optas.builder.OptimizationBuilder Class Reference

OptimizationBuilder allows you to build/specify an optimization problem. More...

Public Member Functions
def	__init__ (self, int T, List[RobotModel] robots=[], List[TaskModel] tasks=[], bool derivs_align=False)
	OptimizationBuilder constructor. More...
List[str]	get_model_names (self)
	Return the names of each model. More...
int	get_model_index (self, str name)
	Return the index of the model in the list of models. More...
Model	get_model (self, str name)
	Return the model with given name. More...
CasADiArrayType	get_model_state (self, str name, int t, int time_deriv=0)
	Get the model state at a given time. More...
CasADiArrayType	get_model_states (self, str name, int time_deriv=0)
	Get the full state trajectory for a given model. More...
CasADiArrayType	get_model_parameters (self, str name, int time_deriv=0)
	Get the array of parameters for a given model. More...
CasADiArrayType	get_model_parameter (self, str name, int t, int time_deriv=0)
	Get the model parameter at a given time. More...
CasADiArrayType	get_robot_states_and_parameters (self, str name, int time_deriv=0)
	Get the vector of states and parameters for a given model. More...
cs.DM	is_cost_quadratic (self)
	True DM(1) when cost function is quadratic in the decision variables, False DM(0) otherwise. More...
cs.SX	add_decision_variables (self, str name, int m=1, int n=1, bool is_discrete=False)
	Add decision variables to the optimization problem. More...
cs.SX	add_parameter (self, str name, int m=1, int n=1)
	Add a parameter to the optimization problem. More...
None	add_cost_term (self, str name, cs.SX cost_term)
	Add cost term to the optimization problem. More...
None	add_geq_inequality_constraint (self, str name, CasADiArrayType lhs, Union[None, CasADiArrayType] rhs=None)
	Add the inequality constraint lhs >= rhs to the optimization problem. More...
None	add_leq_inequality_constraint (self, str name, CasADiArrayType lhs, Union[None, CasADiArrayType] rhs=None)
	Add the inequality constraint lhs <= rhs to the optimization problem. More...
None	add_bound_inequality_constraint (self, str name, CasADiArrayType lhs, CasADiArrayType mid, CasADiArrayType rhs)
	Add the inequality constraint lhs <= mid <= rhs to the optimization problem. More...
None	add_equality_constraint (self, str name, CasADiArrayType lhs, Union[None, CasADiArrayType] rhs=None, reduce_constraint=False)
	Add the equality constraint lhs == rhs to the optimization problem. More...
None	sphere_collision_avoidance_constraints (self, name, obstacle_names, link_names=None, base_link=None)
	Add collision avoidance constraints, based on spherical representations attached to the robot links and placed in the world. More...
None	integrate_model_states (self, str name, int time_deriv, CasADiArrayType dt)
	Integrates the model states over time. More...
None	enforce_model_limits (self, str name, int time_deriv=0, Union[None, CasADiArrayType] lo=None, Union[None, CasADiArrayType] up=None, safe_frac=1.0)
	Enforce model limits. More...
None	initial_configuration (self, str name, Union[None, CasADiArrayType] init=None, int time_deriv=0)
	Set initial configuration. More...
None	fix_configuration (self, str name, CasADiArrayType config=None, int time_deriv=0, int t=0)
	Fix configuration. More...
Optimization	build (self)
	Build the optimization problem. More...

Public Attributes
	T
	Number of time steps for the trajectory. More...
	derivs_align
	When true, the time derivatives align for each time step. More...

Private Member Functions
CasADiArrayType	_x (self)
	Return the decision variables as a casadi.SX vector. More...
CasADiArrayType	_p (self)
	Return the parameters as a casadi.SX vector. More...
cs.DM	_is_linear_in_x (self, cs.SX y)
	Returns true DM(1) if y is a linear function of the decision variables, false DM(0) otherwise. More...
CasADiArrayType	_cost (self)
	Returns the cost function. More...

