok daha ak\u0131ll\u0131 h\u00e2le gelecek. Jetson Thor, ara\u015ft\u0131rma ve end\u00fcstrideki robotik sistemler i\u00e7in bir \u201cbeyin\u201d g\u00f6revi g\u00f6recek yeni nesil robotik bilgisayar platformu olarak tasarland\u0131. Robotlar\u0131n ba\u015far\u0131s\u0131, zengin …<\/p>\n<\/div>","protected":false},"author":1,"featured_media":9664,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1,83],"tags":[],"class_list":["post-9663","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-genel","category-haberler"],"_links":{"self":[{"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/posts\/9663","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/comments?post=9663"}],"version-history":[{"count":1,"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/posts\/9663\/revisions"}],"predecessor-version":[{"id":9666,"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/posts\/9663\/revisions\/9666"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/media\/9664"}],"wp:attachment":[{"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/media?parent=9663"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/categories?post=9663"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.itopya.com\/blog\/wp-json\/wp\/v2\/tags?post=9663"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}