TR2021022147A2 - HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE - Google Patents

HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE

Info

Publication number
TR2021022147A2
TR2021022147A2 TR2021/022147A TR2021022147A TR2021022147A2 TR 2021022147 A2 TR2021022147 A2 TR 2021022147A2 TR 2021/022147 A TR2021/022147 A TR 2021/022147A TR 2021022147 A TR2021022147 A TR 2021022147A TR 2021022147 A2 TR2021022147 A2 TR 2021022147A2
Authority
TR
Turkey
Prior art keywords
robot
algorithm
task
artificial intelligence
new
Prior art date
Application number
TR2021/022147A
Other languages
Turkish (tr)
Inventor
Dereli̇ Serkan
Köker Raşi̇t
Çakar Tarik
Original Assignee
İstanbul Geli̇şi̇m Üni̇versi̇tesi̇
Univ Istanbul Gelisim
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by İstanbul Geli̇şi̇m Üni̇versi̇tesi̇, Univ Istanbul Gelisim filed Critical İstanbul Geli̇şi̇m Üni̇versi̇tesi̇
Priority to TR2021/022147A priority Critical patent/TR2021022147A2/en
Publication of TR2021022147A2 publication Critical patent/TR2021022147A2/en
Priority to PCT/TR2022/050654 priority patent/WO2023128957A1/en

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00Programme-control systems
    • G05B19/02Programme-control systems electric
    • G05B19/18Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
    • G05B19/409Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by using manual data input [MDI] or by using control panel, e.g. controlling functions with the panel; characterised by control panel details or by setting parameters
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J13/00Controls for manipulators
    • B25J13/06Control stands, e.g. consoles, switchboards
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B25HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25JMANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00Programme-controlled manipulators
    • B25J9/16Programme controls
    • B25J9/1656Programme controls characterised by programming, planning systems for manipulators
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00Program-control systems
    • G05B2219/30Nc systems
    • G05B2219/39Robotics, robotics to robotics hand
    • G05B2219/39443Portable, adapted to handpalm, with joystick, function keys, display
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/004Artificial life, i.e. computing arrangements simulating life
    • G06N3/006Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Manufacturing & Machinery (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Numerical Control (AREA)

Abstract

Bu buluş; endüstriyel robotların görev veya görev yeri değişiklerinde fonksiyonlarını etkin bir şekilde sürdürebilmeleri için herhangi bir ortamda, herhangi bir zamanda ve herhangi bir kişi tarafından uzmanlık gerektirmeden eklem açılarına göre yeni konum, pozisyon ve oryantasyon bilgilerine ilişkin simülasyonların hızlı bir şekilde elde edildiği dinamik robot yönlendirmesi için el terminali ve algoritması ile ilgili olup, özelliği; kullanıcının sisteme giriş paneli (1) tarafından robot DH parametrelerinin kaydedildiği veri giriş birimi (10), görevi veya görev yeri değişen robotun yeni görevini gerçekleştireceği eklem açılarının sistem tarafından farklı yapay zekâ algoritmaları tarafından anakart (3) tarafından hesaplandığı ve simülasyonların oluşturulduğu işlem birimi (20) içermesidir.This invention; Handheld terminal for dynamic robot guidance, where simulations of new position, position and orientation information according to joint angles are obtained quickly, in any environment, at any time and by any person, so that industrial robots can continue their functions effectively in task or job changes. and its algorithm, its feature is; The data entry unit (10, where the robot DH parameters are recorded by the user's input panel (1) to the system), the processing unit (20) where the joint angles of the robot whose task or place of duty is to perform its new task are calculated by the motherboard (3) by different artificial intelligence algorithms by the system and simulations are created (20). ) is included.

