JP2022013477A - Next-generation robot with combination of next-generation artificial intelligence and robot (parallel link (robotic pod) and so on) - Google Patents
Next-generation robot with combination of next-generation artificial intelligence and robot (parallel link (robotic pod) and so on) Download PDFInfo
- Publication number
- JP2022013477A JP2022013477A JP2020125798A JP2020125798A JP2022013477A JP 2022013477 A JP2022013477 A JP 2022013477A JP 2020125798 A JP2020125798 A JP 2020125798A JP 2020125798 A JP2020125798 A JP 2020125798A JP 2022013477 A JP2022013477 A JP 2022013477A
- Authority
- JP
- Japan
- Prior art keywords
- robot
- control
- signal
- generation
- robots
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000013473 artificial intelligence Methods 0.000 title claims description 8
- 230000001537 neural effect Effects 0.000 abstract description 8
- 238000006243 chemical reaction Methods 0.000 abstract description 6
- 230000000007 visual effect Effects 0.000 abstract description 4
- 230000007935 neutral effect Effects 0.000 abstract 1
- 238000013528 artificial neural network Methods 0.000 description 15
- 238000000034 method Methods 0.000 description 10
- 238000012545 processing Methods 0.000 description 6
- 238000010422 painting Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000003466 welding Methods 0.000 description 4
- 230000000474 nursing effect Effects 0.000 description 3
- 230000008602 contraction Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005498 polishing Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000002858 neurotransmitter agent Substances 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 238000007517 polishing process Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
Images
Landscapes
- Manipulator (AREA)
Abstract
Description
従来は結び付けられることの無かった、人工知能とロボットを結びつけるものである。It connects artificial intelligence and robots, which was not previously connected.
人工知能分野(工学)では、ニューラルネットとベイズ推定を結びつけ、ハードウェアの進歩、高速化により、深層学習が実用になり、ニューラルネットの機能が見直されることになった。人工知能は身体性が無い、又は少ない、とされて来た。
脳神経科学分野(医学、生物学)では、神経伝達信号の形式が、インパルス密度変調であることが明らかとなった。ロボット技術分野(工学)では、パラレルリンク方式が、垂直多関節ロボット、水平多関節ロボット(スカラ型)と並んで実用化された。以上の三分野(工学2分野、医学生物学1分野)の発展を結びつけることによって本発明が生まれた。In the field of artificial intelligence (engineering), deep learning has become practical due to the combination of neural networks and Bayesian inference, and advances in hardware and speeding up, and the functions of neural networks have been reviewed. Artificial intelligence has been described as having no or little physicality.
In the field of neuroscience (medicine, biology), it has become clear that the form of neurotransmitter signals is impulse density modulation. In the field of robot technology (engineering), the parallel link method has been put into practical use along with vertical articulated robots and horizontal articulated robots (scalar type). The present invention was born by combining the developments of the above three fields (engineering 2 fields, medical biology 1 field).
