Constraint Class Reference

A constraint is a function + lower and upper bounds. More...

#include <drake/solvers/Constraint.h>

Inheritance diagram for Constraint:

Collaboration diagram for Constraint:

Public Member Functions
	Constraint (size_t num_constraints)
template<typename DerivedLB , typename DerivedUB >
	Constraint (size_t num_constraints, Eigen::MatrixBase< DerivedLB > const &lb, Eigen::MatrixBase< DerivedUB > const &ub)
virtual	~Constraint ()
virtual void	eval (const Eigen::Ref< const Eigen::VectorXd > &x, Eigen::VectorXd &y) const =0
virtual void	eval (const Eigen::Ref< const Drake::TaylorVecXd > &x, Drake::TaylorVecXd &y) const =0
Eigen::VectorXd const &	lower_bound () const
Eigen::VectorXd const &	upper_bound () const
size_t	num_constraints () const

Protected Attributes
Eigen::VectorXd	lower_bound_
Eigen::VectorXd	upper_bound_

Detailed Description

A constraint is a function + lower and upper bounds.

Some thoughts: It should support evaluating the constraint, adding it to an optimization problem, and have support for constraints that require slack variables (adding additional decision variables to the problem). There should also be some notion of parameterized constraints: e.g. the acceleration constraints in the rigid body dynamics are constraints on vdot and f, but are "parameterized" by q and v.

Constructor & Destructor Documentation

Constraint ( size_t  num_constraints )

inlineexplicit

Constraint ( size_t  num_constraints, Eigen::MatrixBase< DerivedLB > const &  lb, Eigen::MatrixBase< DerivedUB > const &  ub  )

inline

virtual ~Constraint ( )

inlinevirtual

Here is the call graph for this function:

Member Function Documentation

virtual void eval ( const Eigen::Ref< const Eigen::VectorXd > &  x, Eigen::VectorXd &  y  ) const

pure virtual

Implemented in LinearComplementarityConstraint, BoundingBoxConstraint, LinearConstraint, PolynomialConstraint, QuasiStaticConstraintWrapper, QuadraticConstraint, SingleTimeKinematicConstraintWrapper, and DynamicConstraint.

Here is the caller graph for this function:

virtual void eval ( const Eigen::Ref< const Drake::TaylorVecXd > &  x, Drake::TaylorVecXd &  y  ) const

pure virtual

Implemented in LinearComplementarityConstraint, BoundingBoxConstraint, LinearConstraint, PolynomialConstraint, QuadraticConstraint, and DynamicConstraint.

Eigen::VectorXd const& lower_bound ( ) const

inline

size_t num_constraints ( ) const

inline

Here is the caller graph for this function:

Eigen::VectorXd const& upper_bound ( ) const

inline

Member Data Documentation

Eigen::VectorXd lower_bound_

protected

Eigen::VectorXd upper_bound_

protected

---

The documentation for this class was generated from the following file:

• drake/solvers/Constraint.h