Description

TARIFNAME DINAMIK ROBOT YÖNLENDIRMESI IÇIN EL TERMINALI VE ALGORITMASI Teknolojik Alan: Bu bulus; endüstriyel robotlarin görev veya görev yeri degisiklerinde fonksiyonlarini etkin bir sekilde sürdürebilmeleri için herhangi bir ortamda, herhangi bir zamanda ve herhangi bir kisi tarafindan uzmanlik gerektirmeden eklem açilarina göre yeni konum, pozisyon ve oryantasyon bilgilerine iliskin simülasyonlarin hizli bir sekilde elde edildigi dinamik robot yönlendirmesi için el terminali ve algoritmasi ile ilgilidir. Teknigin Bilinen Durumu: Mevcut teknikte seri robotlarin yeni pozisyon ve oryantasyon bilgilerine ihtiyaç duyulacagi is hatti veya konum degisikliklerinde eklem açilarinin yeni degerlerinin saptanabilmesi manuel olarak ya da bilgisayar yardimiyla lineer olmayan ve karmasik matematiksel islemler yapmayi gerektirmektedir. Özellikle de robotlar devamli olarak is hatti veya konum degisikligi yapiyorlarsa bu islemler için mühendislerin bu ise ayri bir mesai ayirmasi gerekmekte ve bu da hem is yüklerinin artmasina hem de gereksiz zaman kayiplarina yol açmaktadir. Karmasik islemler bilgisayar yardimiyla gerçeklestirilmek istenense dahi farkli algoritmalar ayri ayri kodlanmakta ve sonuçlar elde edildikten sonra her bir sonuç ayri ayri platformlarda analiz edilmektedir. Daha sonrasinda elde edilen eklem açilarina iliskin çikarimlar ya da tahminler sonucu uygun deger bulunmaya çalisilmaktadir. Tahmin ve çikarima dayali elde edilen pozisyon bilgisi tek bir degerin optimizasyonu sonucu ortaya çikmis olup, daha iyi degerler varsa bile bu degerlerin ek hesaplamalarla ortaya konulmasi gerekmektedir. Böylece sürecin isleyisi gecikmekte ve zaman kaybina sebep olmaktadir. Günümüzde robotlar için yapilan bu hesaplamalarda en iyi sonucu veren degerin kisilerin çikarimlarina bagli olmasi güvenilmezlik yaratmakta ve robotlarin çalisma verimini etkileme ihtimali nedeniyle dezavantajli bir durum ortaya çikarmaktadir. Çünkü hesaplama sürecinde uzmanlarin çalisma prensibine göre islem adimlari veya sonuca ulasma biçimleri degisiklik gösterebilmektedir. Bu durum uzmanlarin ayni islem için farkli sonuçlar elde etmesine sebep olmakla beraber kurumsal yapida standartlasmanin da önüne geçmektedir. Uzmanlar tarafindan ulasilan hesaplamalar dogru dahi olsa robot yeni pozisyon bilgileriyle sahada test edilerek gerekli düzeltmeler yapilmak zorundadir ve bu durum da sürecin gereksiz yere uzamasina sebep olmaktadir. Bu tür problemlerin üstesinden gelmek için literatürde Kinematik Hesaplama Yöntemi" baslikli konu ele alinmaktadir. Mevcut bulus; çok eklemli robot kolda, ters kinematik hesaplama yaparak robot kolunun konumundan ve durusundan eklem açilarini ortaya koymak için bir ters kinematik hesaplama cihazi ve yöntemi ile ilgilidir. Giris cihazi, sayisal hesaplayiei, eklem açisi hesaplayici, kontrolör, yakinsama tespit cihazi ve çikis eihazindan meydana gelen bir ters kinematik hesap birimi ile çok eklemli bir robot kolunda ters kinematik hesaplama gerçeklestirmektedir. CN108427282 numarali basvuruda "Ögrenmeyi Ögretmeye Dayali Robot Için Ters Kinematik Çözüm Yöntemi" anlatilmaktadir. Mevcut bulus, robot ters kinematigini hesaplamak için Gauss karisim modelini kullanan bir makine ögrenme algoritmasiyla çalismakta ve ögrenmeyi ögretmeye dayali bir robot ters kinematik çözüm yöntemi sunmaktadir. Bahsi geçen bulusta; belli bir sayida robot grubunun eklem açisi, bir uç efektörün kartezyen konumu ve Euler açisi toplanarak bir veri seti elde edilir ve bu veri seti optimize edildikten sonra Gauss karisim modeli parametrelerini elde etmek için veri seti yinelemeli olarak hesaplanir. Yukarida ele alinan basvurularda robotlarda ters kinematik hesaplama ile eklem açi degerlerinin elde edilmesine yönelik buluslar ele alinmistir. Söz konusu yöntemlerde robotlarin eklem açilari farkli tekniklerle hesaplanmaktadir. Ancak bu sistemlerde robotlarin eklem açilarina göre yeni konum, pozisyon ve oryantasyon bilgilerinin yer aldigi simülasyonlar yer almamakta ve ters kinematik hesaplama ile eklem açilari belirlenen robotlar görsel bir sekilde tablet tarzi bir el terminalinde canlandirilamamaktadir. Bu durumda dogrulama süreci fazladan bir zaman sartiyatma sebep olmakta ve gereksiz is yükü yaratmaktadir. Dogrulama süreci basarili bir sekilde gerçeklestirilse dahi robotlarin yeni eklem ve açi degerlerine ait pozisyon bilgileriyle sahada test edilmesi ve bu dogrultuda gerekli düzenlemelerin yapilmasi gerekmektedir. Bu nedenle sürecin gereksiz yere uzamasina sebep olmaktadir. Sonuç olarak yukarida bahsedilen dezavantajlarin üstesinden gelebilen ve robotlarin görev veya görev yeri degisiklerinde fonksiyonlarini etkin bir sekilde sürdürebilmeleri için her durum veya her kisi için ayni islemi ayni biçimde gerçeklestiren ve zaman, mekân ve kisiden bagimsiz bir sekilde bulusa entegre olmus ana kart üzerinde gömülü olarak bulunan donanim birimi ile farkli yapay zeka algoritmalari kullanilarak hesaplanan ve bulus tarafindan sunulan en uygun eklem açi degerleri baz alinarak yeni konum, pozisyon ve oryantasyon bilgilerine iliskin simülasyonlarin görsel olarak elde edilebilecegi elde tasinabilir, küçük boyutlu basit bir cihaza ihtiyaç duyulmaktadir. Bulusun Tanimi: Bu bulus; endüstriyel robot manipülatörlerin görev veya is hatti degisikliklerinde yapay zekâ algoritmalari yardimiyla eklem açilarinin hesaplanarak en uygun konum ve oryantasyon bilgilerini göre olusturulan simülasyonlarin gözlemlenebilecegi dinamik robot yönlendirmesi için el terminali ve algoritmasi ile ilgilidir. Bulus; tasinabilir, küçük boyutlu, zamandan, kisiden, makineden, mekândan bagimsiz olarak herhangi bir ortamda, herhangi bir zamanda ve herhangi bir personel tarafindan kullanilabilen bir el terminali cihazini içermektedir. Bulus konusu ürün; genel itibariyle dokunmatik ekran, batarya, ana kart ve tüm bu yapilari koruyan plastik bir arka kapaktan meydana gelmektedir. Sisteme veri girisini saglayan dokunmatik ekran; kizilötesi tipi dokunmatik ekran teknolojileri kullanilarak olusturulmustur. Ekran kenarinda LED ve foto algilayioilar bulunmakta olup, algilayicilar LED bütünlügünü kontrol etmektedir. Bulus konusu ürün bir el terminal cihazi oldugu için tasinabilir özelligi olmasi gerekmekte olup, bulusta bataryanin mevcudiyeti bu Özelligi kullanilabilir kilmaktadir. Burada bahsi geçen bulus bir ürün niteliginde olup, gömülü sayisal tasarim yöntemi ile donanimsal yazilim entegre edilerek ortaya konmustur. Bahsi geçen bulusa entegre edilen gömülü sistem olarak kullanilan F ield-programmable gate array (Alanda programlanabilir kapi dizisi) (F PGA) donanimsal yazilim sayesinde eklem sayisi fark etmeksizin yapay zeka tabanli islemlerle robotlarin kinematik hesaplamalari milisaniyeler içinde gerçeklestirilmekte ve sonuçlar LCD ekranda simüle edilmektedir. Bulus bünyesinde yer alan ana kart; farkli yapay zeka algoritmalarinin tek bir karta entegre halde bulunmasina ve yapilan hesaplamalarin tek bir ortamda farkli yapay zeka teknikleri ile elde edilmesine ve yine tek bir platformda analiz edilmesine imkan tanimaktadir. Bulusa entegre ana kart sayesinde tek bir platformda gerçeklestirilen hesaplamalar ve bu hesaplamalar dogrultusunda olusturulan simülasyonlar kullaniciya sunulmaktadir. Yapay zeka tabanli islemler sayesinde