Loading...
Searching...
No Matches
gtsam::ConcentratedGaussian< T > Class Template Reference
Detailed Description
template<class T>
class gtsam::ConcentratedGaussian< T >
class gtsam::ConcentratedGaussian< T >
A nonlinear density, inherits from ExtendedPriorFactor.
This class models a (left) extended concentrated Gaussian (L-ECG) with Gaussian noise models.
Inheritance diagram for gtsam::ConcentratedGaussian< T >:
Public Member Functions | |
Standard Constructors | |
| ConcentratedGaussian () | |
| Default constructor for serialization. | |
| ConcentratedGaussian (Key key, const T &origin, const sharedGaussianNoiseModel &model) | |
| Constructor with noise model and optional mean in tangent space. | |
| ConcentratedGaussian (Key key, const T &origin, const Vector &mean, const sharedGaussianNoiseModel &model) | |
| Constructor with noise model and optional mean in tangent space. | |
| ConcentratedGaussian (Key key, const T &origin, const Matrix &covariance) | |
| Constructor with covariance matrix (zero mean in tangent space). | |
| ConcentratedGaussian (Key key, const T &origin, const Vector &mean, const Matrix &covariance) | |
| Constructor with mean (in tangent space) and covariance matrix. | |
Standard Destructor | |
Testable | |
| void | print (const std::string &s, const KeyFormatter &keyFormatter=DefaultKeyFormatter) const override |
| print | |
| bool | equals (const NonlinearFactor &expected, double tol=1e-9) const override |
| equals | |
Standard API | |
| T | retractMean (Matrix *xHm) const |
| Return T element corresponding to the mean, with optional Jacobian. | |
| T | retractMean () const |
| Return the mean without requesting its Jacobian. | |
| T | retractMean (Matrix &xHm) const |
| Return the mean and write its Jacobian into xHm. | |
| double | negLogConstant () const |
| Calculate the normalization constant for the density. | |
| double | logProbability (const T &x) const |
| Calculate the log-probability of the given value. | |
| double | logProbability (const Values &values) const |
| Log-probability overload taking a Values container. | |
| double | evaluate (const T &x) const |
| Evaluate the probability density at the given value. | |
| double | evaluate (const Values &values) const |
| Evaluate density P(x) using a Values container. | |
Transport and Fusion | |
| ConcentratedGaussian | reset () const |
| Create a new ConcentratedGaussian with zero mean by moving the origin to x̂ = Retract(origin, mean). | |
| ConcentratedGaussian | transportTo (const T &x_hat) const |
| Transport this density to a new origin x̂, returning a density at x̂ with nonzero mean in that chart. | |
| ConcentratedGaussian | operator* (const ConcentratedGaussian &other) const |
| Fusion operator implementing the (approximate) three-step Fusion method in: Y. | |
| Public Member Functions inherited from gtsam::ExtendedPriorFactor< T > | |
| ExtendedPriorFactor () | |
| default constructor - only use for serialization | |
| ExtendedPriorFactor (Key key, const T &origin, const SharedNoiseModel &model) | |
| Constructor with noise model and optional mean in tangent space. | |
| ExtendedPriorFactor (Key key, const T &origin, const Vector &mean, const SharedNoiseModel &model) | |
| Constructor with noise model and optional mean in tangent space. | |
| ExtendedPriorFactor (Key key, const T &origin, const Matrix &covariance) | |
| Constructor with covariance matrix (zero mean in tangent space). | |
| ExtendedPriorFactor (Key key, const T &origin, const Vector &mean, const Matrix &covariance) | |
| Constructor with mean (in tangent space) and covariance matrix. | |
| void | print (const std::string &s, const KeyFormatter &keyFormatter=DefaultKeyFormatter) const override |
| print | |
| bool | equals (const NonlinearFactor &expected, double tol=1e-9) const override |
| equals | |
| gtsam::NonlinearFactor::shared_ptr | clone () const override |
| Vector | evaluateError (const T &x, OptionalMatrixType H) const override |
| vector of errors | |
| double | error (const T &x) const |
| Compute the negative log-likelihood of a given value. | |
| double | likelihood (const T &x) const |
| Compute the likelihood of a given value. | |
| const T & | origin () const |
