Optimal regulation and reinforcement learning for the nonholonomic integrator
Optimal regulation and reinforcement learning for the nonholonomic integrator
复制标题
非完整积分器的最优调节和强化学习
DOI：
--
复制
发表时间：
2000
期刊：
影响因子：
--
通讯作者：
R. Brockett
中科院分区：
文献类型：
--
作者：
K. Morgansen;R. Brockett
基于Hamilton-Jacobi-Bellman方程的强化学习方法已被证明对线性系统是有效的。我们考虑将这些方法推广到一类线性化不可控的非线性系统。针对连续时间和离散时间三维非完整积分器，提出并验证了基于光滑齐次范数的贴现无限水平代价函数的最优值。
Reinforcement learning methods based on the Hamilton-Jacobi-Bellman equation have proven to be effective for linear systems. We consider the extension of these methods to a class of nonlinear systems whose linearizations are not controllable. Optimal values for a discounted, infinite horizon cost function based on a smooth homogeneous norm are proposed and validated both for the continuous-time and for the discrete-time three-dimensional nonholonomic integrator.