« 大家一起來抗癌 作者 / 蕭聰愋 | Main | Disruptive Civil Technologies Six Technologies with Potential Impacts on US Interests out to 2025 »

September 14, 2008

Comments

Fred

You will find that alpha-beta filter is like your first order PLL (phase lock loop) or PD (position and derivative) control. Yes, it is simple solution for single sensor feedback controller. If you have more than one sensor then the model will need to be expanded.

Tze-Chien Chu

oh, the state in Kalman filter means "2 more variable ?" .

Tze-Chien Chu

I think to use each signal a A-B Filter and another upper level system to control the A & B based on the filtered signal ? ....

Fred

The state in Kalman means a vector of variables, which typically greater than 2. Degenerative vector is a single variable which could be estimated by PI, PD or PID classic control algorithm.

If you give each variable a different control then their information is not shared hence the solution is not consider jointly optimal. For example, a 3D space position can be defined by X, Y and Z three variables. Your goal is to minimize the "position" error not just X, Y, Z individually. A multi-level control is doable but not "optimal".

Tze-Chien Chu

Yes. From the state-map's view . All data are map to a point. But my NXT is not fast. I need another strategy to solve the problem.

The comments to this entry are closed.

My Photo

October 2021

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31            
Blog powered by Typepad
Member since 10/2004