前記三分野それぞれの文献は多数あるが、本発明のように三分野をつなぎ合わせ、組み合わせた先行文献は存在しない。一部、人工知能分野、脳神経科学分野、ロボット技術分野、をつなぎ合わせた論文はあるか、論文の目標が、学術的に如何に自然の人間・生物の挙動を模擬できるか、に焦点が当たっており、ロボットの作用端からのフィードバック信号を活かして、精密、正確、でありながら剛柔取り交ぜた制御を目標にした論文は無い。
次世代人工知能とロボット(パラレルリンク・ロボット(ロボティック・ポッド)など)を組み合わせることによって、従来の産業用ロボットが適用されていた応用分野を含み、従来ロボットではできないとされ、熟練労働者を必要としていた多くの分野に実用的に適用できる。産業用に限らず、事務所用、家庭用、遊戯用、車載用、介護用、医療用、など新しい産業分野を創る可能性を持っている。
構成: パソコン/タブレットPC/スマホ信号発生器+ニューラルネット/直通並行指令+パラレルリンク/ロボティック・ポッド(垂直多関節ロボット、水平多関節ロボット、も同様)+ロボット手先の位置センサー+位置情報のフィードバック(+触覚センサー+触感フィードバック)(+聴覚センサー(マイクロフォン)+聴覚情報フィードバック)(+視覚センサー(ビジョン)+視覚(ビジョン)情報フィードバック) (()内はオプション)
信号の内容:パソコン/タブレットPC/スマホから、インパルス密度変調方式によるパルス列をロボットへの指令信号としてニューラルネットに入力する。ニューラルネットの出力点をアクチュエータの軸数分(実用上2点から6点だが、これに限らず)設け、それぞれから出力されるパルスをリニア伸縮ポッドに指令信号として入力する。入力されるパルス列のパルス密度の違いにより、二本のポッドの伸縮度合いが異なり、ポッド端に取り付けた手先を指令通り振ることができる。振られた手先に取り付けられた位置センサーの信号を、パソコン/タブレットPC/スマホ又はニューラルネットの入力にフィードバックする。これにより、手先をフィードバック制御の対象として動作させることができ、従来の産業用ロボットではできなかった、柔軟な動き、複雑な動き、剛柔取り混ぜた動き、誤差ゼロに追い込む制御、倣い制御、危険回避、柔らかいものを掴む、をプリプログラムドでなく、フィードバック制御により、望む動きをさせることができる。望む動きになった時点で、ニューラルネットを指令信号経路から外し、代わりに直通並行指令経路、とする。
また、フィードバック信号を複数種類持ち、ニューラルネットの複数変数の重みづけ自動調整能力を利用し、より精密な制御を実現できる。
以上の動作原理により、塗装、溶接、組み立て、と言う従来ロボットの典型的応用だけでなく、複雑な形状のワークの磨き仕上げ、多層塗り、三次元構造の基板への微細部品取り付け、など応用範囲が広がる。パラレルリンクの特長である軽量高剛性を生かして、手先に重量物を取り付け、塗装・溶接大型機搭載や、塗装・溶接以外の用途にも使用できる。三次元プリンタの出力物を仕上げることにも使用できる。By combining next-generation artificial intelligence and robots (parallel link robots (robotic pods, etc.)), it includes application fields to which conventional industrial robots have been applied, and it is said that conventional robots cannot do it. It can be practically applied to many fields that were needed. Not limited to industrial use, it has the potential to create new industrial fields such as office use, home use, amusement use, in-vehicle use, nursing care use, and medical use.
Configuration: PC / Tablet PC / Smartphone signal generator + Neural net / Direct parallel command + Parallel link / Robotic pod (same for vertical articulated robot and horizontal articulated robot) + Robot hand position sensor + Position information Feedback (+ tactile sensor + tactile feedback) (+ auditory sensor (microphone) + auditory information feedback) (+ visual sensor (vision) + visual (vision) information feedback) (() is optional)
Signal content: A pulse train by the impulse density modulation method is input to the neural net as a command signal to the robot from a personal computer / tablet PC / smartphone. The output points of the neural network are provided for the number of axes of the actuator (practically 2 to 6 points, but not limited to this), and the pulses output from each are input to the linear expansion / contraction pod as a command signal. Depending on the difference in the pulse density of the input pulse train, the degree of expansion and contraction of the two pods differs, and the hand attached to the end of the pod can be shaken as instructed. The signal of the position sensor attached to the shaken hand is fed back to the input of the personal computer / tablet PC / smartphone or neural network. As a result, the hand can be operated as a target of feedback control, and flexible movement, complicated movement, mixed movement, rigid and flexible movement, control to drive to zero error, imitation control, and danger, which were not possible with conventional industrial robots. Avoiding, grabbing soft objects, is not pre-programmed, but feedback control allows you to make the desired movement. When the desired movement is achieved, the neural network is removed from the command signal path and replaced with a direct parallel command path.
In addition, it has multiple types of feedback signals and can realize more precise control by utilizing the automatic weighting adjustment ability of multiple variables of the neural network.
