Optimal Cislunar Architecture Design Using Monte Carlo Tree Search Methods

被引:3
作者
Klonowski, Michael [1 ]
Holzinger, Marcus J. [1 ]
Fahrner, Naomi Owens [2 ]
机构
[1] Univ Colorado, Smead Aerosp Engn Sci, 3775 Discovery Dr, Boulder, CO 80303 USA
[2] Ball Aerosp, 10 Longs Peak Dr, Broomfield, CO 80021 USA
关键词
Monte Carlo Tree Search; Space domain awareness; Reinforcement learning; Cislunar architecture; MULTIOBJECTIVE OPTIMIZATION;
D O I
10.1007/s40295-023-00383-x
中图分类号
V [航空、航天];
学科分类号
08 ; 0825 ;
摘要
A novel multi-objective Monte Carlo Tree Search (MO-MCTS) algorithm is developed and implemented for use in architecture design problems. This algorithm is used with two well-known problems with known solutions in order to verify its performance. It is then used in a highly nonlinear Cislunar architecture design problem with no known analytical solutions. The results of this implementation display the ability of MO-MCTS to effectively navigate the state space of mixed integer nonlinear programming problems and emphasize the versatility of MO-MCTS for designing critical Cislunar architecture.
引用
收藏
页数:30
相关论文
共 31 条
  • [1] Wind farm layout optimization using adaptive evolutionary algorithm with Monte Carlo Tree Search reinforcement learning
    Bai, Fangyun
    Ju, Xinglong
    Wang, Shouyi
    Zhou, Wenyong
    Liu, Feng
    [J]. ENERGY CONVERSION AND MANAGEMENT, 2022, 252
  • [2] Mixed-integer nonlinear optimization
    Belotti, Pietro
    Kirches, Christian
    Leyffer, Sven
    Linderoth, Jeff
    Luedtke, James
    Mahajan, Ashutosh
    [J]. ACTA NUMERICA, 2013, 22 : 1 - 131
  • [3] Pymoo: Multi-Objective Optimization in Python']Python
    Blank, Julian
    Deb, Kalyanmoy
    [J]. IEEE ACCESS, 2020, 8 : 89497 - 89509
  • [4] A Survey of Monte Carlo Tree Search Methods
    Browne, Cameron B.
    Powley, Edward
    Whitehouse, Daniel
    Lucas, Simon M.
    Cowling, Peter I.
    Rohlfshagen, Philipp
    Tavener, Stephen
    Perez, Diego
    Samothrakis, Spyridon
    Colton, Simon
    [J]. IEEE TRANSACTIONS ON COMPUTATIONAL INTELLIGENCE AND AI IN GAMES, 2012, 4 (01) : 1 - 43
  • [5] Chaslot G., 2008, P AAAI C ARTIFICIAL, V4, P216
  • [6] Multi-objective design of optical systems for space situational awareness
    Coder, Ryan D.
    Holzinger, Marcus J.
    [J]. ACTA ASTRONAUTICA, 2016, 128 : 669 - 684
  • [7] Couetoux Adrien, 2011, Learning and Intelligent Optimization. 5th International Conference, LION 5. Selected Papers, P433, DOI 10.1007/978-3-642-25566-3_32
  • [8] A fast and elitist multiobjective genetic algorithm: NSGA-II
    Deb, K
    Pratap, A
    Agarwal, S
    Meyarivan, T
    [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 (02) : 182 - 197
  • [9] Duffy L., 2022, 2022 IEEE INT SYST C, P1
  • [10] Sensor tasking in the cislunar regime using Monte Carlo Tree Search
    Fedeler, Samuel
    Holzinger, Marcus
    Whitacre, William
    [J]. ADVANCES IN SPACE RESEARCH, 2022, 70 (03) : 792 - 811