4.1. Hodographic shaping¶
The hodographic shapebased method shapes the velocity of the spacecraft using cylindrical coordinates. Two different implementations of this shaping method exists, one using time as independent variable, and the other using polar angle. The first one has been implemented here. This shaping method relies on the combination of integrable and differentiable base functions to shape the spacecraft velocity. The contribution of each of those base functions is weighted by a coefficient. Three base functions per velocity component at least are required so that their associated coefficients can be chosen to ensure the boundary conditions are fulfilled. The addition of any other base function, and thus of socalled free coefficients, introduces an extra degree of freedom in the problem, making it possible to optimise the shapebased trajectory to minimise the required deltaV.
4.1.1. Setting up a hodographically shaped trajectory¶
A lowthrust trajectory designed by hodographic shaping can be created with the HodographicShaping
class, which inherits from the base class ShapeBasedMethodLeg
. A settings class for hodographic shaping has also been implemented (HodographicShapingLegSettings
, see Setting up a lowthrust trajectory for more details) to construct HodographicShaping
objects more easily.

class
HodographicShaping
¶
The HodographicShaping
class is defined as follows:
HodographicShaping(
const Eigen::Vector6d initialState,
const Eigen::Vector6d finalState,
const double timeOfFlight,
const int numberOfRevolutions,
simulation_setup::NamedBodyMap& bodyMap,
const std::string bodyToPropagate,
const std::string centralBody,
std::vector< std::shared_ptr< shape_based_methods::BaseFunctionHodographicShaping > >& radialVelocityFunctionComponents,
std::vector< std::shared_ptr< shape_based_methods::BaseFunctionHodographicShaping > >& normalVelocityFunctionComponents,
std::vector< std::shared_ptr< shape_based_methods::BaseFunctionHodographicShaping > >& axialVelocityFunctionComponents,
const Eigen::VectorXd freeCoefficientsRadialVelocityFunction,
const Eigen::VectorXd freeCoefficientsNormalVelocityFunction,
const Eigen::VectorXd freeCoefficientsAxialVelocityFunction,
std::shared_ptr< numerical_integrators::IntegratorSettings< double > > integratorSettings =
std::shared_ptr< numerical_integrators::IntegratorSettings< > >( ) )
where the inputs are:
initialState
State of the spacecraft at departure.
finalState
State of the spacecraft at arrival.
timeOfFlight
Time of flight required for the shapebased trajectory (boundary condition automatically fulfilled from the way the hodographic shaping is implemented).
numberOfRevolutions
Required number of revolutions before the spacecraft reaches its final state.
bodyMap
Map of pointers to
Body
objects involved in the lowthrust trajectory.
bodyToPropagate
Name of the spacecraft to be propagated.
centralBody
Name of the central body of the lowthrust trajectory.
radialVelocityFunctionComponents
Vector of
BaseFunctionHodographicShaping
objects, containing the definition of the base functions used to shape the radial velocity of the spacecraft during the lowthrust trajectory.
normalVelocityFunctionComponents
Vector of
BaseFunctionHodographicShaping
objects, used to shape the normal velocity of the spacecraft during the lowthrust trajectory.
axialVelocityFunctionComponents
Vector of
BaseFunctionHodographicShaping
objects, used to shape the axial velocity of the spacecraft during the lowthrust trajectory.
freeCoefficientsRadialVelocityFunction
Vector of free coefficients for the radial velocity function (weighting coefficients for the additional base functions (any base function starting from the fourth one, if more than 3 base functions are provided in
radialVelocityFunctionComponents
)).
freeCoefficientsNormalVelocityFunction
Vector of free coefficients for the normal velocity function (weighting coefficients for the additional base functions (any base function starting from the fourth one, if more than 3 base functions are provided in
normalVelocityFunctionComponents
)).
freeCoefficientsAxialVelocityFunction
Vector of free coefficients for the axial velocity function (weighting coefficients for the additional base functions (any base function starting from the fourth one, if more than 3 base functions are provided in
axialVelocityFunctionComponents
)).
integratorSettings
Integrator settings (empty by default), used to propagate the spacecraft mass or thrust profiles, or to numerically propagate the fully perturbed trajectory (as a means to assess the quality of the analytical shapedbased preliminary design).
4.1.2. Hodographic shaping base functions¶
The constructor of the HodographicShaping
class requires as inputs one vector of BaseFunctionHodographicShaping
objects for each of the three velocity components. All the base functions to hodographically shape a trajectory are derived from the base class BaseFunctionHodographicShaping
. The shaping functions for the three velocity components are not fixed in hodographic shaping, but rather are left to be selected by the user (to be chosen among a set of base functions). All these hodographic shaping base functions are defined in the classes presented below.