Private Attributes
	_models
	List of models. More...
	_decision_variables
	SXContainer containing decision variables. More...
	_parameters
	SXContainer containing parameters. More...
	_cost_terms
	SXContainer containing the cost terms. More...
	_lin_eq_constraints
	SXContainer containing the linear equality constraints. More...
	_lin_ineq_constraints
	SXContainer containing the linear inequality constraints. More...
	_ineq_constraints
	SXContainer containing the inequality constraints. More...
	_eq_constraints
	SXContainer containing the equality constraints. More...

Detailed Description

OptimizationBuilder allows you to build/specify an optimization problem.

Constructor & Destructor Documentation

__init__()

def optas.builder.OptimizationBuilder.__init__	(		self,
		int	T,
		List[RobotModel]	robots = [],
		List[TaskModel]	tasks = [],
		bool	derivs_align = False
	)

OptimizationBuilder constructor.

   @param T Number of time steps for the trajectory.
   @param robots A list of robot models.
   @param tasks A list of task models.
   @param derivs_align When true, the time derivatives align for each time step. Default is False.
   @return An instance of the OptimizationBuilder class.

Member Function Documentation

_cost()

CasADiArrayType optas.builder.OptimizationBuilder._cost (   self )

private

Returns the cost function.

   @return The symbolic cost function.

_is_linear_in_x()

cs.DM optas.builder.OptimizationBuilder._is_linear_in_x (   self, cs.SX  y  )

private

Returns true DM(1) if y is a linear function of the decision variables, false DM(0) otherwise.

   @param y Symbolic function of interest.
   @return True if y is linear in x.

_p()

CasADiArrayType optas.builder.OptimizationBuilder._p (   self )

private

Return the parameters as a casadi.SX vector.

   @return Symbolic parameters.

_x()

CasADiArrayType optas.builder.OptimizationBuilder._x (   self )

private

Return the decision variables as a casadi.SX vector.

   @return Symbolic decision variables.

add_bound_inequality_constraint()

None optas.builder.OptimizationBuilder.add_bound_inequality_constraint	(		self,
		str	name,
		CasADiArrayType	lhs,
		CasADiArrayType	mid,
		CasADiArrayType	rhs
	)

Add the inequality constraint lhs <= mid <= rhs to the optimization problem.

   @param name Name for the constraint.
   @param lhs Left-hand side for the inequality constraint.
   @param mid Middle part of the inequality constraint.
   @param rhs Right-hand side for the inequality constraint.

add_cost_term()

None optas.builder.OptimizationBuilder.add_cost_term	(		self,
		str	name,
		cs.SX	cost_term
	)

Add cost term to the optimization problem.

   @param name Name for cost function.
   @param cost_term Cost term, must be an array with shape 1-by-1.

add_decision_variables()

cs.SX optas.builder.OptimizationBuilder.add_decision_variables	(		self,
		str	name,
		int	m = 1,
		int	n = 1,
		bool	is_discrete = False
	)

Add decision variables to the optimization problem.

   @param name Name of decision variable array.
   @param m Number of rows in decision variable array. Default is 1.
   @param n Number of columns in decision variable array. Default is 1.
   @param is_discret If true, then the decision variables are treated as discrete variables. Default is False.
   @return Array of the decision variables.

add_equality_constraint()

None optas.builder.OptimizationBuilder.add_equality_constraint	(		self,
		str	name,
		CasADiArrayType	lhs,
		Union[None, CasADiArrayType]	rhs = None,
			reduce_constraint = False
	)

Add the equality constraint lhs == rhs to the optimization problem.

   @param name Name for the constraint.
   @param lhs Left-hand side for the inequality constraint.
   @param rhs Right-hand side for the inequality constraint. If None, then it is replaced with the zero array with the same shape as lhs.
   @param reduce_constraint When True, the constraints are specified by ||lhs - rhs||^2 = 0. This has the effect of reducing the number of equality constraints in the problem. However, if the constraints are linear they will be treated as nonlinear. Default is False.

add_geq_inequality_constraint()

None optas.builder.OptimizationBuilder.add_geq_inequality_constraint	(		self,
		str	name,
		CasADiArrayType	lhs,
		Union[None, CasADiArrayType]	rhs = None
	)

Add the inequality constraint lhs >= rhs to the optimization problem.

