Online Learning of Low Dimensional Strategies for High-Level Push Recovery in Bipedal Humanoid Robots

被引:0
|
作者
Yi, Seung-Joon [1 ,2 ]
Zhang, Byoung-Tak [2 ]
Hong, Dennis [3 ]
Lee, Daniel D. [1 ]
机构
[1] Univ Penn, Grasp Lab, Philadelphia, PA 19104 USA
[2] Seoul Natl Univ, BI Lab, Seoul 151, South Korea
[3] Virginia Tech, RoMeLa Lab, Blacksburg, VA 24061 USA
关键词
humanoid robot; biomechanically motivated push recovery; low-dimensional policy; online learning;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Bipedal humanoid robots will fall under unforeseen perturbations without active stabilization. Humans use dynamic full body behaviors in response to perturbations, and recent bipedal robot controllers for balancing are based upon human biomechanical responses. However these controllers rely on simplified physical models and accurate state information, making them less effective on physical robots in uncertain environments. In our previous work, we have proposed a hierarchical control architecture that learns from repeated trials to switch between low-level biomechanically-motivated strategies in response to perturbations. However in practice, it is hard to learn a complex strategy from limited number of trials available with physical robots. In this work, we focus on the very problem of efficiently learning the high-level push recovery strategy, using simulated models of the robot with different levels of abstraction, and finally the physical robot. From the state trajectory information generated using different models and a physical robot, we find a common low dimensional strategy for high level push recovery, which can be effectively learned in an online fashion from a small number of experimental trials on a physical robot. This learning approach is evaluated in physics-based simulations as well as on a small humanoid robot. Our results demonstrate how well this method stabilizes the robot during walking and whole body manipulation tasks.
引用
收藏
页码:1649 / 1655
页数:7
相关论文
共 43 条
  • [31] Learning fuzzy cognitive maps using evolution strategies: A novel schema for modeling and simulating high-level behavior
    Koulouriotis, DE
    Diakoulakis, IE
    Emiris, DM
    [J]. PROCEEDINGS OF THE 2001 CONGRESS ON EVOLUTIONARY COMPUTATION, VOLS 1 AND 2, 2001, : 364 - 371
  • [32] Joint learning networks of low-level and high-level features for multi-label ship recognition in complex backgrounds
    Tian, Yang
    Meng, Hao
    Ling, Yue
    [J]. APPLIED INTELLIGENCE, 2023, 53 (20) : 24327 - 24345
  • [33] Joint learning networks of low-level and high-level features for multi-label ship recognition in complex backgrounds
    Yang Tian
    Hao Meng
    Yue Ling
    [J]. Applied Intelligence, 2023, 53 : 24327 - 24345
  • [34] Improving Federated Learning through Low-Entropy Client Sampling Based on Learned High-Level Features
    Abebe, Waqwoya
    Munoz, Pablo
    Jannesari, Ali
    [J]. 2024 IEEE 17TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING, CLOUD 2024, 2024, : 20 - 29
  • [35] A Teaching Approach for Bridging the Gap Between Low-Level and High-Level Programming Using Assembly Language Learning for Small Microcontrollers
    Bolanakis, Dimosthenis E.
    Evangelakis, Georgios A.
    Glavas, Euripidis
    Kotsis, Konstantinos T.
    [J]. COMPUTER APPLICATIONS IN ENGINEERING EDUCATION, 2011, 19 (03) : 525 - 537
  • [36] Low recovery rates of high-level aminoglycoside-resistant enterococci could be attributable to restricted usage of aminoglycosides in Indian settings
    Sekar, Ramalingam
    Srivani, Ramesh
    Vignesh, Ramachandran
    Kownhar, Hayath
    Shankar, Esaki Muthu
    [J]. JOURNAL OF MEDICAL MICROBIOLOGY, 2008, 57 (03) : 397 - 398
  • [37] Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data
    Gsaxner, Christina
    Roth, Peter M.
    Wallner, Juergen
    Egger, Jan
    [J]. PLOS ONE, 2019, 14 (03):
  • [38] Exploring Low-level and High-level Transfer Learning for Multi-task Facial Recognition with a Semi-supervised Neural Network
    Barros, Pablo
    Fliesswasser, Erik
    Kerzel, Matthias
    Wermter, Stefan
    [J]. 2019 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2019, : 1378 - 1384
  • [39] Evaluating low- mid- and high-level fusion strategies for combining Raman and infrared spectroscopy for quality assessment of red meat
    Robert, Chima
    Jessep, William
    Sutton, Joshua J.
    Hicks, Talia M.
    Loeffen, Mark
    Farouk, Mustafa
    Ward, James F.
    Bain, Wendy E.
    Craigie, Cameron R.
    Fraser-Miller, Sara J.
    Gordon, Keith C.
    [J]. FOOD CHEMISTRY, 2021, 361
  • [40] Spatial Frequency Training Modulates Neural Face Processing: Learning Transfers from Low- to High-Level Visual Features
    Peters, Judith C.
    van den Boomen, Cartin
    Kemner, Chantal
    [J]. FRONTIERS IN HUMAN NEUROSCIENCE, 2017, 11