class
BaseFunctionHodographicShaping
¶
This is the base class to derive the base functions used for hodographic shaping. Each class derived from this base class retains the following methods:
evaluateFunction
Returns the value of the base function for a specific independent variable value.
evaluateDerivative
Returns the derivative function of the base function at a specific independent variable value.
evaluateIntegral
Returns the integral function of the base function at the specific independent variable.
The following classes inherit from the base class BaseFunctionHodographicShaping
:

class
ConstantFunctionHodographicShaping
¶
Class for the constant function. The class is defined as follows and the function always returns the value 1.0:
ConstantFunctionHodographicShaping( )

class
SineFunctionHodographicShaping
¶
Class for the sine function. The class is defined as follows:
SineFunctionHodographicShaping( double frequency )
where the input is the frequency of the sine function.

class
CosineFunctionHodographicShaping
¶
Class for the cosine function. The class is defined as follows:
CosineFunctionHodographicShaping( double frequency )
where the input is the frequency of the cosine function.

class
ExponentialFunctionHodographicShaping
¶
Class for the exponential function. The class is defined as follows:
ExponentialFunctionHodographicShaping( double exponent )
where the input is the multiplying coefficient of the independent variable within the exponential function.

class
ScaledExponentialFunctionHodographicShaping
¶
Class for the scaled exponential function. The class is defined as follows:
ScaledExponentialFunctionHodographicShaping( double exponent, double scaleFactor )
where the inputs are the multiplying coefficient of the independent variable within the exponential function, and a scaling factor to be applied in front of the multiplying coefficient.

class
ExponentialSineFunctionHodographicShaping
¶
Class for the exponential sine function. The class is defined as follows:
ExponentialSineFunctionHodographicShaping( double exponentExponentialFunction, double frequencySineFunction )
where the inputs are, in order, the exponent coefficient of the exponential function, and the frequency of the sine function. The exponential sine function returns the product of the sine and exponential functions.

class
ScaledExponentialSineFunctionHodographicShaping
¶
Class for the scaled exponential sine function. The class is defined as follows:
ScaledExponentialSineFunctionHodographicShaping( double exponentExponentialFunction, double frequencySineFunction, double scaleFactor )
where the inputs are the exponent of the exponential function, the frequency of the sine function, and a scaling factor (to be applied to the exponential function, as implemented in ScaledExponentialFunctionHodographicShaping
). This function returns the product of the scaled exponential function with the sine function.

class
ExponentialCosineFunctionHodographicShaping
¶
Class for the exponential cosine function. The class is defined as follows:
ExponentialCosineFunctionHodographicShaping( double exponentExponentialFunction, double frequencyCosineFunction )
where the inputs are the exponent of the exponential function, and the frequency of the cosine function. This function returns the product of the cosine and exponential functions.

class
ScaledExponentialCosineFunctionHodographicShaping
¶
Class for the scaled exponential cosine function. The class is defined as follows:
ScaledExponentialCosineFunctionHodographicShaping( double exponentExponentialFunction, double frequencyCosineFunction, double scaleFactor )
where the inputs are the exponent of the exponential function, the frequency of the cosine function, and a scaling factor (to be applied to the exponential function, as implemented in ScaledExponentialFunctionHodographicShaping
). This function returns the product of the scaled exponential function with the cosine function.

