Extended Kalman Filter

Class implementing the extended Kalman filter algorithm.

This class implements the extended Kalman filter algorithm. It accepts a model of the explicit discrete time-invariant form:

   x(k+1) = stateEqn(x(k),u(k),dt) + N(0,Q)
     y(k) = outputEqn(x(k),u(k)) + N(0,R)

where N is the standard normal distribution with zero mean and covariance Q or R. This class implements the Observer interface.

The state and output Jacobian matrices must also be provided. These are functions, both taking as input x and u.

Methods

• initialize(EKF,t0,x0,u0) - Initialize filter given initial time, state, and inputs
• estimate(EKF,t,u,z) - Update the state and output estimates given new input and output data.
• getStateEstimate(EKF) - Return a state estimate structure with mean and covariance.

See also Kalman Filter, Unscented Kalman Filter.