gerçeklestirilen hesaplamalarda sonsuz sayida yeni deger elde edilmekte ve en uygun degerler sistem tarafindan önerilmektedir. Bulus konusu sistemin donanimi üzerinde gerçeklestirilen islemlerin kullaniciya aktarilmasi ve/veya kullanicinin donanimda islenmesini istedigi bilgileri sisteme aktarabilmesi için ayrica bir kullanici arabirim yazilimi bulunmaktadir. Bu yazilimin çalismasi için kullanici sisteme robot DH parametrelerini kaydetmekte ve istedigi herhangi bir zamanda veya mekanda herhangi bir kisi tarafindan görevi veya görev yeri degisen robotun yeni görevini en kusursuz sekilde gerçeklestirecegi eklem açilari sistem tarafindan dört farkli yapay zekâ algoritmasiyla hesaplanmaktadir. Daha sonrasinda kullanici ekran üzerinde hesaplanan degerlerin simülasyonuna ulasmakta ve kullanici tercihine sunulan birçok yeni deger ve görsel simülasyonlar arasindan en uygun olanini seçmekte ve ardindan sisteme bu bilgileri girip robotu çalistirabilmektedir. Bulus için gelistirilen kullanici arabirim yazilimi ile robotun eklem sayisi ve robotun bulundugu yerin ismi gibi hatirlatma bilgilerinin de yer alabilecegi tanimli robot listesine yeni bir ekleme yapilmaktadir. Kullanicinin görsel arabirimi kullanarak islem yapacagi robotlara ait DH parametreleri ROM belleginde saklanmakta ve bu adimlardan sonra eklenen ve depolanan DH parametrelerine göre ekrana robotun simülasyonu getirilmekte ve dogrulama gerçeklestirilmektedir. Bulusun asil hedefini gerçeklestirmek üzere sisteme kayitli bir robot için bir görev ya da görev yeri degisikligi saptandiginda robota ait en ideal pozisyon ve konum bilgileri; bulusun bilesenlerinden biri olan FPGA ana kart içerisine gömülü olarak bulunan farkli ve yepyeni bir yöntem kullanilarak olusturulan donanim birimi, yapay zekâ algoritmalari ile robotun ters kinematik hesabini paralel bir sekilde gerçeklestirerek sonuçlari LCD ekrana aktarmaktadir. Sonuçlar ates böcegi (ABO), yapay ari kolonisi (YAKA), parçacik sürü optimizasyonu (PSO), kuantum parçacik sürü optimizasyonu (QPSO) olmak üzere bu dört farkli yapay zeka algoritmasi ile hesaplanmaktadir. RAM bellekten verileri alan yapay zeka algoritmalari paralel bir hat olusturarak hesaplamalarin hizli bir sekilde gerçeklestirilmesini saglamaktadir. Bulusa entegre ana kart, yapay zeka algoritmalarinin tek bir karta entegre olmasini saglamanin yaninda tamamen donanimsal olarak çalisma gerçeklestirdigi için islem basamaklarinin çok daha hizli bir sekilde gerçeklesmesine imkan tanimaktadir. Sistemin çalisma algoritmasina göre öncelikle ROM belleginde saklanan robot bilgileri RAM bellegine aktarildiktan sonra RAM bellekten verileri alan yapay zeka algoritmalari paralel bir hat olusturarak çalistirilmaktadir. Kinematik hesaplamalar gerçeklestirilmekte ve daha sonra her bir yapay zeka algoritmasiyla gerçeklestirilen hesaplamalardan yola çikarak pozisyon hatalari tespit edilmektedir. Bu esnada kayan noktali islemler (FPU) kinematik hesabi ile; aritinetik mantik islemler (ALU) pozisyon hatasi ile iliskili gerçeklesecek sayisal islemleri yapan arabirimler olarak sistemde yer almaktadir. En sonunda yapilan hesaplamalar sonucunda hata payi en az olan eklem açilari RAM bellege kaydedilmekte ve sonuçlar ekrana yansitilmaktadir. Yapay zeka algoritmalariyla eklem açilarinin hesaplanmasi için robotun uç elemaninin x, y ve 2 koordinat bilgileri sisteme girilmektedir ve sistemin ana kartinda yer alan gömülü sistem tasarimi ile islemler çok kisa sürelerde gerçeklestirildikten sonra sonuçlar ekranda sunulmaktadir. Yapay zeka algoritmalariyla en dogru sonucu veren eklem açilari temel alinarak olusturulan robot simülasyonlari ekranda yer almaktadir. Bulus konusu ürün yapisal ve karakteristik özellikleri ve tüm avantajlari asagida verilen sekiller ve bu sekillere atiflar yapmak suretiyle yazilan detayli açiklama sayesinde daha net anlasilacaktir ve bu nedenle degerlendirmenin de bu sekiller ve detayli açiklama göz önünde bulundurularak yapilmasi gerekmektedir. Sekillerin Açiklanmasi: Bulus, ilisikteki sekillere atifta bulunularak anlatilacaktir, böylece bulusun özellikleri daha açikça anlasilacak ve takdir edilecektir, fakat bunun amaci bulusu bu belli düzenlemeler ile sinirlamak degildir. Tam tersine; bulusun ilisikteki istemler tarafindan tanimlandigi alani içine dahil edilebilecek bütün alternatifleri, degisiklikleri ve denkliklerinin kapsanmasi amaçlanmistir. Gösterilen ayrintilar, sadece mevcut bulusun tercih edilen düzenlemelerinin anlatimi amaciyla gösterildigi ve hem yöntemlerin sekillendirilmesinin hem de bulusun kurallari ve kavramsal özelliklerinin en kullanisli ve kolay anlasilir tanimini saglamak amaciyla sunulduklari anlasilmalidir. Bu çizimlerde; Sekil 1 Bulusu meydana getiren parçalarin sembolik görünümüdür. Sekil 2a Bulus için gelistirilen kullanici arabirim yaziliminda veri giris birimde yeni robot tanimlama ekraninin sembolik görünümüdür. Sekil 2b Bulus için gelistirilen kullanici arabirim yaziliminda veri giris birimde Robot DH parametrelerini girme ekraninin sembolik görünümüdür. Sekil 2c Bulus için gelistirilen kullanici arabirim yaziliminda robot simülasyon ekraninin sembolik görünümüdür. Sekil 3a Bulus için gelistirilen kullanici arabirim yaziliminda veri giris birimde yeni görev olusturma ekraninin sembolik görünümüdür. Sekil 3b Bulus için gelistirilen kullanici arabirim yaziliminda yeni görev simülasyon ekraninin sembolik görünümüdür. Sekil 4 Ters kincmatik hesaplayici blok semasi görünümüdür. Bu bulusun anlasilmasina yardimci olacak sekiller ekli resimde belirtildigi gibi numaralandirilmis olup isimleri ile beraber asagida verilmistir. Referanslarin Açiklanmasi: 1. Giris Paneli 2. LCD Ekran 3. Anakart . Veri Giris Birimi . Islem Birimi Bulusun Açiklanmasiz Bulus; sisteme disardan veri girisinin yapildigi giris paneli (1), anakartta (3) gerçeklestirilen islemlerin sonuçlarinin görüntülendigi, simülasyon isleminin gerçeklestirildigi ve anakartta (3) gerçeklestirilen islemlerin sonucunun gösterildigi LCD ekran (2), yapay zeka algoritmalarinin tek bir karta entegre seklinde eklenmesini saglayarak islemleri hizli bir sekilde gerçeklestiren ve çevresel giris/çikis birimleriyle (tus takimi, ekran gibi) sunucu üzerinden haberlesmeyi saglayan anakart (3) ve sisteme kablosuz tasinma özelligi kazandiran bataryayi (4) içermektedir. Bulus konusu el terminalinde yer alan giris paneli (l) dokunmatik ekran olmaktadir. Bulus konusu el terminalinde yer alan anakartta (3) gömülü sistem olarak içerisinde Field-programmable gate array (Alanda programlanabilir kapi dizisi) (FPGA) teknolojisi kullanilmaktadir. Bulus konusu el terminalinde yer alan anakartta (3) gömülü sistem ile eklem sayisi fark etmeksizin robotlarin ters kinematik problemi çözülmektedir. Bulus konusu el terminalinde yer alan LCD ekran (2) anakart (3) tarafindan üretilen sonuçlari simüle ederek kullaniciya göstermektedir. Bulus konusu el terminalinin algoritmasi kullanicinin sisteme giris paneli (1) tarafindan robot DH parametrelerinin kaydettigi veri giris birimi (10), görevi veya görev yeri degisen robotun yeni görevini gerçeklestirecegi eklem açilarinin sistem tarafindan farkli yapay zekâ algoritmalan tarafindan anakart (3) tarafindan hesaplandigi ve simülasyonlarin olusturuldugu islem birimi (20) içermektedir. Bulus