class
PowerFunctionHodographicShaping
¶
Class for the power function. The class is defined as follows:
PowerFunctionHodographicShaping( double exponent )
where the input parameter is the exponent of the power function. This function returns the value of the input parameter to the power indicated by the input variable.

class
ScaledPowerFunctionHodographicShaping
¶
Class for the scaled power function. The class is defined as follows:
ScaledPowerFunctionHodographicShaping( double exponent, double scaleFactor )
where the inputs are the exponent of the power function, and a scaling factor to be applied in front of this power function. This function returns the product between a scaling factor and the power function, as defined by the class PowerFunctionHodographicShaping
.

class
PowerSineFunctionHodographicShaping
¶
Class for the power sine function. The class is defined as follows:
PowerSineFunctionHodographicShaping( double exponentPowerFunction, double frequencySineFunction )
where the inputs are the exponent of the power function and the frequency of the sine function. This function returns the product of the sine and the power functions (defined by the classes SineFunctionHodographicShaping
and PowerFunctionHodographicShaping
, respectively).

class
ScaledPowerSineFunctionHodographicShaping
¶
Class for the scaled power sine function. The class is defined as follows:
ScaledPowerSineFunctionHodographicShaping( double exponentPowerFunction, double frequencySineFunction, double scaleFactor )
where the inputs are the exponent of the power function, the frequency of the sine function, and a scaling factor (to be applied to the power function, as implemented in ScaledPowerFunctionHodographicShaping
). This function returns the product of the scaled power function (ScaledPowerFunctionHodographicShaping
) with the sine function (SineFunctionHodographicShaping
).

class
PowerCosineFunctionHodographicShaping
¶
Class for the power cosine function. The class is defined as follows:
PowerCosineFunctionHodographicShaping( double exponentPowerFunction, double frequencyCosineFunction )
where the inputs are the exponent of the power function and the frequency of the cosine function. This function returns the product of the cosine and the power functions (CosineFunctionHodographicShaping
and PowerFunctionHodographicShaping
, respectively).

class
ScaledPowerCosineFunctionHodographicShaping
¶
Class for the scaled power cosine function. The class is defined as follows:
ScaledPowerCosineFunctionHodographicShaping( double exponentPowerFunction, double frequencyCosineFunction, double scaleFactor )
where the inputs are the exponent of the power function, the frequency of the cosine function, and a scaling factor (to be applied to the power function, as implemented in ScaledPowerFunctionHodographicShaping
). This function returns the product of the cosine function (CosineFunctionHodographicShaping
) with the scaled power function (ScaledPowerFunctionHodographicShaping
).
4.1.3. Setting up the hodographic base functions¶
To facilitate the creation of the base functions described above, settings classes have been implemented for all of them. A BaseFunctionHodographicShaping
object can then be created from the corresponding settings with the function createBaseFunctionHodographicShaping
. It takes the type of the base function, as well as a BaseFunctionHodographicShapingSettings
object as inputs.

class
BaseFunctionHodographicShapingSettings
¶
This is the base class for base function settings for hodographic shaping. Several base function settings classes inherit from this base class:

class
TrigonometricFunctionHodographicShapingSettings
¶
Class of settings for cosine and sine base functions. It is defined as follows, taking the frequency of the trigonometric function as input:
TrigonometricFunctionHodographicShapingSettings( const double frequency )

class
ExponentialFunctionHodographicShapingSettings
¶
Class of settings for exponential functions (either simply exponential or scaled exponential functions). It is defined as follows (the scaled factor is by default set to 1):
ExponentialFunctionHodographicShapingSettings( const double exponent,
const double scaleFactor )

class
ExponentialTimesTrigonometricFunctionHodographicShapingSettings
¶
Class of settings for base functions obtained by multiplying exponential functions with trigonometric ones (This includes the following hodographic shaping base functions: ExponentialSineFunctionHodographicShaping
, ExponentialCosineFunctionHodographicShaping
,
ScaledExponentialSineFunctionHodographicShaping
, ScaledExponentialCosineFunctionHodographicShaping
). The class is defined as follows (the scaled factor is by default set to 1.0):
ExponentialTimesTrigonometricFunctionHodographicShapingSettings( const double exponent,
const double frequency,
const double scaleFactor )