   @param name Name for the constraint.
   @param lhs Left-hand side for the inequality constraint.
   @param rhs Right-hand side for the inequality constraint. If None, then it is replaced with the zero array with the same shape as lhs.

add_leq_inequality_constraint()

None optas.builder.OptimizationBuilder.add_leq_inequality_constraint	(		self,
		str	name,
		CasADiArrayType	lhs,
		Union[None, CasADiArrayType]	rhs = None
	)

Add the inequality constraint lhs <= rhs to the optimization problem.

   @param name Name for the constraint.
   @param lhs Left-hand side for the inequality constraint.
   @param rhs Right-hand side for the inequality constraint. If None, then it is replaced with the zero array with the same shape as lhs.

add_parameter()

cs.SX optas.builder.OptimizationBuilder.add_parameter	(		self,
		str	name,
		int	m = 1,
		int	n = 1
	)

Add a parameter to the optimization problem.

   @param name Name of parameter array.
   @param m Number of rows in parameter array. Default is 1.
   @param n Number of columns in parameter array. Default is 1.
   @return Array of the parameters.

build()

Optimization optas.builder.OptimizationBuilder.build

(

self

)

Build the optimization problem.

enforce_model_limits()

None optas.builder.OptimizationBuilder.enforce_model_limits	(		self,
		str	name,
		int	time_deriv = 0,
		Union[None, CasADiArrayType]	lo = None,
		Union[None, CasADiArrayType]	up = None,
			safe_frac = 1.0
	)

Enforce model limits.

   @param name Name of model.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.).
   @param lo Lower limits, if None then model limits specified in the model class are used.
   @param up Upper limits, if None then model limits specified in the model class are used.
   @param safe_frac When safe_frac < 1.0, the joint limits are reduced by a fraction of their original values.

fix_configuration()

None optas.builder.OptimizationBuilder.fix_configuration	(		self,
		str	name,
		CasADiArrayType	config = None,
		int	time_deriv = 0,
		int	t = 0
	)

Fix configuration.

   @param name Name of model.
   @param config The configuration. When None is passed then it is considered to be zero.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.).
   @param t Index for the configuration in trajectory (by default this is the first element but it could also be the last for example in moving horizon estimation).

get_model()

Model optas.builder.OptimizationBuilder.get_model	(		self,
		str	name
	)

Return the model with given name.

   @param name Name of the model.
   @return A task or robot model.

get_model_index()

int optas.builder.OptimizationBuilder.get_model_index	(		self,
		str	name
	)

Return the index of the model in the list of models.

   @param name Name of the model.
   @return Index of the model in the list of models.

get_model_names()

List[str] optas.builder.OptimizationBuilder.get_model_names

(

self

)

Return the names of each model.

   @return List of model names.

get_model_parameter()

CasADiArrayType optas.builder.OptimizationBuilder.get_model_parameter	(		self,
		str	name,
		int	t,
		int	time_deriv = 0
	)

Get the model parameter at a given time.

   @param name Name of the model.
   @param t Index of the desired state.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.).
   @return The parameter vector where dim is the model number of parameters

get_model_parameters()

CasADiArrayType optas.builder.OptimizationBuilder.get_model_parameters	(		self,
		str	name,
		int	time_deriv = 0
	)

Get the array of parameters for a given model.

   @param name Name of the model.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.).
   @return The array of parameters where dim is the number of model parameters, and T is the number of time-steps in the trajectory.

get_model_state()

CasADiArrayType optas.builder.OptimizationBuilder.get_model_state	(		self,
		str	name,
		int	t,
		int	time_deriv = 0
	)

Get the model state at a given time.