konusu el terminalinin algoritmasi; el terminali üzerinde gerçeklestirilen islemlerin kullaniciya gösterilmesi veya kullanicinin terminalde islenmesini istedigi bilgileri disardan sisteme aktardigi veri giris birimini (10) içennektedir. Bulus konusu el terminalinin algoritmasi; robota ait eklem sayisi, konum veya bölüm bilgisi gibi bilgileri üzerinden sisteme robot tanimlamalarinin yapildigi veri giris birimini (10) içermektedir. Bulus konusu el terminalinin algoritmasi; ates böcegi (ABO), yapay ari kolonisi (YAKA), parçacik sürü optimizasyonu (PSO), kuantum parçacik sürü optimizasyonu (QPSO) gibi yapay zeka algoritmalarini içeren islem birimini (20) içermektedir. Bulusun Detayli Açiklanmasi: Bulus konusu el terminalini olusturulan temel unsurlar giris paneli (1), LCD ekran (2), anakart (3) ve batarya (4) olmaktadir. Yukarida belirtilen unsurlar özellikleri asagidaki gibidir: Dokunmatik Ekran olan giris paneli (1): Bulusun bu parçasi sisteme disardan veri girmek amaciyla kullanilmaktadir. Kizilötesi teknolojisi ile olusturulmustur ve 7" boyutundadir. Kizilötesi teknolojisinde ekran kenarinda LED ve foto algilayicilar bulunmaktadir. Algilayicilar LED bütünlügünü kontrol etmektedir. Elle dokunulan yerin LED bütünlügü bozuldugundan dokunulan yer kontrolcü tarafindan belirlenmektedir. LCD ekran (2): Anakartta (3) gerçeklestirilen islemlerin sonuçlarini görüntülemek, simülasyon islemini gerçeklestirmek amaciyla kullanilmaktadir. Anakartta (3) gerçeklestirilen islemlerin sonucunu görüntülemek için kullanilmaktadir. Anakart (3): Bulusun asil olarak özgün olmasini saglayan parçasidir. Yapay zeka algoritmalarinin tek bir karta entegre seklinde eklenmesini saglayarak islemlerin çok hizli bir sekilde gerçeklesmesini saglamaktadir. Bunun yaninda çevresel giris/çikis birimleriyle de (tus takimi, ekran gibi) haberlesmeyi saglamaktadir. Bulus nano teknoloji ile üretimi gerçeklestirilen Artix 7 yongasini üzerinde barindiran 500 MHZ frekans araliginda çalisabilen bir FPGA kartidir. Bulus ASIC yapida bir entegrede olabilmektedir. Batarya (4): Bulus bir el terrninalini andirdigindan dolayi tasinabilir özellige sahiptir. Bu özelligin bulusa kazandirilmasini saglayan elemandir. Lityum iyon yapida, 5000 mAh gücünde 3.7 V gerilim üretebilen bu donanim birimidir. Boyutu 8 7 10 cm arali gindadir. Bulus konusu sistemin tüm bu parçalarin yaninda donanim üzerinde gerçeklestirilen islemlerin kullaniciya gösterilmesi veya kullanicinin donanimda islenmesini istedigi bilgileri disardan sisteme aktarabilmesi için ayrica bir kullanici arabirim yazilimi olusturulmustur. Bu arabirim yazilimi su algoritma ile çalismaktadir: a) Kullanici sisteme veri giris birimi (10) üzerinden robot DH parametrelerini kaydeder, b) Görevi veya görev yeri degisen robotun yeni görevini en kusursuz sekilde gerçeklestirecegi eklem açilari sistem tarafindan farkli yapay zekâ algoritmalariyla hesaplanir, c) islem birimi (20) tarafindan yapilan bu degerlerin simülasyonu LCD ekran (2) üzerinden olusturulur ve kullanici bu degerlere bakar, d) En uygun pozisyonu robota uygular Bu algoritmanin kullanicilardan bagimsiz en kolay sekilde gerçeklesmesi için gelistirilen kullanici arabirim yaziliminin adim adim çalismasi asagidaki Sekil-2'de gösterilmistir. Sekil 2°de kullanici arabirim yazilimi ile bulusun ROM bellegine yeni bir robot eklenmektedir. Burada robota hem kaç eklemli oldugu hem de hangi bölümde bulundugunu animsatan bir isim verilmektedir. Robot ekleme islemi robota ait DH parametrelerinin eklenmesiyle gerçeklestirilmektedir. Sekil 2c,te eklenen robotun eklem simülasyonu gösterilerek robotun dogrulamasi yapilmaktadir. Sekil 39te ise sisteme kaydedilmis olan robota ait yeni bir görev tanimlandiginda en ideal konumu bulmak için gerçeklestirilen projeksiyon adimlari görünmektedir. Bu projeksiyon için dört farkli yapay zeka algoritmasi (ates böcegi (ABO), yapay ari kolonisi (YAKA), parçacik sürü optimizasyonu (PSO), kuantum parçacik sürü optimizasyonu (QPSO)) kullanilabilmektedir. Sekil 3a,de robotun uç elemaninin x, y ve z koordinat bilgileri sisteme girildikten sonra islemler hizli bir sekilde gerçeklestirilip sonuçlar ekrana getirilmektedir. Sekil 3b°de ise elde edilen sonuçlara göre robot simülasyonunu göstermektedir. Bulusun asil itibariyle özü anakarttir (3) (FPGA karti). Bu kart içerisinde gömülü olarak bulunan, farkli ve yepyeni bir yöntem kullanilarak olusturulan donanim birimi dört farkli yapay zekâ yöntemi ile robotun ters kinematik hesabini paralel bir sekilde gerçeklestirerek sonuçlari çok hizli bir sekilde LCD ekrana (2) aktarmaktadir. Sekil 4"te önerilen yeni yöntemin blok semasi görünmektedir. Bu semada görünen her bir birim donanimsal olarak olusturulmus olup sayisal sistem literatüründe küçük/orta ölçekli entegre ismini almaktadir. Yöntemin tamami ise büyük ölçekli entegre (VLSI) olarak isimlendirilmektedir. Yöntem tamamiyla donanimsal olarak çalismakta oldugundan islemler çok hizli bir sekilde gerçeklesmekte ve sonuçlar çok kisa sürelerde elde edilmektedir. Sekil-4°te göründügü üzere yapay zekâ algoritmalari paralel bir hat olusturarak ayni anda "RAM" bellekten verileri alarak islem yapmaktadirlar. Dolayisiyla hesaplamalarin paralel olmasi ayrica sistemi hizlandirmaktadir. Sekil-4'te yer alan "ROM" ise kullanicinin görsel arabirimi kullanarak islem yapacagi robotlara ait DH parametrelerini tutan donanim birimidir. Disardan bilgi girmek için ise sistemde ayrica "KeyPad" isminde ayri bir donanim birimi olusturulmustur. "ALU" ve "FPU" sistemde gerçeklesecek sayisal islemleri yapan arabirimlerdir. Sistemin çalisma algoritmasi su sekildedir: l- Seçilen robot bilgilerini ROM°dan RAMia aktar. 2- RAM°da olusan baslangiç bilgileriyle yapay zekâ algoritmalarini paralel olarak çalistir. 3- Algoritmalardan elde edilen bilgilerle ileri kinematik hesabini gerçeklestir. 4- Her bir algoritma ile elde edilen pozisyon hatalarini elde et. - En küçük pozisyon hatasini veren eklem açilarini RAM"a kaydet. 6- Sonuçlari LCD ekranda (3) görüntüle. Sekil ?de parçalari verilen tablet tarzi el terminali gerekse de Sekil 4`te blok semasi verilen ve anakart (3) üzerinde gömülü olarak bulunan, hesaplama islemlerini yeni bir yöntem seklinde gerçeklestiren donanim birimi bulustaki yeniliklerdir. Hali hazirda endüstride bu tarz hesaplamalar üst düzey yetkili mühendisler tarafindan yapilmaktadir. Ancak her mühendisin çalisma yöntemine göre islem adimlari veya sonuca ulasma biçimi firmadan firmaya veya kisiden kisiye göre degisiklik arz etmektedir. Bu durum kurumsal yapida standartlasmanin önüne geçmektedir. Iste, burada bahsi geçen bulus sayesinde her durum veya her sorumlu kisi için ayni islem ayni sekilde gerçeklesmektedir. Böylece robotun is hatti veya görev degisikligi islemi kisiden veya kurumdan bagimsiz bir sekilde gerçeklestirilmektedir. TR TR TR DESCRIPTION HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE Technological Field: This invention; Handheld terminal for dynamic robot guidance, where simulations of new location, position and orientation information can be quickly obtained according to joint angles, in any environment, at any time and by anyone, without requiring expertise, in order for industrial robots to continue their functions effectively in case of task or location changes. It's