Based on the above operating principles, not only the typical applications of conventional robots such as painting, welding, and assembly, but also the range of applications such as polishing and finishing of workpieces with complicated shapes, multi-layer coating, and mounting of fine parts on a substrate with a three-dimensional structure. Spreads. Taking advantage of the light weight and high rigidity that are the features of the parallel link, it can be used for mounting heavy objects on the hand, for mounting on large painting / welding machines, and for applications other than painting / welding. It can also be used to finish the output of a 3D printer.
1.パソコン/タブレットPC/スマホで、作業目的、作業段階、作業内容、被加工物形状、を本アプリに入力する。職人芸の内容を入力する(ティーチング)。
2.パソコン/タブレットPC/スマホ上の本アプリは、アクチュエータの動作経路、加工動作に展開する。経路、動作を実現するパルス密度変調信号に変換する。
3.作られたパルス密度変調信号をニューラルネットの入力に渡す。
4.ニューラルネットの出力を、アクチュエータに渡す。
5.アクチュエータ作用端の複数のセンサー信号をパソコン/タブレットPC/スマホ、又はニューラルネット入力にフィードバックする。
6.以上を繰り返し、所期の動き・精度に達したら、ニューラルネットを外し、直通並行指令とする。
7.以上により、目的の作業を完遂する。1. 1. Enter the work purpose, work stage, work content, and workpiece shape into this application on a personal computer / tablet PC / smartphone. Enter the content of craftsmanship (teaching).
2. 2. This application on a personal computer / tablet PC / smartphone expands to the actuator operation path and processing operation. Convert to a pulse density modulated signal that realizes the path and operation.
3. 3. The created pulse density modulation signal is passed to the input of the neural network.
4. The output of the neural network is passed to the actuator.
5. A plurality of sensor signals at the actuator action end are fed back to a personal computer / tablet PC / smartphone or a neural network input.
6. Repeat the above, and when the desired movement and accuracy are reached, remove the neural network and use a direct parallel command.
7. With the above, the desired work is completed.
従来のロボットは、高速、精密、正確、が目標であり条件であった。一方深層学習人工知能は、確率計算による、疑似正解漸近方式であり、精密、正確、では必ずしも無く、オープンループ制御の従来ロボットの制御装置には不向きであった。本発明は、この相性の悪い両者を結びつけて、次世代のAIと次世代のロボットを同時に実現するものである。
応用分野が従来の産業用に限らず、事務所用、家庭用、遊戯用、車載用、介護用、医療用、など新しい産業分野を創る可能性を持っている。
AI側から見ると、AIは従来身体性が無い、又は少ない(視覚、聴覚、くらいしか無い)とされていたが、本発明により、AIにロボットと言う身体性を与えるものである。
IoT(Internet of Things)と言う全てのものがインターネットに接続される時代に、自律的に精密制御を行うロボットである本発明は重要なThing(構成要素)となることができる。
機構にパラレルリンク(ロボティック・ポッド)を採用する意味:パルス密度変調のパルス列は、座標変換もDA変換も必要なく、直接ポッド或いはリンクに指令信号として入力ができ、製品設計・製造・部品数がより簡素になる。垂直多関節ロボットは関節にサーボモーターを配置するので、大量の直交座標と極座標変換、AD、DA変換、を必要とし、応答速度、コンピュータパワー、部品数、自重、可搬重量、消費電力、設計工数、など各要素で不利である。しかし乍ら、本方式を垂直多関節ロボットに適用しても従来の多関節ロボットに比べてより高いレベルの制御性を実現できる。水平多関節(スカラ型)ロボットについても同じである。Conventional robots have the goal and condition of high speed, precision, and accuracy. On the other hand, deep learning artificial intelligence is a pseudo-correct answer approach method by probability calculation, and it is not always precise and accurate, and it is not suitable for the control device of the conventional robot of open loop control. The present invention combines these incompatible two to realize a next-generation AI and a next-generation robot at the same time.
The application field is not limited to the conventional industrial use, but has the potential to create new industrial fields such as office use, home use, play use, in-vehicle use, nursing care use, and medical use.