class
PowerFunctionHodographicShapingSettings
¶
Class of settings for power functions (either simply power or scaled power base functions). The class is defined as follows (the scaled factor is by default set to 1.0):
PowerFunctionHodographicShapingSettings( const double exponent,
const double scaleFactor )

class
PowerTimesTrigonometricFunctionHodographicShapingSettings
¶
Class of settings for base functions defined by multiplying power functions with trigonometric ones (this includes the following hodographic shaping base functions: PowerSineFunctionHodographicShaping
, PowerCosineFunctionHodographicShaping
, ScaledPowerSineFunctionHodographicShaping
, ScaledPowerCosineFunctionHodographicShaping
). This settings class is defined as follows (the scaled factor is by default set to 1.0):
PowerTimesTrigonometricFunctionHodographicShapingSettings( const double exponent,
const double frequency,
const double scaleFactor )
4.1.4. Optimising the hodographically shaped trajectory¶
As explained before, hodographic shaping allows for a flexible number of base functions for each velocity component. Three base functions at least must be provided so that the boundary conditions can be satisfied, but a higher number of base functions can also be used, which turns into the creation of n9 degrees of freedom (n being the total number of base functions provided, over the three velocity components). The weighting coefficients for these additional base functions are free parameters and can be tuned to minimise the deltaV required by the shaped trajectory. This thus transforms the hodographically shaping problem into an optimisation problem where the best set of free parameters, leading to the minimum deltaV, is to be found.
A predefined optimisation problem compatible with the PAGMO library has been implemented to this end.

class
HodographicShapingOptimisationProblem
¶
This class sets up the optimisation of the hodographically shaped trajectory, and its constructor is defined as follows:
HodographicShapingOptimisationProblem(
Eigen::Vector6d initialState,
Eigen::Vector6d finalState,
const double timeOfFlight,
const int numberOfRevolutions,
simulation_setup::NamedBodyMap& bodyMap,
const std::string bodyToPropagate,
const std::string centralBody,
std::vector< std::shared_ptr< shape_based_methods::BaseFunctionHodographicShaping > >& radialVelocityFunctionComponents,
std::vector< std::shared_ptr< shape_based_methods::BaseFunctionHodographicShaping > >& normalVelocityFunctionComponents,
std::vector< std::shared_ptr< shape_based_methods::BaseFunctionHodographicShaping > >& axialVelocityFunctionComponents,
std::vector< std::vector< double > >& freeCoefficientsBounds )
where the input parameters are:
initialState
State vector at departure
finalState
State vector at arrival.
timeOfFlight
Timeofflight of the shaped trajectory.
numberOfRevolutions
Expected number of revolutions of the shaped trajectory.
bodyMap
Map of pointers to
Body
objects defining the trajectory environment.
bodyToPropagate
Name of the spacecraft to be propagated.
centralBody
Name of the central body of the trajectory.
radialVelocityFunctionComponents
Vector of
BaseFunctionHodographicShaping
objects, containing the definition of the base functions used to shape the radial velocity.
normalVelocityFunctionComponents
Vector of
BaseFunctionHodographicShaping
objects, containing the definition of the base functions used to shape the normal velocity.
axialVelocityFunctionComponents
Vector of
BaseFunctionHodographicShaping
objects, containing the definition of the base functions used to shape the axial velocity.
freeCoefficientsBounds
Vector containing the lower and upper bounds for the free coefficients of the hodographic shaping method.
The fitness
function creates the hodographically shaped trajectory corresponding to the base functions provided as inputs, and to a given set of free parameters. It then returns the deltaV associated with this shaped trajectory.
The get_bounds
function simply returns the bounds for the free coefficients to be optimised, which are already provided as inputs of the HodographicShapingOptimisationProblem
constructor.