about the algorithm. Known State of the Technique: In the current technique, determining the new values of joint angles in work line or position changes where new position and orientation information of serial robots will be needed requires performing non-linear and complex mathematical operations manually or with the help of a computer. Especially if the robots constantly change the work line or location, engineers need to allocate a separate shift for these operations, which leads to both an increase in workload and unnecessary time loss. Even if complex operations are desired to be performed with the help of computers, different algorithms are coded separately and after the results are obtained, each result is analyzed on separate platforms. Afterwards, the appropriate value is tried to be found as a result of inferences or predictions regarding the obtained joint angles. The position information obtained based on prediction and inference has emerged as a result of the optimization of a single value, and even if there are better values, these values need to be revealed with additional calculations. Thus, the operation of the process is delayed and causes loss of time. Today, the fact that the value that gives the best result in these calculations for robots depends on people's conclusions creates unreliability and creates a disadvantageous situation due to the possibility of affecting the working efficiency of robots. Because during the calculation process, the process steps or the way they reach the result may vary depending on the working principle of the experts. This situation causes experts to obtain different results for the same process and also prevents standardization in the institutional structure. Even if the calculations reached by experts are correct, the robot must be tested in the field with new position information and necessary corrections must be made, and this causes the process to be unnecessarily prolonged. In order to overcome such problems, the subject titled "Kinematic Calculation Method" is discussed in the literature. The present invention is about an inverse kinematic calculation device and method to reveal the joint angles from the position and posture of the robot arm by performing inverse kinematic calculation in a multi-joint robot arm. Input device. , performs inverse kinematic calculation on a multi-joint robot arm with an inverse kinematic calculation unit consisting of a numerical calculator, joint angle calculator, controller, convergence detection device and output device. "Inverse Kinematics Solution Method for Robot Based on Teaching to Learn" is described in the application numbered CN108427282. The current invention works with a machine learning algorithm that uses the Gaussian mixture model to calculate robot inverse kinematics and offers a robot inverse kinematics solution method based on teaching learning; the joint angle of a certain number of robot groups, the Cartesian position of an end effector and the Euler angle are collected. A data set is obtained and after this data set is optimized, the data set is calculated iteratively to obtain the Gaussian mixture model parameters. In the applications discussed above, inventions regarding obtaining joint angle values by inverse kinematic calculation in robots are discussed. In these methods, joint angles of robots are calculated using different techniques. However, these systems do not include simulations that include new position, position and orientation information according to the joint angles of the robots, and robots whose joint angles are determined by inverse kinematic calculation cannot be visualized visually on a tablet-style handheld terminal. In this case, the verification process requires extra time and creates unnecessary workload. Even if the verification process is carried out successfully, the robots must be tested in the field with the position information of the new joints and angle values and the necessary arrangements must be made accordingly. Therefore, it causes the process to be unnecessarily prolonged. As a result, the hardware embedded in the main board that can overcome the above-mentioned disadvantages and perform the same operation in the same way for every situation or person, independent of time, place and person, so that the robots can continue their functions effectively when the task or place of duty changes. There is a need for a handheld, small-sized, simple device that can visually obtain simulations of new location, position and orientation information based on the most appropriate joint angle values calculated using different artificial intelligence algorithms and presented by the invention. Description of Invention: This invention; It is related to the hand terminal and algorithm for dynamic robot guidance, where simulations created according to the most appropriate position and orientation information can be observed by calculating joint angles with the help of artificial intelligence algorithms in case of task or work line changes of industrial robot manipulators. Meet; It includes a portable, small-sized handheld terminal device that can be used in any environment, at any time and by any personnel, regardless of time, person, machine or place. The product of the invention; It generally consists of the touch screen, battery, main board and a plastic back cover that protects all these structures. Touch screen that allows data entry into the system; It was created using infrared type touch screen technologies. There are LED and photo sensors on the edge of the screen, and the sensors control the integrity of the LED. Since the product subject to the invention is a handheld terminal device, it must have a portable feature, and the presence of the battery in the invention makes this feature usable. The invention mentioned here is a product and was created by integrating embedded digital design method and hardware software. Thanks to the Field-programmable gate array (F PGA) hardware software used as the embedded system integrated into the aforementioned invention, kinematic calculations of robots are carried out within milliseconds with artificial intelligence-based processes, regardless of the number of joints, and the results are simulated on the LCD screen. The main card included in the invention; It allows different artificial intelligence algorithms to be integrated on a single card and the calculations made to be obtained with different artificial intelligence techniques in a single environment and analyzed on a single platform. Thanks to the motherboard integrated into the invention, calculations performed on a single platform and simulations created in line with these calculations are presented to the user. An infinite number of new values are obtained in the calculations carried out thanks to artificial intelligence-based processes, and the most appropriate values are suggested by the system. There is also a user interface software to transfer the operations performed on the hardware of the system subject to the invention to the user and/or to transfer the information that the user wants to be processed in the hardware to the system. In order for this software to work, the user records the