update README.md

README.md

@@ -12,9 +12,10 @@ This library implements following features:
 * Extended Kalman Filter (when using functions for dynamics and observation matrixes)
 * Correlation Matrix
 
-## Demos
+## Demos/Examples
 
 * [Browser interactive constant acceleration on bikes](http://piercus.github.io/kalman-filter)
+* [Sinusoidale Extended Kalman-Filter](https://observablehq.com/d/a033acc0859cc0de)
 * [Code pen GPS Data smoothing with constant speed](https://codepen.io/piercus/pen/wvoqPww)
 * [Partial Observation](https://github.com/piercus/kalman-filter/issues/34)
 * [Smooth 3x3 rotation matrix](https://github.com/piercus/kalman-filter/issues/37)
@@ -136,6 +137,7 @@ For advanced usage, here is the correspondance table with the matrix name of the
 | $H_k$, the observation model | `observation.stateProjection` |
 | $Q_k$, the covariance of the process noise | `dynamic.covariance` |
 | $R_k$, the covariance of the observation noise | `observation.covariance` |
+| $B_k u_k$, the control-input model multiplied by the control vector | `dynamic.constant` |
 |$\mathbf{P}_{0\mid 0}$| `dynamic.init.covariance` |
 |$\mathbf{x}_{0\mid 0}$| `dynamic.init.mean` |
 
@@ -151,11 +153,45 @@ Available default models as :
 
 This will automatically configure the `dynamic.transition` matrix.
 
-| `dynamic.name` | State | Transition Equation | `dynamic.transition` |
-|--|--|--|--|
-| `constant-position` | $\begin{bmatrix} x_t \end{bmatrix}$ | $x_t \sim x_{t-1}$ | $\begin{bmatrix} 1 \end{bmatrix}$ |
-| `constant-speed` | $\begin{bmatrix} x_t \\\\ speed_t \end{bmatrix}$ | $\begin{split}x_t \sim x_{t-1} + speed_{t-1},\\\\ speed_t \sim speed_{t-1}\end{split}$| $\begin{bmatrix} 1 & 1 \\\\ 0 & 1 \end{bmatrix}$ |
-| `constant-acceleration` | $\begin{bmatrix} x \\\\ speed_t \\\\ acc_t \end{bmatrix}$ | $\begin{split}x_t \sim x_{t-1} + speed_{t-1} \\\\ speed_t \sim speed_{t-1} + acc_{t-1} \\\\ acc_t \sim acc_{t-1}\end{split}$ | $\begin{bmatrix} 1 & 1 & 0 \\\\ 0 & 1 & 1 \\\\ 0 & 0 & 1\end{bmatrix}$ |
+##### constant-position
+
+```math
+\begin{align}
+State :& \begin{bmatrix} x_t \end{bmatrix}\\ 
+Transition Equation :& x_t \sim x_{t-1} \\ 
+dynamic.transition :& \begin{bmatrix} 1 \end{bmatrix}
+\end{align}
+```
+
+##### constant-speed
+
+```math
+\begin{align}
+State :& \begin{bmatrix} x_t \\ speed_t \end{bmatrix} \\ 
+Transition Equation :& 
+\begin{split}
+x_t &\sim x_{t-1} + speed_{t-1},\\ 
+speed_t &\sim speed_{t-1}
+\end{split} \\ 
+dynamic.transition :& \begin{bmatrix} 1 & 1 \\ 0 & 1 \end{bmatrix}
+\end{align}
+```
+
+##### constant-acceleration
+
+```math
+\begin{align}
+State :& \begin{bmatrix} x_t \\ speed_t \\ acc_t \end{bmatrix} \\ 
+
+Transition Equation :& 
+\begin{split}
+x_t &\sim x_{t-1} + speed_{t-1} \\ 
+speed_t &\sim speed_{t-1} + acc_{t-1} \\ 
+acc_t &\sim acc_{t-1}
+\end{split} \\ 
+dynamic.transition :& \begin{bmatrix} 1 & 1 & 0 \\ 0 & 1 & 1 \\ 0 & 0 & 1\end{bmatrix}
+\end{align}
+```
 
 #### 'constant-position' on 2D data
 
@@ -337,6 +373,7 @@ In order to use the Kalman-Filter with a dynamic or observation model which is n
 * `observation.covariance`
 * `dynamic.transition`
 * `dynamic.covariance`
+* `dynamic.constant`
 
 In this situation this `function` will return the value of the matrix at each step of the kalman-filter.
 
@@ -421,7 +458,14 @@ You will need to put your non-linear functions in the following parameters
 * `observation.fn`
 * `dynamic.fn`
 
-See an example in `test/issues/56.js`
+See an example in [Sinusoidale Extended Kalman-Filter](https://observablehq.com/d/a033acc0859cc0de)
+
+### Using Control model
+
+If you want to add a constant parameter in the dynamic model (also called `control input`), you can use `dynamic.constant` function
+
+See an example code in `demo/bouncing-ball` or the result in [Bouncing Ball example](https://observablehq.com/d/a033acc0859cc0de)
+
 
 ## Use your kalman filter
 
@@ -456,7 +500,7 @@ If you want to use KalmanFilter in more advanced usage, you might want to dissoc
 let previousCorrected = null;
 const results = [];
 observations.forEach(observation => {
-	const predictedState = kFilter.predict({
+	const predicted = kFilter.predict({
 		previousCorrected
 	});
 
@@ -589,7 +633,7 @@ let previousCorrected = null;
 const results = [];
 
 observations.forEach(observation => {
-	const predictedState = kFilter.predict({
+	const predicted = kFilter.predict({
 		previousCorrected
 	});
 
@@ -620,7 +664,7 @@ let previousCorrected = null;
 const results = [];
 
 observations.forEach(observation => {
-	const predictedState = kFilter.predict({
+	const predicted = kFilter.predict({
 		previousCorrected
 	});
 

package.json

@@ -1,6 +1,6 @@
 {
   "name": "kalman-filter",
-  "version": "1.10.1",
+  "version": "1.12.0",
   "description": "Kalman filter (and Extended Kalman Filter) Multi-dimensional implementation in Javascript",
   "main": "index.js",
   "scripts": {
@@ -14,6 +14,10 @@
     "type": "git",
     "url": "git+https://github.com/piercus/kalman-filter.git"
   },
+  "files": [
+    "index.js",
+    "lib"
+  ],
   "release": {
     "branches": [
       "master"