Standard RK integrator. More...
#include <crocoddyl/core/integrator/rk.hpp>
Public Types | |
typedef ActionDataAbstractTpl< Scalar > | ActionDataAbstract |
typedef IntegratedActionModelAbstractTpl< Scalar > | Base |
typedef ControlParametrizationModelAbstractTpl< Scalar > | ControlParametrizationModelAbstract |
typedef IntegratedActionDataRKTpl< Scalar > | Data |
typedef DifferentialActionModelAbstractTpl< Scalar > | DifferentialActionModelAbstract |
typedef MathBaseTpl< Scalar > | MathBase |
typedef MathBase::MatrixXs | MatrixXs |
typedef MathBase::VectorXs | VectorXs |
Public Member Functions | |
IntegratedActionModelRKTpl (boost::shared_ptr< DifferentialActionModelAbstract > model, boost::shared_ptr< ControlParametrizationModelAbstract > control, const RKType rktype, const Scalar time_step=Scalar(1e-3), const bool with_cost_residual=true) | |
Initialize the RK integrator. More... | |
IntegratedActionModelRKTpl (boost::shared_ptr< DifferentialActionModelAbstract > model, const RKType rktype, const Scalar time_step=Scalar(1e-3), const bool with_cost_residual=true) | |
Initialize the RK integrator. More... | |
virtual void | calc (const boost::shared_ptr< ActionDataAbstract > &data, const Eigen::Ref< const VectorXs > &x) |
Integrate the total cost value for nodes that depends only on the state using RK scheme. More... | |
virtual void | calc (const boost::shared_ptr< ActionDataAbstract > &data, const Eigen::Ref< const VectorXs > &x, const Eigen::Ref< const VectorXs > &u) |
Integrate the differential action model using RK scheme. More... | |
virtual void | calcDiff (const boost::shared_ptr< ActionDataAbstract > &data, const Eigen::Ref< const VectorXs > &x) |
Compute the partial derivatives of the cost. More... | |
virtual void | calcDiff (const boost::shared_ptr< ActionDataAbstract > &data, const Eigen::Ref< const VectorXs > &x, const Eigen::Ref< const VectorXs > &u) |
Compute the partial derivatives of the RK integrator. More... | |
virtual bool | checkData (const boost::shared_ptr< ActionDataAbstract > &data) |
Checks that a specific data belongs to this model. | |
virtual boost::shared_ptr< ActionDataAbstract > | createData () |
Create the RK integrator data. More... | |
std::size_t | get_ni () const |
Return the number of nodes of the integrator. | |
virtual void | print (std::ostream &os) const |
Print relevant information of the RK integrator model. More... | |
virtual void | quasiStatic (const boost::shared_ptr< ActionDataAbstract > &data, Eigen::Ref< VectorXs > u, const Eigen::Ref< const VectorXs > &x, const std::size_t maxiter=100, const Scalar tol=Scalar(1e-9)) |
Computes the quasic static commands. More... | |
Public Attributes | |
EIGEN_MAKE_ALIGNED_OPERATOR_NEW typedef _Scalar | Scalar |
Protected Attributes | |
boost::shared_ptr< ControlParametrizationModelAbstract > | control_ |
Model of the control parametrization. | |
boost::shared_ptr< DifferentialActionModelAbstract > | differential_ |
< Control parametrization | |
std::size_t | nu_ |
< Differential action model | |
boost::shared_ptr< StateAbstract > | state_ |
< Dimension of the control | |
Scalar | time_step2_ |
< Model of the state | |
Scalar | time_step_ |
< Square of the time step used for integration | |
bool | with_cost_residual_ |
< Time step used for integration | |
Standard RK integrator.
It applies a standard RK integration schemes to a differential (i.e., continuous time) action model. The available integrators are: RK2, RK3, and RK4.
This standard RK scheme introduces also the possibility to parametrize the control trajectory inside an integration step, for instance using polynomials. This requires introducing some notation to clarify the difference between the control inputs of the differential model and the control inputs to the integrated model. We have decided to use \(\mathbf{w}\) to refer to the control inputs of the differential model and \(\mathbf{u}\) for the control inputs of the integrated action model.
IntegratedActionModelRKTpl | ( | boost::shared_ptr< DifferentialActionModelAbstract > | model, |
boost::shared_ptr< ControlParametrizationModelAbstract > | control, | ||
const RKType | rktype, | ||
const Scalar | time_step = Scalar(1e-3) , |
||
const bool | with_cost_residual = true |
||
) |
Initialize the RK integrator.
[in] | model | Differential action model |
[in] | control | Control parametrization |
[in] | rktype | Type of RK integrator |
[in] | time_step | Step time (default 1e-3) |
[in] | with_cost_residual | Compute cost residual (default true) |
IntegratedActionModelRKTpl | ( | boost::shared_ptr< DifferentialActionModelAbstract > | model, |
const RKType | rktype, | ||
const Scalar | time_step = Scalar(1e-3) , |
||
const bool | with_cost_residual = true |
||
) |
Initialize the RK integrator.
This initialization uses ControlParametrizationPolyZeroTpl
for the control parametrization.
[in] | model | Differential action model |
[in] | rktype | Type of RK integrator |
[in] | time_step | Step time (default 1e-3) |
[in] | with_cost_residual | Compute cost residual (default true) |
|
virtual |
Integrate the differential action model using RK scheme.
[in] | data | RK integrator data |
[in] | x | State point \(\mathbf{x}\in\mathbb{R}^{ndx}\) |
[in] | u | Control input \(\mathbf{u}\in\mathbb{R}^{nu}\) |
|
virtual |
Integrate the total cost value for nodes that depends only on the state using RK scheme.
It computes the total cost and defines the next state as the current one. This function is used in the terminal nodes of an optimal control problem.
[in] | data | RK integrator data |
[in] | x | State point \(\mathbf{x}\in\mathbb{R}^{ndx}\) |
|
virtual |
Compute the partial derivatives of the RK integrator.
[in] | data | RK integrator data |
[in] | x | State point \(\mathbf{x}\in\mathbb{R}^{ndx}\) |
[in] | u | Control input \(\mathbf{u}\in\mathbb{R}^{nu}\) |
|
virtual |
Compute the partial derivatives of the cost.
It updates the derivatives of the cost function with respect to the state only. This function is used in the terminal nodes of an optimal control problem.
[in] | data | RK integrator data |
[in] | x | State point \(\mathbf{x}\in\mathbb{R}^{ndx}\) |
|
virtual |
Create the RK integrator data.
|
virtual |
Computes the quasic static commands.
The quasic static commands are the ones produced for a the reference posture as an equilibrium point, i.e. for \(\mathbf{f^q_x}\delta\mathbf{q}+\mathbf{f_u}\delta\mathbf{u}=\mathbf{0}\)
[in] | data | RK integrator data |
[out] | u | Quasic static commands |
[in] | x | State point (velocity has to be zero) |
[in] | maxiter | Maximum allowed number of iterations |
[in] | tol | Tolerance |
|
virtual |
Print relevant information of the RK integrator model.
[out] | os | Output stream object |