robot DH parameters in the system and the joint angles at which the robot will perform its new task in the most perfect way, if its task or location is changed by anyone at any time or place, are calculated by the system with four different artificial intelligence algorithms. Afterwards, the user accesses the simulation of the calculated values on the screen and selects the most appropriate one among the many new values and visual simulations offered to the user's preference, and then can enter this information into the system and operate the robot. With the user interface software developed for the invention, a new addition is made to the defined robot list, which can also include reminder information such as the number of joints of the robot and the name of the location where the robot is located. DH parameters of the robots that the user will operate using the visual interface are stored in the ROM memory, and after these steps, the simulation of the robot is displayed on the screen and verification is performed according to the DH parameters added and stored. In order to realize the main goal of the invention, when a task or task location change is detected for a robot registered in the system, the ideal position and location information of the robot; The hardware unit, which is created using a different and brand new method embedded in the FPGA main board, one of the components of the invention, performs the inverse kinematics calculation of the robot in parallel with artificial intelligence algorithms and transfers the results to the LCD screen. The results are calculated with four different artificial intelligence algorithms: firefly (ABO), artificial bee colony (YAKA), particle swarm optimization (PSO), and quantum particle swarm optimization (QPSO). Artificial intelligence algorithms that retrieve data from RAM memory create a parallel line and enable calculations to be carried out quickly. The motherboard integrated with the invention not only enables artificial intelligence algorithms to be integrated into a single card, but also allows the processing steps to be carried out much faster since it operates entirely in hardware. According to the operating algorithm of the system, first the robot information stored in the ROM memory is transferred to the RAM memory, and then the artificial intelligence algorithms that receive the data from the RAM memory are run by creating a parallel line. Kinematic calculations are performed and then position errors are detected based on the calculations performed by each artificial intelligence algorithm. Meanwhile, with floating point operations (FPU) kinematic calculation; Arithinetic logic operations (ALU) are included in the system as interfaces that perform digital operations related to position error. As a result of the final calculations, the joint angles with the least margin of error are saved to RAM memory and the results are reflected on the screen. In order to calculate joint angles with artificial intelligence algorithms, x, y and 2 coordinate information of the end element of the robot is entered into the system and the results are presented on the screen after the operations are carried out in a very short time with the embedded system design on the main board of the system. Robot simulations created with artificial intelligence algorithms based on the joint angles that give the most accurate results are displayed on the screen. The structural and characteristic features and all the advantages of the product subject to the invention will be more clearly understood thanks to the figures given below and the detailed explanation written by referring to these figures, and therefore the evaluation should be made by taking these figures and the detailed explanation into consideration. Description of Drawings: The invention will be described with reference to the accompanying drawings, so that the features of the invention will be more clearly understood and appreciated, but the purpose of this is not to limit the invention to these particular embodiments. Just the opposite; It is intended to cover all alternatives, modifications and equivalences of the invention that may be included within the scope of the invention as defined by the appended claims. It should be understood that the details shown are for the sole purpose of illustrating preferred embodiments of the present invention and are presented for the purpose of providing the most useful and easily understandable description of both the embodiment of the methods and the rules and conceptual features of the invention. In these drawings; Figure 1 is the symbolic view of the parts that make up the invention. Figure 2a is the symbolic view of the new robot definition screen in the data entry unit in the user interface software developed for the invention. Figure 2b is the symbolic view of the screen for entering Robot DH parameters in the data entry unit in the user interface software developed for the invention. Figure 2c is the symbolic view of the robot simulation screen in the user interface software developed for the invention. Figure 3a is the symbolic view of the new task creation screen in the data entry unit in the user interface software developed for the invention. Figure 3b is the symbolic view of the new task simulation screen in the user interface software developed for the invention. Figure 4 is the inverse kinematic calculator block diagram view. The figures that will help understand this invention are numbered as indicated in the attached picture and are given below with their names. References Explanation: 1. Access Panel 2. LCD Screen 3. Motherboard. Data Entry Unit. Processing Unit Invention Without Disclosure; The input panel (1) where data is entered into the system from outside, the LCD screen (2) where the results of the operations performed on the motherboard (3) are displayed, the simulation process is carried out and the results of the operations performed on the motherboard (3) are displayed (2), the artificial intelligence algorithms are integrated into a single card, making the operations faster. It includes the motherboard (3), which communicates with peripheral input/output units (such as keypad, screen) via the server, and the battery (4), which provides the system with wireless portability. The input panel (l) in the handheld terminal that is the subject of the invention is a touch screen. Field-programmable gate array (FPGA) technology is used as an embedded system on the motherboard (3) in the handheld terminal of the invention. The inverse kinematics problem of robots is solved, regardless of the number of joints, with the system embedded in the motherboard (3) in the handheld terminal of the invention. The LCD screen (2) in the handheld terminal of the invention simulates the results produced by the motherboard (3) and shows them to the user. The algorithm of the handheld terminal that is the subject of the invention is the data input unit (10) where the robot DH parameters are recorded by the user's system entry panel (1), the joint angles at which the robot will perform its new task when its task or location changes are calculated by the system by the motherboard (3) by different artificial