   @param name Name of the model.
   @param t Index of the desired state.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.)
   @return The state vector where dim is the model dimension.

get_model_states()

CasADiArrayType optas.builder.OptimizationBuilder.get_model_states	(		self,
		str	name,
		int	time_deriv = 0
	)

Get the full state trajectory for a given model.

   @param name Name of the model.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.).
   @return The state vector where dim is the model dimension, and T is the number of time-steps in the trajectory.

get_robot_states_and_parameters()

CasADiArrayType optas.builder.OptimizationBuilder.get_robot_states_and_parameters	(		self,
		str	name,
		int	time_deriv = 0
	)

Get the vector of states and parameters for a given model.

   Note that method only applies to to RobotModel.
   To be replaced by get_model_states_and_parameters once parameters are added to base class Model.

   @param name Name of the model.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.)
   @return The vector of parameters where dim is the number of model states and parameters (for a robot it should correspond to the degrees of freedom), and T is the number of time-steps in the trajectory.

initial_configuration()

None optas.builder.OptimizationBuilder.initial_configuration	(		self,
		str	name,
		Union[None, CasADiArrayType]	init = None,
		int	time_deriv = 0
	)

Set initial configuration.

   @param name Name of model.
   @param init Initial configuration. If None is passed then this is assumed to be zero.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.). The default is 0.

integrate_model_states()

None optas.builder.OptimizationBuilder.integrate_model_states	(		self,
		str	name,
		int	time_deriv,
		CasADiArrayType	dt
	)

Integrates the model states over time.

   @param name Name of the model.
   @param time_deriv The time-deriviative required (i.e. position is 0, velocity is 1, etc.).
   @param dt Integration time step. If the value is a scalar then the value is used as the time step for the trajectory. When an array is passed, it should have the correct length: ```n = T - (1 if derivs_align else time_deriv)``` where T and derivs_align are flags passed to the optimization builder constructor.

is_cost_quadratic()

cs.DM optas.builder.OptimizationBuilder.is_cost_quadratic

(

self

)

True DM(1) when cost function is quadratic in the decision variables, False DM(0) otherwise.

   @return Truth value if the cost function is quadratic or not.

sphere_collision_avoidance_constraints()

None optas.builder.OptimizationBuilder.sphere_collision_avoidance_constraints	(		self,
			name,
			obstacle_names,
			link_names = None,
			base_link = None
	)

Add collision avoidance constraints, based on spherical representations attached to the robot links and placed in the world.

   @param name Name of the model.
   @param obstacle_names A list of strings with names of obstacles. For each object, parameters will be created with labels NAME_position (size 3) and NAME_radii (size 1) that represent the obstacle position and radius respectively.
   @param link_names The links you want to attach collision avoidance constraints onto; for each link a parameter is created with label NAME_radii (size 1). When None is passed, a sphere is attcahed to each link specified in the URDF.
   @param base_link Name of the base frame link. When None is passed, the root link in the URDF is used.

Member Data Documentation

_cost_terms

optas.builder.OptimizationBuilder._cost_terms

private

SXContainer containing the cost terms.

_decision_variables

optas.builder.OptimizationBuilder._decision_variables

private

SXContainer containing decision variables.

_eq_constraints

optas.builder.OptimizationBuilder._eq_constraints

private

SXContainer containing the equality constraints.

_ineq_constraints

optas.builder.OptimizationBuilder._ineq_constraints

private

SXContainer containing the inequality constraints.

_lin_eq_constraints

optas.builder.OptimizationBuilder._lin_eq_constraints

private

SXContainer containing the linear equality constraints.

_lin_ineq_constraints

optas.builder.OptimizationBuilder._lin_ineq_constraints

private

SXContainer containing the linear inequality constraints.

_models

optas.builder.OptimizationBuilder._models

private

List of models.

_parameters

optas.builder.OptimizationBuilder._parameters

private

SXContainer containing parameters.

derivs_align

optas.builder.OptimizationBuilder.derivs_align

When true, the time derivatives align for each time step.

T

optas.builder.OptimizationBuilder.T

Number of time steps for the trajectory.

---

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

• builder.py