Confusing description of built-in estimator noise model on MPC documentation page

2 次查看（过去 30 天）
Bill Tubbs
Bill Tubbs 2022-10-18
编辑： Bill Tubbs 2022-10-18
I'm confused by the following text which appears on the documentation page: 'Implement Custom State Estimator Equivalent to Built-In Kalman Filter':
In this scenario, there are three additive noises in the Kalman filter design: (1) process noise added to the manipulated variable; (2) process noise added to the input of God; (3) measurement noise added to the input of Gmn:
[ Gd.B(1) 0 0 ] [ wn1 ]
w[k] = [ Gd.B(2) 0 0 ] * [ wn2 ] = B_est * white noise
[ 0 God.B 0 ] [ wn3 ]
v[k] = [ Gd.D God.D Gmn.D ] * wn4 = D_est * white noise
To me, it looks like there are 4 white noises (wn1, wn2, wn3, wn4), not 3. It would make sense to me if it was written instead as:
[ Gd.B(1) 0 0 ] [ wn1 ]
w[k] = [ Gd.B(2) 0 0 ] * [ wn2 ] = B_est * white noise
[ 0 God.B 0 ] [ wn3 ]
[ wn1 ]
v[k] = [ Gd.D God.D Gmn.D ] * [ wn2 ] = D_est * white noise
[ wn3 ]
Maybe I am misunderstanding it.
For info, when I run the code, I get size(B_est) is 3x3 and size(D_est) is 1x3.

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Model Predictive Control Toolbox 的更多信息

标签

产品


版本

R2021b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by