|
DUECA/DUSIME
|
As simulations are developed with DUECA and DUSIME, additional tools for simulations will be developed, and once these tools are made re-usable, they will be included in the DUECA distribution. Note that these tools have nothing (or maybe little) to do with DUECA itself, they can be seen as separate from the communication and activation tools offered by DUECA. This documentation section is a sort of gathering of such tools.
To facilitate OpenGL drawing, a class dueca::DuecaGLWindow has been developed. The common use for modules that want to have GL output, for example a module that provides an outside visual or instrument display, would be to create a class (normally not the module class itself, but if you want to experiment with multiple inheritance, you might consider it) that derives from DuecaGLWindow. This derived class should implement at least an implementation for the virtual function dueca::DuecaGLWindow::display. This implementation should use opengl commands to draw the image. Further details are in the DuecaGLWindow documentation.
You can either run the GL window in priority level 0. If you want to run the GL window in another priority level, you must use the GlutSweeper. This is a dueca module that takes care of glut initialisation, and ensures that an update is called after you requested a redraw for your window(s).
The function dueca::randNormal() provides a random generator for random numbers with a normal (Gaussian) distribution and a standard deviation of 1.
The class dueca::GtkGladeWindow can create an experiment window directly from a file created with glade, a GUI creation tool. By using a table with callback functions, of type dueca::GladeCallbackTable, you can easily connect the actions from the widgets to a C++ class handling the interface.
Linear table interpolation for tables up to four dimensions is available from the dueca::Interpolator1, dueca::Interpolator2, dueca::Interpolator3 and dueca::Interpolator4 classes.
A RigidBody class provides the basis for implementing the dynamics of a single rigid body. One can make a derived class, in which the forces and gravitational pull on the body are defined. Optionally this class can extend the state vector of the RigidBody parent. With integration routines dueca::integrate_euler() or dueca::integrate_rungekutta() , the combination may be integrated numerically. These integration function templates can also be used separately.
dueca::LinearSystem is a class for calculating linear system responses. It accepts the linear system as a continuous-time transfer function or state-space system. The variant dueca::LimitedLinearSystem implements a linear system with limiter.
For converting and calibrating signals that go to and from hardware (AD and DA converters, and the like), and possibly for other things as well, a set of classes has been defined:
dueca::SingletonPointer, can be used to ensure only one instance of a hardware controlling class is created.
dueca::OutputCalibrator, for calculating integer output signals from a floating point value put in.
dueca::InputCalibrator, for doing the reverse.
dueca::Polynomial and dueca::PolynomialN, for implementing a calibration polynomial.
dueca::Steps and dueca::StepsN, for implementing a stepping output, e.g. for reading flap selections in complete increments.
dueca::Circular and dueca::CircularWithPoly, for implementing a conversion for circular (e.g. synchro) values.
dueca::Inverse, for implementing an inverting (division) conversion, e.g. from interval time to rate.
dueca::InputRatioCalibrator, for calculating an value as a ratio with, e.g., a reference voltage.
The class dueca::FindFiles creates a vector of file names matching a glob patter.
The class dueca::UniqueFile creates a numbered, unique file name based on a specified format string.
The templated class dueca::ConglomerateFactory implements an extensible factory pattern. This is used in for example the WorldView project to create families of graphical objects.
The file AxisTransforms.hxx contains objects for common axis transformations.