> For the complete documentation index, see [llms.txt](https://yams.yassrobotics.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://yams.yassrobotics.com/api-reference/java-reference/math/derivative-time-filter.md).

# DerivativeTimeFilter

**Package:** `yams.math`

`DerivativeTimeFilter` computes a numerical derivative of a signal while debouncing noise below a minimum rate of change. Useful for estimating velocity from a position signal.

## Constructors

```java
DerivativeTimeFilter(double initial, Time debouncerPeriod)
DerivativeTimeFilter(Time debouncerPeriod)   // Initial value = 0
```

| Parameter         | Description                                                                                               |
| ----------------- | --------------------------------------------------------------------------------------------------------- |
| `debouncerPeriod` | Minimum period over which a derivative is considered valid. Changes faster than this rate are suppressed. |

## Methods

| Method                                | Returns  | Description                                                        |
| ------------------------------------- | -------- | ------------------------------------------------------------------ |
| `derivative(double current, Time dt)` | `double` | Computes derivative given explicit elapsed time `dt`.              |
| `derivative(double current)`          | `double` | Computes derivative using FPGA clock elapsed time since last call. |

## Example

```java
DerivativeTimeFilter filter = new DerivativeTimeFilter(Seconds.of(0.02));

// In periodic():
double velocity = filter.derivative(encoder.getPosition());
```