From the AI side, AI has conventionally been considered to have no or little physicality (sight, hearing, and so on), but according to the present invention, AI is given the physicality of a robot.
In an era when everything called IoT (Internet of Things) is connected to the Internet, the present invention, which is a robot that autonomously performs precision control, can be an important Thing (component).
Meaning of adopting parallel link (robotic pod) for the mechanism: The pulse train of pulse density modulation does not require coordinate conversion or DA conversion, and can be directly input to the pod or link as a command signal, and product design, manufacturing, and number of parts. Becomes simpler. Since a vertical articulated robot places a servomotor at a joint, it requires a large amount of rectangular and polar coordinate conversion, AD, DA conversion, response speed, computer power, number of parts, own weight, payload, power consumption, design. It is disadvantageous in each factor such as manpower. However, even if this method is applied to a vertical articulated robot, a higher level of controllability can be achieved compared to conventional articulated robots. The same applies to horizontal articulated (scalar type) robots.
高速、精密、正確、高剛性、一本やりであった従来ロボットを使いながら制御を革新することにより、高速と低速、精密と粗さ、正確と大雑把、剛体と柔軟・軟体、と言う、相反する動作を同時に実現する。High speed, precision, accuracy, high rigidity, by innovating control while using a conventional robot that was single-handed, high speed and low speed, precision and roughness, accuracy and rough, rigid body and flexible / soft body, conflicting movements At the same time.
プリプログラムドで選択肢無し・柔軟性無しの制御方式から、次世代人工知能ソフトウェアとニューラルネットを結びつけることにより、高速と低速、精密と粗さ、正確と大雑把、剛体と柔軟・軟体、と言う、相反する動作を同時に実現する制御方式に転換する。臨機応変の制御をも実現する。From a pre-programmed control method with no choices and no flexibility, by connecting next-generation artificial intelligence software and neural networks, high speed and low speed, precision and roughness, accuracy and rough, rigid body and flexible body, soft body, etc. Switch to a control method that simultaneously realizes conflicting operations. It also realizes flexible control.
精密な倣い動作中に、被加工物に突起が偶々発見された際、自動的に突起を避ける動き、突起を摩耗除去する動き、両方とも実現できる。その結果、従来加工が難しかった軟体、非定型物体・表面、自然物体(例えば、加工前の岩石など)を望む通り加工できるようになる。When a protrusion is accidentally found on the workpiece during the precise copying operation, the movement to automatically avoid the protrusion and the movement to remove the protrusion by wear can be realized. As a result, soft bodies, atypical objects / surfaces, and natural objects (for example, rocks before processing), which were difficult to process in the past, can be processed as desired.
パソコン/タブレットPC/スマホをインパルス密度変調方式の制御指令信号発生器とし、発生した信号をニューラルネットに入力しニューラルネットの出力を、AD変換、DA変換せずに、アクチュエータ(パラレルリンク、ロボティック・ポッド、など)に指令信号として入力する。アクチュエータの末端に位置センサーを取り付け、位置信号を、パソコン/タブレットPC/スマホ信号発生器及びニューラルネットの入力にフィードバック信号として入力することによりフィードバック制御すなわち状況に応じた精密制御を実現する。位置だけでなく、触感、聴覚、視覚、もセンサーを取り付け、フィードバック信号として、制御に使用し、被加工物の表面性状に応じた、精密柔軟な制御を実現する。A personal computer / tablet PC / smartphone is used as a control command signal generator of the impulse density modulation method, and the generated signal is input to the neural net, and the output of the neural net is not AD-converted or DA-converted, and the actuator (parallel link, robotic) is used.・ Input as a command signal to the pod, etc.). A position sensor is attached to the end of the actuator, and the position signal is input as a feedback signal to the input of the personal computer / tablet PC / smartphone signal generator and the neural net to realize feedback control, that is, precise control according to the situation. Not only the position but also the tactile, auditory, and visual sensors are attached and used for control as a feedback signal to realize precise and flexible control according to the surface texture of the workpiece.
