The Value of Planning for Infinite-Horizon Model Predictive Control

被引:5
|
作者
Hatch, Nathan
Boots, Byron
机构
来源
2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021) | 2021年
关键词
D O I
10.1109/ICRA48506.2021.9561718
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Model Predictive Control (MPC) is a classic tool for optimal control of complex, real-world systems. Although it has been successfully applied to a wide range of challenging tasks in robotics, it is fundamentally limited by the prediction horizon, which, if too short, will result in myopic decisions. Recently, several papers have suggested using a learned value function as the terminal cost for MPC. If the value function is accurate, it effectively allows MPC to reason over an infinite horizon. Unfortunately, Reinforcement Learning (RL) solutions to value function approximation can be difficult to realize for robotics tasks. In this paper, we suggest a more efficient method for value function approximation that applies to goal-directed problems, like reaching and navigation. In these problems, MPC is often formulated to track a path or trajectory returned by a planner. However, this strategy is brittle in that unexpected perturbations to the robot will require replanning, which can be costly at runtime. Instead, we show how the intermediate data structures used by modern planners can be interpreted as an approximate value function. We show that that this value function can be used by MPC directly, resulting in more efficient and resilient behavior at runtime.
引用
收藏
页码:7372 / 7378
页数:7
相关论文
共 50 条
  • [31] Parametrized Infinite-Horizon Model Predictive Control for Linear Time-Invariant Systems with Input and State Constraints
    Muehlebach, Michael
    D'Andrea, Raffaello
    2016 AMERICAN CONTROL CONFERENCE (ACC), 2016, : 2669 - 2674
  • [32] INFINITE-HORIZON DYNAMIC-PROGRAMMING MODELS - PLANNING-HORIZON FORMULATION
    MORTON, TE
    OPERATIONS RESEARCH, 1979, 27 (04) : 730 - 742
  • [33] Pseudospectral methods for infinite-horizon optimal control problems
    Fahroo, Fariba
    Ross, I. Michael
    JOURNAL OF GUIDANCE CONTROL AND DYNAMICS, 2008, 31 (04) : 927 - 936
  • [34] Infinite-horizon minimax control with pointwise cost functional
    Piccardi, C.
    Journal of Optimization Theory and Applications, 1993, 78 (02): : 317 - 336
  • [35] Model Checking Optimal Infinite-horizon Control for Probabilistic Gene Regulatory Networks
    Wang, Lisong
    Feng, Tao
    Song, Junhua
    Guo, Zonghao
    Hu, Jun
    IEEE ACCESS, 2018, 6 : 77299 - 77307
  • [36] On One Adjoint Trajectory in Infinite-Horizon Control Problems
    D. V. Khlopin
    Proceedings of the Steklov Institute of Mathematics, 2024, 327 (Suppl 1) : S155 - S173
  • [37] A framework for receding-horizon control in infinite-horizon aggregative games
    Fele, Filiberto
    De Paola, Antonio
    Angeli, David
    Strbac, Goran
    ANNUAL REVIEWS IN CONTROL, 2018, 45 : 191 - 204
  • [38] Optimal infinite-horizon control for probabilistic Boolean networks
    Pal, Ranadip
    Datta, Aniruddha
    Dougherty, Edward R.
    IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2006, 54 (06) : 2375 - 2387
  • [39] VALUE AND POLICY FUNCTION APPROXIMATIONS IN INFINITE-HORIZON OPTIMIZATION PROBLEMS
    Gnecc, Giorgio
    Sanguineti, Marcello
    JOURNAL OF DYNAMICAL SYSTEMS AND GEOMETRIC THEORIES, 2008, 6 (02) : 123 - 147
  • [40] INFINITE-HORIZON MINIMAX CONTROL WITH POINTWISE COST FUNCTIONAL
    PICCARDI, C
    JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS, 1993, 78 (02) : 317 - 336