Dynamic Programming for Optimal Control

by allenlu2007

 

Bellman equation

Discrete LQR

How to solve P(t)? Dynamic Programming

P(t) 可由解 (6.13) 得到,或由 dynamic programming 來解。Dynamic programming 其實是 divide-and-conquer.

先由 optimal control (u(t), x(t), and J) for [to, tf] 開始。這個 solution 對任何 to < t1 < tf 的 [t1, tf] sub-interval 都是 optimal control 如下。

NewImage

 再來是相反由小含大

NewImage

Advertisements