大筋の磨き加工工程及び職人芸の内容を指示として、パソコンに入力、パソコンソフトで詳細加工工程に分解、出力パルス信号(パルス密度変調)を生成、ニューラルネットを経由して、アクチュエータの各軸に指令信号として入力する。
アクチュエータ端に取り付けられた位置センサーを含む複数センサーからの信号をフィードバック信号として、パソコンに入力する。複数のセンサーフィードバック信号で調整した出力バルス信号を生成、ニューラルネットを経由してアクチュエータの各軸に指令信号として入力する。ニューラルネットは、複数のフィードバック信号に重みづけをして、指令信号を作り出す。
以上の繰り返しの後、最適出力パルス信号が得られる。この信号を、ニューラルネットを経由せずに、アクチュエータの各軸に指令信号として入力する。各種センサー信号をパソコンにフィードバックする。この工程を繰り返すことにより、所期の加工が完了する。Input to a personal computer with instructions on the rough polishing process and the contents of craftsmanship, decompose into detailed machining processes with personal computer software, generate an output pulse signal (pulse density modulation), and connect to each axis of the actuator via a neural net. Input as a command signal.
Signals from multiple sensors including the position sensor attached to the end of the actuator are input to the personal computer as feedback signals. An output bals signal adjusted by multiple sensor feedback signals is generated and input as a command signal to each axis of the actuator via a neural network. The neural network weights a plurality of feedback signals to generate a command signal.
After repeating the above, the optimum output pulse signal is obtained. This signal is input as a command signal to each axis of the actuator without going through the neural network. Feed back various sensor signals to the personal computer. By repeating this process, the desired processing is completed.
従来技術では、ロボットの制御指令は、すべてプリプログラムドで、ティーチングプレイバック方式であっても、CADデータによる制御でも、サブミリのレベルまで、すべて予め決めた通りの制御、即ち高速、精密、正確、が大前提で、本実施例のように、フィードバック信号を用いて、精度を上げて行くことは従来出来なかった。その為職人芸と呼ばれる超精密仕上げは従来のロボットでは出来なかった。In the conventional technology, all control commands of the robot are pre-programmed, and whether it is a teaching feedback method or CAD data control, all control up to the submillimeter level is as predetermined, that is, high speed, precision, and accuracy. , Is a major premise, and it has not been possible to improve the accuracy by using the feedback signal as in this embodiment. Therefore, ultra-precision finishing called craftsmanship could not be done with conventional robots.
従来の塗装、溶接、組み立て、は勿論、不連続面の研磨、仕上げ塗装、仕上げ清掃、塗料や釉薬の上塗り、精密組み立て、精密部品実装、自動車の内装部品取り付け、それぞれに必要な倣い制御による加工、芸術作品の加工、手術ロボット、医療ロボット、介護ロボット、食品製造ロボット、食品盛り付けロボット、介助支援ロボット、リハビリ支援ロボット、事務所用ロボット、家庭用ロボット、遊戯用ロボット、など無限とも思える利用可能性がある。特に手術ロボットは、現状人間の術者のリモートコントロールであり、マジックハンドとも言える。本発明により、手術ロボットをより自動化できる。Conventional painting, welding, assembly, as well as polishing of discontinuous surfaces, finish painting, finish cleaning, topcoating of paints and glazes, precision assembly, precision component mounting, automobile interior component mounting, processing by copying control required for each , Art work processing, surgical robots, medical robots, nursing robots, food manufacturing robots, food serving robots, assistance support robots, rehabilitation support robots, office robots, home robots, game robots, etc. There is sex. In particular, surgical robots are currently remote controls for human surgeons and can be called magic hands. INDUSTRIAL APPLICABILITY According to the present invention, a surgical robot can be further automated.