| const std::optional< Vector > & | mean () const |
| noiseModel::Gaussian::shared_ptr | gaussianModel (const std::string &method="<unknown>", bool throwOnFailure=false) const |
| Get the Gaussian noise model, or return nullopt/throw if not Gaussian. | |
| std::optional< Matrix > | covariance (const std::string &method="<unknown>", bool throwOnFailure=false) const |
| Get Gaussian covariance, or return nullopt/throw if not Gaussian. | |
| std::optional< Gaussian > | gaussian (const std::string &method="<unknown>", bool throwOnFailure=false) const |
| Get a Gaussian, or return nullopt if not Gaussian and throw is false. | |
| Public Member Functions inherited from gtsam::NoiseModelFactorT< Vector, ValueTypes... > | |
| Key | key () const |
| Returns a key. | |
| Vector | unwhitenedError (const Values &x, std::vector< Matrix > &H) const |
| support taking in the actual vector instead of the pointer as well to get access to this version of the function from derived classes one will need to use the "using" keyword and specify that like this: public: using NoiseModelFactor::unwhitenedError; | |
| NoiseModelFactorT () | |
| Default Constructor for I/O. | |
| NoiseModelFactorT (const SharedNoiseModel &noiseModel, KeyType< ValueTypes >... keys) | |
| Constructor. | |
| NoiseModelFactorT (const SharedNoiseModel &noiseModel, CONTAINER keys) | |
| Constructor. | |
| virtual Vector | evaluateError (const ValueTypes &... x, OptionalMatrixTypeT< ValueTypes >... H) const=0 |
| Override evaluateError to finish implementing an n-way factor. | |
| Vector | evaluateError (const ValueTypes &... x, MatrixTypeT< ValueTypes > &... H) const |
| If all the optional arguments are matrices then redirect the call to the one which takes pointers. | |
| Vector | evaluateError (const ValueTypes &... x) const |
| No-Jacobians requested function overload. | |
| AreAllMatrixRefs< Vector, OptionalJacArgs... > | evaluateError (const ValueTypes &... x, OptionalJacArgs &&... H) const |
| Some (but not all) optional Jacobians are omitted (function overload) and the jacobians are l-value references to matrices. | |
| AreAllMatrixPtrs< Vector, OptionalJacArgs... > | evaluateError (const ValueTypes &... x, OptionalJacArgs &&... H) const |
| Some (but not all) optional Jacobians are omitted (function overload) and the jacobians are pointers to matrices. | |
| Key | key1 () const |
| Key | key2 () const |
| Key | key3 () const |
| Key | key4 () const |
| Key | key5 () const |
| Key | key6 () const |
| std::shared_ptr< GaussianFactor > | linearize (const Values &values) const override |
| Linearize factors whose error and argument dimensions are all fixed to an arbitrary-arity FixedJacobianFactor. | |
| Vector | unwhitenedError (const Values &x, OptionalMatrixVecType H=nullptr) const override |
| This implements the unwhitenedError virtual function by calling the n-key specific version of evaluateError, which is pure virtual so must be implemented in the derived class. | |
| Public Member Functions inherited from gtsam::NoiseModelFactor | |
| NoiseModelFactor () | |
| Default constructor for I/O only. | |
| ~NoiseModelFactor () override | |
| Destructor. | |
| template<typename CONTAINER> | |
| NoiseModelFactor (const SharedNoiseModel &noiseModel, const CONTAINER &keys) | |
| Constructor. | |
| void | print (const std::string &s="", const KeyFormatter &keyFormatter=DefaultKeyFormatter) const override |
| Print. | |
| bool | equals (const NonlinearFactor &f, double tol=1e-9) const override |
| Check if two factors are equal. | |
| size_t | dim () const override |
| get the dimension of the factor (number of rows on linearization) | |
| const SharedNoiseModel & | noiseModel () const |
| access to the noise model | |
| Vector | unwhitenedError (const Values &x, std::vector< Matrix > &H) const |
| support taking in the actual vector instead of the pointer as well to get access to this version of the function from derived classes one will need to use the "using" keyword and specify that like this: public: using NoiseModelFactor::unwhitenedError; | |
| Vector | whitenedError (const Values &c) const |
| Vector of errors, whitened This is the raw error, i.e., i.e. | |
