The Information filter stores and works with the inverse of the covariance matrix. The information filter has a more computationally complex prediction, and a less complex update. This makes it suitable for situations where large numbers of measurements are used for state estimates, or when the state of the system does not need to be known too frequently.
Matrix<double>
x0Initial estimate of state variables.
Matrix<double>
P0Covariance of state variable estimates.
This behaves the same as other constructors if the given boolean is false. Otherwise, in relation to the given state/covariance should satisfy:
Matrix<double>
stateThe initial estimate of the state of the system.
Matrix<double>
covThe covariance of the initial state estimate.
bool
invertedHas covariance/state been converted to information filter form?
Matrix<double>
FState transition matrix.
Performs a prediction of the next state of the Kalman Filter, where there is plant noise. The covariance matrix of the plant noise, in this case, is a square matrix corresponding to the state transition and the state of the system.
Matrix<double>
FState transition matrix.
Matrix<double>
QA plant noise covariance matrix.
Performs a prediction of the next state of the Kalman Filter, given a description of the dynamic equations of the system, the covariance of the plant noise affecting the system and the equations that describe the effect on the system of that plant noise.
Matrix<double>
FState transition matrix.
Matrix<double>
GNoise coupling matrix.
Matrix<double>
QPlant noise covariance.
Matrix
to describe the noise of the system.
Matrix<double>
zThe measurements of the system.
Matrix<double>
HMeasurement model.
Matrix<double>
RCovariance of measurements.
Examination of system state requires an inversion of the covariance matrix for the information filter, and is quite expensive for large systems.