Scaffolded Learning of In-place Trotting Gait for a Quadruped Robot with Bayesian Optimization

被引:1
|
作者
Zhai, Keyan [1 ]
Li, Chu'an [1 ]
Rosendo, Andre [1 ]
机构
[1] ShanghaiTech Univ, Sch Informat Sci & Technol, 393 Middle Huaxia Rd, Shanghai 201210, Peoples R China
关键词
Quadruped locomotion; Gait optimization; Parametric search;
D O I
10.1007/978-3-030-95892-3_28
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
During learning trials, systems are exposed to different failure conditions which may break robotic parts before a safe behavior is discovered. Humans contour this problem by grounding their learning to a safer structure/control first and gradually increasing its difficulty. This paper presents the impact of a similar supports in the learning of a stable gait on a quadruped robot. Based on the psychological theory of instructional scaffolding, we provide different support settings to our robot, evaluated with strain gauges, and use Bayesian Optimization to conduct a parametric search towards a stable Raibert controller. We perform several experiments to measure the relation between constant supports and gradually reduced supports during gait learning, and our results show that a gradually reduced support is capable of creating a more stable gait than a support at a fixed height. Although gaps between simulation and reality can lead robots to catastrophic failures, our proposed method combines speed and safety when learning a new behavior.
引用
收藏
页码:365 / 373
页数:9
相关论文
共 50 条
  • [41] Heuristic Gait Learning of Quadruped Robot Based on Deep Deterministic Policy Gradient Algorithm
    Wang, Mingchao
    Ruan, Xiaogang
    Zhu, Xiaoqing
    2020 CHINESE AUTOMATION CONGRESS (CAC 2020), 2020, : 1046 - 1049
  • [42] Optimization-based parallel learning of quadruped robot locomotion skills
    Zhang, Siyuan
    Zhu, Xiaoqing
    Chen, Jiangtao
    Liu, Xinyuan
    Wang, Tao
    Qinghua Daxue Xuebao/Journal of Tsinghua University, 2024, 64 (10): : 1696 - 1705
  • [43] Research on Foothold Optimization of the Quadruped Crawling Robot based on Reinforcement Learning
    Liu X.
    Wang P.
    Dong R.
    Recent Patents on Mechanical Engineering, 2024, 17 (01) : 11 - 22
  • [44] Multi-objective Parameter CPG Optimization for Gait Generation of a Quadruped Robot Considering Behavioral Diversity
    Oliveira, Miguel
    Santos, Cristina P.
    Costa, Lino
    Matos, Itor
    Ferreira, Manuel
    2011 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, 2011,
  • [45] Bayesian Optimization Meets Riemannian Manifolds in Robot Learning
    Jaquier, Noemie
    Rozo, Leonel
    Calinon, Sylvain
    Burger, Mathias
    CONFERENCE ON ROBOT LEARNING, VOL 100, 2019, 100
  • [46] Combining Simulations and Real-Robot Experiments for Bayesian Optimization of Bipedal Gait Stabilization
    Rodriguez, Diego
    Brandenburger, Andre
    Behnke, Sven
    ROBOT WORLD CUP XXII, ROBOCUP 2018, 2019, 11374 : 70 - 82
  • [47] A study of a gait generation of a quadruped robot based on rhythmic control - optimization of CPG parameters by a fast dynamics simulation environment
    Takemura, H
    Ueda, J
    Matsumoto, Y
    Ogasawara, T
    CLIMBING AND WALKING ROBOTS, 2002, : 759 - 766
  • [48] ROBOT SKILL SYNTHESIS THROUGH HUMAN VISUO-MOTOR LEARNING Humanoid Robot Statically-stable Reaching and In-place Stepping
    Babic, Jan
    Hajdinjak, Blaz
    Oztop, Erhan
    ICINCO 2010: PROCEEDINGS OF THE 7TH INTERNATIONAL CONFERENCE ON INFORMATICS IN CONTROL, AUTOMATION AND ROBOTICS, VOL 2, 2010, : 212 - 215
  • [49] Fault-tolerant gait learning and morphology optimization of a polymorphic walking robot
    Christensen, David Johan
    Larsen, Jorgen Christian
    Stoy, Kasper
    EVOLVING SYSTEMS, 2014, 5 (01) : 21 - 32
  • [50] Development of Humanoid Robot Simulator for Gait Learning by Using Particle Swarm Optimization
    Kuo, Ping-Huan
    Ho, Ya-Fang
    Lee, Kai-Fan
    Tai, Li-Heng
    Li, Tzuu-Hseng S.
    2013 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC 2013), 2013, : 2683 - 2688