| Vector | unweightedWhitenedError (const Values &c) const |
| Vector of errors, whitened, but unweighted by any loss function. | |
| double | weight (const Values &c) const |
| Compute the effective weight of the factor from the noise model. | |
| double | error (const Values &c) const override |
| Calculate the error of the factor. | |
| shared_ptr | cloneWithNewNoiseModel (const SharedNoiseModel newNoise) const |
| Creates a shared_ptr clone of the factor with a new noise model. | |
| double | error (const HybridValues &c) const override |
| All factor types need to implement an error function. | |
| NonlinearFactor () | |
| Default constructor for I/O only. | |
| template<typename CONTAINER> | |
| NonlinearFactor (const CONTAINER &keys) | |
| Constructor from a collection of the keys involved in this factor. | |
| void | print (const std::string &s="", const KeyFormatter &keyFormatter=DefaultKeyFormatter) const override |
| print | |
| double | error (const HybridValues &c) const override |
| All factor types need to implement an error function. | |
| virtual bool | active (const Values &c) const |
| Checks whether a factor should be used based on a set of values. | |
| virtual void | qcqpFactors (NonlinearFactorGraph *costs, NonlinearEqualityConstraints *constraints, size_t columnDimension=1) const |
| Add this factor's QCQP cost and constraints over matrix-valued QCQP variables with the given column dimension. | |
| virtual shared_ptr | rekey (const std::map< Key, Key > &rekey_mapping) const |
| Creates a shared_ptr clone of the factor with different keys using a map from old->new keys. | |
| virtual shared_ptr | rekey (const KeyVector &new_keys) const |
| Clones a factor and fully replaces its keys. | |
| virtual bool | sendable () const |
| Should the factor be evaluated in the same thread as the caller This is to enable factors that has shared states (like the Python GIL lock). | |
| Public Member Functions inherited from gtsam::Factor | |
| virtual | ~Factor ()=default |
| Default destructor. | |
| bool | empty () const |
| Whether the factor is empty (involves zero variables). | |
| Key | front () const |
| First key. | |
| Key | back () const |
| Last key. | |
| const_iterator | find (Key key) const |
| find | |
| const KeyVector & | keys () const |
| Access the factor's involved variable keys. | |
| const_iterator | begin () const |
| Iterator at beginning of involved variable keys. | |
| const_iterator | end () const |
| Iterator at end of involved variable keys. | |
| size_t | size () const |
| virtual void | printKeys (const std::string &s="Factor", const KeyFormatter &formatter=DefaultKeyFormatter) const |
| print only keys | |
| bool | equals (const This &other, double tol=1e-9) const |
| check equality | |
| KeyVector & | keys () |
| iterator | begin () |
| Iterator at beginning of involved variable keys. | |
| iterator | end () |
| Iterator at end of involved variable keys. | |
Public Types | |
| using | Base = ExtendedPriorFactor<T> |
| using | Gaussian = typename Base::Gaussian |
| using | sharedGaussianNoiseModel = noiseModel::Gaussian::shared_ptr |
| Public Types inherited from gtsam::ExtendedPriorFactor< T > | |
| typedef T | T |
| typedef ExtendedPriorFactor< T > | This |
| Mean in the tangent space, default nullopt. | |
| using | Gaussian |
| Simple, non-templated Gaussian fusion in a common tangent space. | |
| Public Types inherited from gtsam::NoiseModelFactorT< Vector, ValueTypes... > | |
| using | ValueType |
| The type of the I'th template param can be obtained as ValueType. | |
| Public Types inherited from gtsam::NoiseModelFactor | |
| typedef std::shared_ptr< This > | shared_ptr |
| Noise model. | |
| Public Types inherited from gtsam::NonlinearFactor | |
| typedef std::shared_ptr< This > | shared_ptr |
| Public Types inherited from gtsam::Factor | |
| typedef KeyVector::iterator | iterator |
| Iterator over keys. | |
| typedef KeyVector::const_iterator | const_iterator |
| Const iterator over keys. | |
Additional Inherited Members | |
| Static Public Attributes inherited from gtsam::NoiseModelFactorT< Vector, ValueTypes... > | |
| static constexpr auto | N |
| N is the number of variables (N-way factor). | |
