Sim2Real in Robotics and Automation: Applications and Challenges

被引:61
|
作者
Hofer, Sebastian [1 ]
Bekris, Kostas [2 ,3 ]
Handa, Ankur [4 ]
Gamboa, Juan Camilo [5 ]
Mozifian, Melissa [5 ]
Golemo, Florian [6 ,7 ]
Atkeson, Chris [8 ]
Fox, Dieter [4 ,9 ]
Goldberg, Ken [10 ]
Leonard, John [11 ]
Karen Liu, C. [12 ]
Peters, Jan [13 ]
Song, Shuran [14 ]
Welinder, Peter [15 ]
White, Martha [16 ]
机构
[1] Amazon, Robot AI, D-10117 Berlin, Germany
[2] Amazon, Robot AI, Seattle, WA 98109 USA
[3] Rutgers State Univ, Comp Sci Dept, Piscataway, NJ 08854 USA
[4] NVIDIA, Seattle, WA 98105 USA
[5] McGill Univ, Sch Comp Sci, Montreal, PQ H3A 0E9, Canada
[6] Mila, Montreal, PQ H2S 3H1, Canada
[7] ElementAI, Montreal, PQ H2S 3G9, Canada
[8] CMU, Robot Inst, Pittsburgh, PA 15213 USA
[9] Univ Washington, Paul G Allen Sch Comp Sci & Engn, Seattle, WA 98195 USA
[10] Univ Calif UC Berkeley, Ind Engn Operat Res IEOR Dept, Berkeley, CA 94720 USA
[11] MIT, CSAIL, 77 Massachusetts Ave, Cambridge, MA 02139 USA
[12] Stanford Univ, Dept Comp Sci, Stanford, CA 94305 USA
[13] Tech Univ Darmstadt, Fachbereich Informat, D-64289 Darmstadt, Germany
[14] Columbia Univ, Dept Comp Sci, New York, NY 10027 USA
[15] OpenAI, San Francisco, CA 94110 USA
[16] Univ Alberta, Dept Comp Sci, Edmonton, AB T6G 2R3, Canada
关键词
Circuit simulation;
D O I
10.1109/TASE.2021.3064065
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
To Perform reliably and consistently over sustained periods of time, large-scale automation critically relies on computer simulation. Simulation allows us and supervisory AI to effectively design, validate, and continuously improve complex processes, and helps practitioners to gain insight into the operation and justify future investments. While numerous successful applications of simulation in industry exist, such as circuit simulation, finite element methods, and computeraided design (CAD), state-of-the-art simulators fall short of accurately modeling physical phenomena, such as friction, impact, and deformation.
引用
收藏
页码:398 / 400
页数:3
相关论文
共 50 条
  • [1] Real2Sim or Sim2Real: Robotics Visual Insertion Using Deep Reinforcement Learning and Real2Sim Policy Adaptation
    Chen, Yiwen
    Li, Xue
    Guo, Sheng
    Ng, Xian Yao
    Ang, Marcelo H.
    INTELLIGENT AUTONOMOUS SYSTEMS 17, IAS-17, 2023, 577 : 617 - 629
  • [2] Rethinking Sim2Real: Lower Fidelity Simulation Leads to Higher Sim2Real Transfer in Navigation
    Truong, Joanne
    Rudolph, Max
    Yokoyama, Naoki
    Chernova, Sonia
    Batra, Dhruv
    Rai, Akshara
    CONFERENCE ON ROBOT LEARNING, VOL 205, 2022, 205 : 859 - 870
  • [3] Sim2Real Grasp Pose Estimation for Adaptive Robotic Applications
    Horvath, Daniel
    Bocsi, Kristof
    Erdos, Gabor
    Istenes, Zoltan
    IFAC PAPERSONLINE, 2023, 56 (02): : 5233 - 5239
  • [4] Sim2Real When Data Is Scarce: Image Transformation for Industrial Applications
    Weisenboehler, Moritz
    Augenstein, Philipp
    Hein, Bjoern
    Wurll, Christian
    Furmans, Kai
    INTELLIGENT AUTONOMOUS SYSTEMS 18, VOL 2, IAS18-2023, 2024, 794 : 65 - 76
  • [5] Object Detection Using Sim2Real Domain Randomization for Robotic Applications
    Horvath, Daniel
    Erdos, Gabor
    Istenes, Zoltan
    Horvath, Tomas
    Foldi, Sandor
    IEEE TRANSACTIONS ON ROBOTICS, 2023, 39 (02) : 1225 - 1243
  • [6] Exploring the Sim2Real Gap using Digital Twins
    Sudhakar, Sruthi
    Hanzelka, Jon
    Bobillot, Josh
    Randhavane, Tanmay
    Joshi, Neel
    Vineet, Vibhav
    2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2023), 2023, : 20361 - 20370
  • [7] A Brief Survey of Sim2Real Methods for Robot Learning
    Dimitropoulos, Konstantinos
    Hatzilygeroudis, Ioannis
    Chatzilygeroudis, Konstantinos
    ADVANCES IN SERVICE AND INDUSTRIAL ROBOTICS, RAAD 2022, 2022, 120 : 133 - 140
  • [8] Neural Neural Textures Make Sim2Real Consistent
    Burgert, Ryan
    Shang, Jinghuan
    Li, Xiang
    Ryoo, Michael S.
    CONFERENCE ON ROBOT LEARNING, VOL 205, 2022, 205 : 2215 - 2225
  • [9] Generation of GelSight Tactile Images for Sim2Real Learning
    Gomes, Daniel Fernandes
    Paoletti, Paolo
    Luo, Shan
    IEEE ROBOTICS AND AUTOMATION LETTERS, 2021, 6 (02) : 4177 - 4184
  • [10] Engine Agnostic Graph Environments for Robotics (EAGERx): A Graph-Based Framework for Sim2real Robot Learning
    van der Heijden, Bas
    Luijkx, Jelle
    Ferranti, Laura
    Kober, Jens
    Babuska, Robert
    IEEE ROBOTICS & AUTOMATION MAGAZINE, 2024,