Predict state estimates forward in time for
[___] = predict(___,
specifies arguments used in the state transition functions or state transition Jacobian
functions of the sensor models or the motion model used in the filter, in addition to all
arguments from the previous syntax.
Predict insEKF Filter Object
insEKF filter object. Specify the angular velocity of filter as
[.1 0 0] rad/s.
filter = insEKF; stateparts(filter,"AngularVelocity",[.1 0 0]);
Show the orientation quaternion at time t = 0 seconds.
orientation0 = quaternion(stateparts(filter,"Orientation"))
orientation0 = quaternion 1 + 0i + 0j + 0k
Predict the filter by 1 second and show the orientation quaternion.
[state, statecov] = predict(filter,1); orientation1 = quaternion(stateparts(filter,"Orientation"))
orientation1 = quaternion 0.99875 + 0.049938i + 0j + 0k
dt — Time step of prediction
Time step of prediction, specified as a positive scalar.
varargin — Additional arguments
any data type
Additional arguments passed to the state transition functions and state transition Jacobian functions of the motion model and sensor models used in the filter, specified as any data type accepted by the two functions. You can use these arguments to simulate control or drive inputs, such as a throttle.
state — Predicted state vector
N-element real-valued vector
Predicted state vector, returned as an N-element real-valued vector, where N is the dimension of the filter state.
stateCovariance — State estimate error covariance
N-by-N real-valued positive definite
State estimate error covariance, returned as an N-by-N real-valued positive definite matrix, where N is the dimension of the state.
C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.
Introduced in R2022a