| Protected Types inherited from gtsam::ExtendedPriorFactor< T > | |
| typedef NoiseModelFactorN< T > | Base |
| Protected Types inherited from gtsam::NoiseModelFactorT< Vector, ValueTypes... > | |
| using | Base |
| using | This |
| using | OptionalMatrixTypeT |
| using | KeyType |
| using | MatrixTypeT |
| using | IsConvertible |
| using | IndexIsValid |
| using | ContainerElementType |
| using | IsContainerOfKeys |
| using | AreAllMatrixRefs |
| A helper alias to check if a list of args are all references to a matrix or not. | |
| using | IsMatrixPointer |
| using | IsNullpointer |
| using | AreAllMatrixPtrs |
| A helper alias to check if a list of args are all pointers to a matrix or not. | |
| Protected Types inherited from gtsam::NoiseModelFactor | |
| typedef NonlinearFactor | Base |
| typedef NoiseModelFactor | This |
| Protected Types inherited from gtsam::NonlinearFactor | |
| typedef Factor | Base |
| typedef NonlinearFactor | This |
| Protected Member Functions inherited from gtsam::NoiseModelFactor | |
| NoiseModelFactor (const SharedNoiseModel &noiseModel) | |
| Constructor - only for subclasses, as this does not set keys. | |
| Factor () | |
| Default constructor for I/O. | |
| template<typename CONTAINER> | |
| Factor (const CONTAINER &keys) | |
| Construct factor from container of keys. | |
| template<typename ITERATOR> | |
| Factor (ITERATOR first, ITERATOR last) | |
| Construct factor from iterator keys. | |
| template<typename CONTAINER> | |
| static Factor | FromKeys (const CONTAINER &keys) |
| Construct factor from container of keys. | |
| template<typename ITERATOR> | |
| static Factor | FromIterators (ITERATOR first, ITERATOR last) |
| Construct factor from iterator keys. | |
| Protected Attributes inherited from gtsam::ExtendedPriorFactor< T > | |
| T | origin_ |
| std::optional< Vector > | mean_ |
| The point in manifold at which tangent space is rooted. | |
| Protected Attributes inherited from gtsam::NoiseModelFactor | |
| SharedNoiseModel | noiseModel_ |
| Protected Attributes inherited from gtsam::Factor | |
| KeyVector | keys_ |
| The keys involved in this factor. | |
Member Function Documentation
◆ equals()
template<class T>
|
inlineoverridevirtual |
equals
Reimplemented from gtsam::NonlinearFactor.
◆ evaluate()
template<class T>
|
inline |
Evaluate the probability density at the given value.
P(x) = exp(logProbability(x)).
◆ logProbability() [1/2]
template<class T>
|
inline |
Calculate the log-probability of the given value.
error(x) as defined for a GTSAM factor already equals 0.5 * ||r(x)||^2_Σ (i.e. the negative log-likelihood without the normalization constant). Hence: log P(x) = log k - error(x).
◆ logProbability() [2/2]
template<class T>
|
inline |
Log-probability overload taking a Values container.
This mirrors the linear GaussianConditional interface so densities can be queried in a uniform way when only a Values is available.
◆ negLogConstant()
template<class T>
|
inline |
◆ operator*()
template<class T>
|
inline |
Fusion operator implementing the (approximate) three-step Fusion method in: Y.
Ge, P. van Goor and R. Mahony, "A Geometric Perspective on Fusing Gaussian Distributions on Lie Groups," in IEEE Control Systems Letters, vol. 8, pp. 844-849, 2024, https://ieeexplore.ieee.org/document/10539262
We choose this->origin_ as the reference, express other density in our chart, fuse the Gaussians, then reset to a zero-mean concentrated Gaussian.
◆ print()
template<class T>
|
inlineoverridevirtual |
Reimplemented from gtsam::Factor.
◆ reset()
template<class T>
|
inline |
Create a new ConcentratedGaussian with zero mean by moving the origin to x̂ = Retract(origin, mean).
Returns an ECG with origin=x̂, zero mean, and covariance transported to x̂.
◆ transportTo()
template<class T>
|
inline |
Transport this density to a new origin x̂, returning a density at x̂ with nonzero mean in that chart.
Uses a full first-order Jacobian for the change of coordinates between charts via the chain rule: J = ∂Local(x̂,x)/∂x · ∂Retract(origin,m)/∂m
The documentation for this class was generated from the following file:
- /tmp/gtsam-4.3.0-doxygen.rsXPUS/source/gtsam/nonlinear/ConcentratedGaussian.h
