arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1906.04057eess.SYcs.SY
收藏

使用 NMPC 和策略梯度迈向安全的强化学习:第一部分 - 随机情形

Towards Safe Reinforcement Learning Using NMPC and Policy Gradients: Part I - Stochastic case

Sebastien Gros, Mario Zanon

AI总结:

本文提出一种基于 actor-critic 和鲁棒 NMPC 的随机策略梯度部署方法,通过参数化优化与硬约束在连续空间中有效解决安全策略的采样评估难题,并确保学习全过程的系统安全性。

摘要:

我们提出了一种部署随机策略梯度方法的方法论,使用 actor-critic 技术,当最优策略通过参数化优化问题近似时,允许通过硬约束来实施安全。对于连续输入空间,对随机策略施加安全限制可能使其密度的采样和评估变得困难。本文提出了一种计算有效的方法来解决该问题。我们将专注于基于鲁棒非线性模型预测控制(NMPC)的策略近似,其中安全可以被显式处理。为简洁起见,我们将仅在鲁棒线性 MPC 背景下详细说明安全策略。向非线性情形的扩展是可能的但更为复杂。我们还将额外提出一种在鲁棒线性 MPC 背景下在整个学习过程中维持系统安全的技术。本文有一篇处理确定性策略梯度情形的姊妹论文。

英文摘要:

We present a methodology to deploy the stochastic policy gradient method, using actor-critic techniques, when the optimal policy is approximated using a parametric optimization problem, allowing one to enforce safety via hard constraints. For continuous input spaces, imposing safety restrictions on the stochastic policy can make the sampling and evaluation of its density difficult. This paper proposes a computationally effective approach to solve that issue. We will focus on policy approximations based on robust Nonlinear Model Predictive Control (NMPC), where safety can be treated explicitly. For the sake of brevity, we will detail safe policies in the robust linear MPC context only. The extension to the nonlinear case is possible but more complex. We will additionally present a technique to maintain the system safety throughout the learning process in the context of robust linear MPC. This paper has a companion paper treating the deterministic policy gradient case.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