intelligence algorithms and simulations are made. It contains the processing unit (20) in which it is created. The algorithm of the handheld terminal that is the subject of the invention; It includes the data entry unit (10) through which the operations performed on the handheld terminal are displayed to the user or the information that the user wants to be processed in the terminal is transferred from outside to the system. The algorithm of the handheld terminal that is the subject of the invention; It contains the data input unit (10) where robot definitions are made to the system through information such as the number of joints, location or section information of the robot. The algorithm of the handheld terminal that is the subject of the invention; It contains the processing unit (20) containing artificial intelligence algorithms such as firefly (ABO), artificial bee colony (YAKA), particle swarm optimization (PSO), quantum particle swarm optimization (QPSO). Detailed Explanation of the Invention: The basic elements that make up the handheld terminal of the invention are the input panel (1), LCD screen (2), motherboard (3) and battery (4). The features of the elements mentioned above are as follows: Touch Screen input panel (1): This part of the invention is used to enter data into the system from outside. It is created with infrared technology and is 7" in size. In infrared technology, there are LED and photo sensors on the edge of the screen. The sensors control the integrity of the LED. Since the LED integrity of the area touched by hand is disrupted, the touched area is determined by the controller. LCD screen (2): It is used to display the results of the operations performed on the motherboard (3) and to perform the simulation process. It is used to display the results of the operations performed on the motherboard (3). Motherboard (3): It is the part that makes the invention original. By allowing artificial intelligence algorithms to be integrated into a single card, it ensures that transactions are carried out very quickly. In addition, it also provides communication with peripheral input/output units (such as keypad, screen). The invention is an FPGA card that can operate in the 500 MHZ frequency range, incorporating the Artix 7 chip produced with nano technology. The invention can be an integrated device with an ASIC structure. Battery (4): The invention has a portable feature because it resembles a handheld terminal. It is the element that provides this feature to the invention. This hardware unit has a lithium-ion structure and can produce 3.7 V voltage with a power of 5000 mAh. Its size ranges from 8 to 7 to 10 cm. In addition to all these parts of the system subject to the invention, a user interface software has also been created to show the operations performed on the hardware to the user or to transfer the information that the user wants to be processed in the hardware from outside to the system. This interface software works with the following algorithm: a) The user records the robot DH parameters to the system via the data input unit (10), b) The joint angles at which the robot, whose task or location changes, will perform its new task in the most perfect way are calculated by the system with different artificial intelligence algorithms, c) The simulation of these values made by the processing unit (20) is created on the LCD screen (2) and the user looks at these values, d) Applies the most appropriate position to the robot. The step-by-step operation of the user interface software developed to realize this algorithm in the easiest way, independent of the users, is shown in Figure-2 below. It is shown in . In Figure 2, a new robot is added to the ROM memory of the invention with the user interface software. Here, the robot is given a name that reminds us of how many joints it has and in which section it is located. Adding a robot is done by adding the DH parameters of the robot. In Figure 2c, the joint simulation of the added robot is shown and the robot is verified. Figure 39 shows the projection steps taken to find the ideal position when a new task is defined for the robot registered in the system. Four different artificial intelligence algorithms (firefly (ABO), artificial bee colony (YAKA), particle swarm optimization (PSO), quantum particle swarm optimization (QPSO)) can be used for this projection. In Figure 3a, after the x, y and z coordinate information of the end element of the robot is entered into the system, the operations are carried out quickly and the results are displayed on the screen. Figure 3b shows the robot simulation according to the results obtained. Essentially, the essence of the invention is the motherboard (3) (FPGA card). The hardware unit embedded in this card, created using a different and brand new method, performs the inverse kinematic calculation of the robot in parallel with four different artificial intelligence methods and transfers the results to the LCD screen (2) very quickly. Figure 4 shows the block diagram of the proposed new method. Each unit appearing in this diagram is created with hardware and is called small/medium scale integrated in the digital system literature. The entire method is called large scale integrated (VLSI). Since the method works entirely with hardware The operations are carried out very quickly and the results are obtained in a very short time. As seen in Figure 4, artificial intelligence algorithms create a parallel line and perform operations by simultaneously retrieving data from the "RAM" memory. Therefore, the parallelism of the calculations also speeds up the system. The "ROM" located in is the hardware unit that holds the DH parameters of the robots that the user will operate using the visual interface. In order to enter information from outside, a separate hardware unit called "KeyPad" has been created in the system for the digital operations to be performed in the system. The operating algorithm of the system is as follows: l- Transfer the selected robot information from ROM to RAM. 2- Run artificial intelligence algorithms in parallel with the initial information created in RAM. 3- Perform advanced kinematics calculations with the information obtained from the algorithms. 4- Obtain the position errors obtained with each algorithm. - Save the joint angles that give the smallest position error to RAM. 6- Display the results on the LCD screen (3). Both the tablet-style handheld terminal, whose parts are given in Figure 1, and the block diagram shown in Figure 4, are embedded on the motherboard (3). The hardware unit that performs the calculation operations in a new way is an innovation in the invention. Currently, such calculations are made by senior authorized engineers in the industry. However, the process steps or the way of reaching the result vary from company to company or person to person, depending on the working method of each engineer. It prevents standardization in the corporate structure. Thanks to the invention mentioned here, the same process is performed in the same way for every situation or responsible person. Thus, the robot's work line or task change process is carried out independently of the person or institution.

