/Data_Fusion_Course

The solution of the data fusion course of SJTU. MTALAB code for KF/UKF/EKF/PF/FKF/DKF. 数据融合技术，卡尔曼滤波KF/无迹卡尔曼滤波UKF/拓展卡尔曼滤波EKF等的MATLAB实现

Primary LanguageMATLABMIT LicenseMIT

Data Fusion

This are code repo for the Data Fusion course.

Table of Contents

Optimal Estimation

Problem description

Suppose a voltage is a random variable $X$ with normal distribution, the mean value is $5$, and the variance is $0.1$; The random variable x is measured $20$ times by two instruments, and the measurement error of the two instruments is assumed to be a normally distributed random variable with a mean value of $0$ and a variance of $0.1$ and $0.4$ respectively. Caculate the least square estimation (LSE), weighted least square estimation (WLS) and linear minimum variance estimation (LMMSE) of $X$, and calculate the mean square error of the corresponding estimation. Let the measurement equation be $Z=HZ+V$.

Usage

To handle the problem, run the following file:

1/code_1/main123.m

Wiener Filter

problem description

Let $y (n) =x (n) +v (n)$, where $x(n)=10sin(\frac{\pi n}{128}+\frac{\pi}{3})$,$v(n)$ is white noise with variance of $1.25$. Design FIR and IIR Wiener filter to estimate the signal $x (n)$.

Usage

To handle the problem, run the following file:

1/code_1/main.m

Kalman Filter

Basic Kalman Filter

1/code_3/kalman.m

Constant Gain Kalman Filter

1/code_3/kalman_constant_gain.m

Square root Kalman Filter

1/code_3/kalman_sqrt.m

Forgetting Factor Kalman Filter

1/code_3/kalman_forgetting_factor.m

Adaptive Kalman Filter

1/code_3/kalman_adaptive.m

Limited K Reduction Kalman Filter

1/code_3/kalman_restain_K.m

Extended Kalman Filter

2/code_0/EKF.m

Unscented Kalman Filter

2/code_0/UKF.m

Particle Filter

2/code_0/PF.m

Federated Kalman Filter

2/code_1/federated_filter.m

Decentralized Kalman filter

2/code_1/center_federated_filter.m

Fuzzy Control

Basic method

4/code/TS_model.m

T-S method

4/code/TS_model.m

Contact

changjingliu@sjtu.edu.cn