Iterative Semi-parametric Dynamics Model Learning For Autonomous Racing

被引:0
|
作者
Georgiev, Ignat [1 ,2 ]
Chatzikomis, Christoforos [1 ]
Volkl, Timo [1 ]
Smith, Joshua [2 ]
Mistry, Michael [2 ]
机构
[1] Arrival, Howald, Luxembourg
[2] Univ Edinburgh, Sch Informat, Edinburgh, Midlothian, Scotland
来源
CONFERENCE ON ROBOT LEARNING, VOL 155 | 2020年 / 155卷
关键词
Robots; Dynamics; Model Learning; Autonomous Driving;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Accurately modeling robot dynamics is crucial to safe and efficient motion control. In this paper, we develop and apply an iterative learning semi-parametric model, with a neural network, to the task of autonomous racing with a Model Predictive Controller (MPC). We present a novel non-linear semi-parametric dynamics model where we represent the known dynamics with a parametric model, and a neural network captures the unknown dynamics. We show that our model can learn more accurately than a purely parametric model and generalize better than a purely non-parametric model, making it ideal for real-world applications where collecting data from the full state space is not feasible. We present a system where the model is bootstrapped on pre-recorded data and then updated iteratively at run time. Then we apply our iterative learning approach to the simulated problem of autonomous racing and show that it can safely adapt to modified dynamics online and even achieve better performance than models trained on data from manual driving.
引用
收藏
页码:552 / 563
页数:12
相关论文
共 50 条
  • [41] Dynamic semi-parametric factor model for functional expectiles
    Petra Burdejová
    Wolfgang K. Härdle
    Computational Statistics, 2019, 34 : 489 - 502
  • [42] Testing the trajectory difference in a semi-parametric longitudinal model
    Niu, Feiyang
    Zhou, Jianhui
    Le, Thu H.
    Ma, Jennie Z.
    STATISTICAL METHODS IN MEDICAL RESEARCH, 2017, 26 (03) : 1519 - 1531
  • [43] A Bayesian semi-parametric bivariate failure time model
    Nieto-Barajas, Luis E.
    Walker, Stephen G.
    COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2007, 51 (12) : 6102 - 6113
  • [44] A Bayesian semi-parametric model for colorectal cancer incidences
    Zhang, S
    Sun, DC
    He, CZ
    Schootman, M
    STATISTICS IN MEDICINE, 2006, 25 (02) : 285 - 309
  • [45] Fuzzy weighted scaled coefficients in semi-parametric model
    Wu, JW
    Jang, JB
    Tsai, TR
    ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS, 1996, 48 (01) : 97 - 110
  • [46] Semi-parametric Approaches to Learning in Model-Based Hierarchical Control of Complex Systems
    Zafar, Munzir
    Mehmood, Areeb
    Khan, Mouhyemen
    Zhang, Shimin
    Murtaza, Muhammad
    Aladele, Victor
    Theodorou, Evangelos A.
    Hutchinson, Seth
    Boots, Byron
    PROCEEDINGS OF THE 2018 INTERNATIONAL SYMPOSIUM ON EXPERIMENTAL ROBOTICS, 2020, 11 : 387 - 397
  • [47] A Bayesian semi-parametric model for thermal proteome profiling
    Siqi Fang
    Paul D. W. Kirk
    Marcus Bantscheff
    Kathryn S. Lilley
    Oliver M. Crook
    Communications Biology, 4
  • [48] Fuzzy approach to semi-parametric sample selection model
    Safiih, L. Muhamad
    Kamil, A. A. Basah
    Osman, M. T. Abu
    APPLIED MATHEMATICS FOR SCIENCE AND ENGINEERING, 2007, : 307 - +
  • [49] A semi-parametric claims reserving model with monotone splines
    Chang, Le
    Gao, Guangyuan
    Shi, Yanlin
    ANNALS OF OPERATIONS RESEARCH, 2024,
  • [50] A semi-parametric model for censored and passively registered data
    Jonker, MA
    Van der Vaart, AW
    BERNOULLI, 2001, 7 (01) : 1 - 31