Claims (3)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2020125798A JP2022013477A (en) | 2020-07-01 | 2020-07-01 | Next-generation robot with combination of next-generation artificial intelligence and robot (parallel link (robotic pod) and so on) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2020125798A JP2022013477A (en) | 2020-07-01 | 2020-07-01 | Next-generation robot with combination of next-generation artificial intelligence and robot (parallel link (robotic pod) and so on) |
Publications (1)
Publication Number | Publication Date |
---|---|
JP2022013477A true JP2022013477A (en) | 2022-01-18 |
Family
ID=80169649
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2020125798A Pending JP2022013477A (en) | 2020-07-01 | 2020-07-01 | Next-generation robot with combination of next-generation artificial intelligence and robot (parallel link (robotic pod) and so on) |
Country Status (1)
Country | Link |
---|---|
JP (1) | JP2022013477A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07291450A (en) * | 1994-04-26 | 1995-11-07 | Kawasaki Heavy Ind Ltd | Intelligent palletizing system |
JP2018043338A (en) * | 2016-09-16 | 2018-03-22 | ファナック株式会社 | Machine learning device for learning operation program of robot, robot system, and machine learning method |
JP2018149669A (en) * | 2017-03-14 | 2018-09-27 | オムロン株式会社 | Learning device and learning method |
-
2020
- 2020-07-01 JP JP2020125798A patent/JP2022013477A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07291450A (en) * | 1994-04-26 | 1995-11-07 | Kawasaki Heavy Ind Ltd | Intelligent palletizing system |
JP2018043338A (en) * | 2016-09-16 | 2018-03-22 | ファナック株式会社 | Machine learning device for learning operation program of robot, robot system, and machine learning method |
JP2018149669A (en) * | 2017-03-14 | 2018-09-27 | オムロン株式会社 | Learning device and learning method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Su et al. | Fuzzy approximation-based task-space control of robot manipulators with remote center of motion constraint | |
Su et al. | Improved human–robot collaborative control of redundant robot for teleoperated minimally invasive surgery | |
Smith et al. | Dual arm manipulation—A survey | |
Pastor et al. | Learning and generalization of motor skills by learning from demonstration | |
GB2272077A (en) | An adaptive control system | |
Petrescu et al. | Velocities and accelerations at the 3R mechatronic systems | |
Rollinson et al. | Torque control strategies for snake robots | |
Luo et al. | Repulsive reaction vector generator for whole-arm collision avoidance of 7-DoF redundant robot manipulator | |
Mistry et al. | An exoskeleton robot for human arm movement study | |
Frketic et al. | Dexterous printing and fabrication of multi-functional parts: Design for science and engineering education | |
Luo et al. | A vision-based virtual fixture with robot learning for teleoperation | |
Hazarika et al. | Robotics: history, trends, and future directions | |
Aziz et al. | Flexible joint robotic manipulator: Modeling and design of robust control law | |
JP2022013477A (en) | Next-generation robot with combination of next-generation artificial intelligence and robot (parallel link (robotic pod) and so on) | |
CH et al. | 2DxoPod-a modular robot for mimicking locomotion in vertebrates | |
Bian et al. | Dynamical system based variable admittance control for physical human-robot interaction | |
Yin et al. | A deep reinforcement learning-based decentralized hierarchical motion control strategy for multiple amphibious spherical robot systems with tilting thrusters | |
Buczyłowski et al. | A Hand-Held Teleoperation Device for the Control of the Mars Rover Analogue Manipulator | |
Zhou et al. | Coordination control of dual‐arm modular robot based on position feedback using Optotrak3020 | |
Cortesao et al. | Haptic control design for robotic-assisted minimally invasive surgery | |
Kebria et al. | Robust collaboration of a haptically-enabled double-slave teleoperation system under random communication delays | |
Nugroho et al. | An improved performance of the designed robotic motion control for NAO robot arms using hybrid neural network-Jacobian | |
Kozlowski | Robot motion and control | |
Ahmadi et al. | Dimitri: a low-cost compliant humanoid torso designed for cognitive robotics research | |
Zhou et al. | Neural network-based variable stiffness impedance control for internal/external forces tracking of dual-arm manipulators under uncertainties |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20200904 |
|
A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20211130 |
|
A02 | Decision of refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A02 Effective date: 20220531 |