Learning Agent Interactions from Density Evolution in 3D Regions With Obstacles

被引:0
|
作者
Tirumalai, Amoolya [1 ,2 ]
Mavridis, Christos N. [1 ,2 ]
Baras, John S. [1 ,2 ]
机构
[1] Univ Maryland, Dept Elect & Comp Engn, College Pk, MD 20742 USA
[2] Univ Maryland, Inst Syst Res, College Pk, MD 20742 USA
来源
2023 62ND IEEE CONFERENCE ON DECISION AND CONTROL, CDC | 2023年
关键词
PARTICLE; BEHAVIORS; ENSEMBLE; DYNAMICS; SCHEMES; SYSTEMS; ORDER;
D O I
10.1109/CDC49753.2023.10384269
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this work, we study the inverse problem of identifying complex flocking dynamics in a domain cluttered with obstacles. We get inspiration from animal flocks moving in complex ways with capabilities far beyond what current robots can do. Owing to the difficulty of observing and recovering the trajectories of the agents, we focus on the dynamics of their probability densities, which are governed by partial differential equations (PDEs), namely compressible Euler equations subject to non-local forces. We formulate the inverse problem of learning interactions as a PDE-constrained optimization problem of minimizing the squared Hellinger distance between the histogram of the flock and the distribution associated to our PDEs. The numerical methods used to efficiently solve the PDE-constrained optimization problem are described. Realistic flocking data are simulated using the Boids model of flocking agents, which differs in nature from the reconstruction models used in our PDEs. Our analysis and simulated experiments show that the behavior of cohesive flocks can be recovered accurately with approximate PDE solutions.
引用
收藏
页码:7156 / 7161
页数:6
相关论文
共 50 条
  • [11] Evolution of the regions of the 3D particle motion in the regular polygon problem of (N+1) bodies with a quasi-homogeneous potential
    Fakis, Demetrios
    Kalvouridis, Tilemahos
    ASTROPHYSICS AND SPACE SCIENCE, 2017, 362 (09)
  • [12] A Sustainability 3D Framework of the 20 Regions of Italy and Comparison With World Countries
    Pulselli, Federico M.
    Sani, Alessandro
    Goffetti, Giulia
    Coscieme, Luca
    Bastianoni, Simone
    FRONTIERS IN ENERGY RESEARCH, 2019, 7
  • [13] High density 3D printed microfluidic valves, pumps, and multiplexers
    Gong, Hua
    Woolley, Adam T.
    Nordin, Gregory P.
    LAB ON A CHIP, 2016, 16 (13) : 2450 - 2458
  • [14] Cubic interactions of arbitrary spin fields in 3d flat space
    Metsaev, R. R.
    JOURNAL OF PHYSICS A-MATHEMATICAL AND THEORETICAL, 2020, 53 (44)
  • [15] Machine Learning Emulation of 3D Cloud Radiative Effects
    Meyer, David
    Hogan, Robin J.
    Dueben, Peter D.
    Mason, Shannon L.
    JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS, 2022, 14 (03)
  • [16] Combining unsupervised learning and discrimination for 3D action recognition
    Chen, Guang
    Clarke, Daniel
    Giuliani, Manuel
    Gaschler, Andre
    Knoll, Alois
    SIGNAL PROCESSING, 2015, 110 : 67 - 81
  • [17] 3D Multi-Agent Formation Control with Rigid Body Maneuvers
    Zhang, Pengpeng
    de Queiroz, Marcio
    ASIAN JOURNAL OF CONTROL, 2019, 21 (03) : 1088 - 1099
  • [18] From a 3D Passive Biped Walker to a 3D Passivity-Based Controlled Robot
    Beigzadeh, Borhan
    Sabaapour, Mohammad Reza
    Yazdi, Mohammad Reza Hairi
    Raahemifar, Kaamran
    INTERNATIONAL JOURNAL OF HUMANOID ROBOTICS, 2018, 15 (04)
  • [19] 3D holography: from discretum to continuum
    Bonzom, Valentin
    Dittrich, Bianca
    JOURNAL OF HIGH ENERGY PHYSICS, 2016, (03):
  • [20] Learning-based robust stabilization for reduced-order models of 2D and 3D Boussinesq equations
    Benosman, Mouhacine
    Borggaard, Jeff
    San, Omer
    Kramer, Boris
    APPLIED MATHEMATICAL MODELLING, 2017, 49 : 162 - 181