NextControl/dev.nextftc.control.filters

Package-level declarations

Signal filters: EMAFilter (exponential moving average), MedianFilter (moving-window median), SlewRateLimiter (rate-of-change limiting), and KalmanFilter (fuses a LinearModel with noisy measurements for optimal state estimation).

Types

Debouncer

class Debouncer @JvmOverloads constructor(var debounceTimeSeconds: Double, var type: Debouncer.DebounceType = DebounceType.RISING)

A simple debounce filter for boolean streams. Requires that the boolean change value from baseline for a specified period of time before the filtered value changes.

EdgeCounterFilter

class EdgeCounterFilter(var requiredEdges: Int, var windowTimeSeconds: Double)

A rising edge counter for boolean streams. Requires that the boolean change value to true for a specified number of times within a specified time window after the first rising edge before the filtered value changes.

EMAFilter

class EMAFilter(val alpha: Double)

KalmanFilter

class KalmanFilter<States : Nat, Inputs : Nat, Outputs : Nat> @JvmOverloads constructor(val plant: LinearModel<States, Inputs, Outputs>, val stateStdDevs: Vector<States>, val measurementStdDevs: Vector<Outputs>, dt: Double = 0.05)

A Kalman filter for linear state estimation.

MedianFilter

class MedianFilter(size: Int)

A class that implements a moving-window median filter. Useful for reducing measurement noise, especially with processes that generate occasional, extreme outliers (such as values from vision processing, LIDAR, or ultrasonic sensors).

SlewRateLimiter

class SlewRateLimiter @JvmOverloads constructor(var positiveRateLimit: Double, var negativeRateLimit: Double, initialValue: Double = 0.0)

A class that limits the rate of change of an input value. Useful for implementing voltage, setpoint, and/or output ramps. A slew-rate limit is most appropriate when the quantity being controlled is a velocity or a voltage.