Module control 

Control theory and systems: PID, state-space, transfer functions, Bode analysis, LQR, Kalman filter, MPC, adaptive control, fuzzy control, and neural controller.

Structs

AdaptiveController

Model Reference Adaptive System (MRAS) controller using the MIT rule for parameter adaptation.

BodeAnalysis

Bode frequency response analysis for a SISO transfer function.

FuzzyController

Mamdani fuzzy controller with triangular MFs and centroid defuzzification.

FuzzyRule

Fuzzy rule: (input_mf_index) → (output_mf_index).

KalmanFilter

Full linear Kalman filter with predict and update steps.

LqrController

Linear Quadratic Regulator (LQR): computes optimal gain matrix K via Discrete Algebraic Riccati Equation (DARE).

ModelPredictiveControl

Simple linear MPC with prediction horizon N, tracking and input penalties.

NeuralController

Simple multi-layer perceptron (MLP) neural network controller.

Pid

Proportional-integral-derivative (PID) controller with anti-windup and derivative filter.

ReplayBuffer

Simple experience replay buffer for neural controller training.

StateSpaceModel

Continuous or discrete linear state-space model: ẋ = Ax + Bu, y = Cx + Du.

TransferFunction

SISO transfer function defined by numerator and denominator polynomials (descending powers).

TriangleMf

Triangular membership function: zero outside [a, c], peak at b.

Constants

TWO_PI

Constant π exported for convenience.

Functions

controllability_matrix

Compute controllability matrix: [B, AB, A²B, …, A^(n-1)B].

pole_placement

Pole placement: compute state feedback gain K such that eigenvalues of (A - B*K) match desired poles (Ackermann’s formula, SISO only).

riccati_solve_dare

Solve the Discrete Algebraic Riccati Equation (DARE) iteratively. Returns gain matrix K = (R + BᵀPB)⁻¹ * BᵀPA.

zoh_discretize

Zero-order hold (ZOH) discretization of continuous-time A matrix. Uses Padé approximation: e^(Adt) ≈ I + Adt + (A*dt)²/2 + …