TR2021/022147A 2021-12-31 2021-12-31 HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE TR2021022147A2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
TR2021/022147A TR2021022147A2 (en) 2021-12-31 2021-12-31 HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE
PCT/TR2022/050654 WO2023128957A1 (en) 2021-12-31 2022-06-24 Handheld terminal and algorithm for dynamic robot orientation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TR2021/022147A TR2021022147A2 (en) 2021-12-31 2021-12-31 HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE

Publications (1)

Publication Number Publication Date
TR2021022147A2 true TR2021022147A2 (en) 2022-02-21

Family

ID=85117810

Family Applications (1)

Application Number Title Priority Date Filing Date
TR2021/022147A TR2021022147A2 (en) 2021-12-31 2021-12-31 HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE

Country Status (2)

Country Link
TR (1) TR2021022147A2 (en)
WO (1) WO2023128957A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115327914A (en) * 2022-08-24 2022-11-11 安徽机电职业技术学院 Robot motion planning method based on artificial gravitational field motion simulation

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116466563B (en) * 2023-04-07 2024-04-19 上海大学 Pump truck energy-saving control method based on improved quantum particle swarm optimization setting PID

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6445964B1 (en) * 1997-08-04 2002-09-03 Harris Corporation Virtual reality simulation-based training of telekinegenesis system for training sequential kinematic behavior of automated kinematic machine
US8024068B2 (en) * 2006-08-04 2011-09-20 Hurco Companies, Inc. Machine tool control system
US11273553B2 (en) * 2017-06-05 2022-03-15 Autodesk, Inc. Adapting simulation data to real-world conditions encountered by physical processes
US10792810B1 (en) * 2017-12-14 2020-10-06 Amazon Technologies, Inc. Artificial intelligence system for learning robotic control policies
CN108427282A (en) 2018-03-30 2018-08-21 华中科技大学 A kind of solution of Inverse Kinematics method based on learning from instruction
CN110531846B (en) * 2018-05-24 2023-05-23 卡兰控股有限公司 Bi-directional real-time 3D interaction of real-time 3D virtual objects within a real-time 3D virtual world representation real-world
JP2021122894A (en) 2020-02-05 2021-08-30 アズビル株式会社 Inverse kinematics calculation device and inverse kinematics calculation method

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115327914A (en) * 2022-08-24 2022-11-11 安徽机电职业技术学院 Robot motion planning method based on artificial gravitational field motion simulation

Also Published As

Publication number Publication date
WO2023128957A1 (en) 2023-07-06

Similar Documents

Publication Publication Date Title
Bicchi On the closure properties of robotic grasping
Kehoe et al. A survey of research on cloud robotics and automation
Bimbo et al. Global estimation of an object’s pose using tactile sensing
TR2021022147A2 (en) HAND TERMINAL AND ALGORITHM FOR DYNAMIC ROBOT GUIDANCE
Malinowski et al. Comparison of embedded system design for industrial applications
Köker Reliability-based approach to the inverse kinematics solution of robots using Elman's networks
US11279023B2 (en) System and method for determining grasping positions for two-handed grasps of industrial objects
Myllyaho et al. Systematic literature review of validation methods for AI systems
Chen et al. Visual dexterity: In-hand reorientation of novel and complex object shapes
Fleer et al. Learning efficient haptic shape exploration with a rigid tactile sensor array
Zodey et al. Matlab toolbox for kinematic analysis and simulation of dexterous robotic grippers
Liu et al. Simulation data driven design optimization for reconfigurable soft gripper system
Zhong et al. A collaborative telerobotics network framework with hand gesture interface and conflict prevention
Wasilewski et al. Measuring the perceived iq of multimodal large language models using standardized iq tests
Leino et al. RL-KLM: Automating keystroke-level modeling with reinforcement learning
Wang et al. Enhancing learning capabilities of movement primitives under distributed probabilistic framework for flexible assembly tasks
Ha et al. A general contact force analysis of an under-actuated finger in robot hand grasping
EP4228868A1 (en) Handheld terminal and algorithm for dynamic robot orientation
Marlier et al. Simulation-based Bayesian inference for robotic grasping
Conforth et al. Reinforcement learning using swarm intelligence-trained neural networks
WO2021178402A1 (en) Automated design tool
Wang et al. Uncertainty-propagated Cartesian coordinated human–robot collaboration on Riemannian manifold with hidden state-space model
Omar et al. A framework for data driven dynamic modeling of serial manipulators
Pohl et al. Humanoid robotic system for grasping and manipulation in decontamination tasks: Humanoides Robotersystem für das Greifen und Manipulieren bei Dekontaminierungsaufgaben
Martínez et al. Deep regression models for local interaction in multi-agent robot tasks