extended Kalman filter
E457842
The extended Kalman filter is a state estimation algorithm that generalizes the Kalman filter to nonlinear systems by linearizing about the current estimate, widely used in robotics and control for tracking and localization.
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf |
Bayesian filter
ⓘ
nonlinear state estimation algorithm ⓘ recursive estimator ⓘ |
| approximates | nonlinear system by local linear model ⓘ |
| assumes |
Gaussian noise
ⓘ
known measurement model ⓘ known process model ⓘ |
| basedOn | Kalman filter NERFINISHED ⓘ |
| canDivergeIf |
initialization is poor
ⓘ
linearization is poor ⓘ |
| comparedTo |
particle filter
ⓘ
unscented Kalman filter NERFINISHED ⓘ |
| computes | Kalman gain ⓘ |
| generalizes | Kalman filter NERFINISHED ⓘ |
| handles |
nonlinear dynamical systems
ⓘ
nonlinear measurement models ⓘ |
| isApproximationOf | optimal Bayesian filter ⓘ |
| isTaughtIn |
control engineering courses
ⓘ
estimation and filtering courses ⓘ robotics courses ⓘ |
| isTypicallyImplementedAs | discrete-time algorithm ⓘ |
| isUsedFor |
attitude estimation
ⓘ
localization ⓘ navigation ⓘ tracking ⓘ |
| isUsedIn |
aerospace navigation
ⓘ
autonomous vehicles ⓘ control systems ⓘ robotics ⓘ sensor fusion ⓘ |
| isUsedWith |
GPS sensors
ⓘ
camera sensors ⓘ inertial measurement units ⓘ lidar sensors ⓘ radar sensors ⓘ |
| linearizesAround | current state estimate ⓘ |
| maintains |
error covariance matrix
ⓘ
state estimate ⓘ |
| originatedIn |
control theory
ⓘ
estimation theory ⓘ |
| performs |
prediction step
ⓘ
update step ⓘ |
| requires |
measurement function
ⓘ
measurement noise covariance ⓘ process noise covariance ⓘ system dynamics model ⓘ |
| uses |
Jacobian matrices
ⓘ
first-order Taylor expansion ⓘ linearization ⓘ |
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.
subject surface form:
Probabilistic Robotics