Aerospace Engineering Research Works

Permanent URI for this collectionhttp://hdl.handle.net/1903/1655

Browse

Author: search.filters.author.Healy, Liam

Search Results

Now showing 1 - 9 of 9
  • Thumbnail Image
    Item
    Comparison Of Accuracy Assessment Techniques For Numerical Integration
    (Univelt, Inc., 2003-02) Berry, Matt; Healy, Liam
    Knowledge of accuracy of numerical integration is important for composing an overall numerical error budget; in orbit determination and propagation for space surveillance, there is frequently a computation time-accuracy tradeoff that must be balanced. There are several techniques to assess the accuracy of a numerical integrator. In this paper we compare some of those techniques: comparison with two-body results, with step-size halving, with a higher-order integrator, using a reverse test, and with a nearby exactly integrable solution (Zadunaisky's technique). Selection of different kinds of orbits for testing is important, and an RMS error ratio may be constructed to condense results into a compact form. Our results show that step- size halving and higher-order testing give consistent results, that the reverse test does not, and that Zadunaisky's technique performs well with a single-step integrator, but that more work is needed to implement it with a multi-step integrator.
  • Thumbnail Image
    Item
    Symbolic and Parallel Computation in Celestial Mechanics
    (Society for Industrial and Applied Mathematics, 1996) Healy, Liam
    One aspect of celestial mechanics is the computation of the long-term orbits of celestial bodies. This type of computation is complicated by the interaction of the many bodies that need to be considered to derive accurate long-term behavior. For reasons explained in this chapter, it is necessary to do this symbolically rather than numerically. Symbolic computations performed on a Lisp machine are described. The visualization of the solution is accomplished in a massively parallel SIMD machine.
  • Thumbnail Image
    Item
    Close Conjunction Detection on Parallel Computer
    (American Institute of Aeronautics and Astronautics, 1995-07) Healy, Liam
    Close conjunction detection is the task of finding which satellites will come within a given distance of other satellites. The algorithms described here are implemented on the Connection Machine (CM) in a program called CM-COMBO. It will find close conjunctions of satellites over a time range for one, a few, or all satellites against the original or another catalog and works with an arbitrary propagator. The problem of comparing an entire catalog against itself is beyond the computing power of current serial machines. This program does not prefilter any orbits and does not make assumptions about the type of orbit (that it be nearly circular, for instance). This paper describes the algorithm for this computation, the implementation on the CM, and resuls of several studies using this program.
  • Thumbnail Image
    Item
    Deterministic Studies of Debris Hazards with Parallel Processors
    (European Space Agency, 1993-04-05) Healy, Liam; Coffey, Shannon
    A new generation of parallel processing computers makes possible the ability to propagate all objects in the space surveillance catalog with simulated objects, and detect close approaches. With this capability, it is possible to test deterministically debris scenarios, without resorting to statistical models. To compare the positions of objects we have developed two methods, an all-to-all comparison and a one-to-all comparison. For the former, a seive significantly reduces computation time; for the latter, direct comparison is possible in parallel. We show results from several simulations, including simulated multiple sources of debris, hazard to the space station, and close contacts amongst the catalog itself, to show potential for debris studies. The techniques described here have potential application the general problem of catalog maintenance.
  • Thumbnail Image
    Item
    Parallel Computing for Space Surveillance
    (MIT Lincoln Laboratory, 1992) Healy, Liam; Coffey, Shannon
    This paper reports on an application of massively parallel processors to multiple satellite propagation and the calculation of miss distances between objects (COMBO). Unlike serial computations, we do not pre-filter the data but rather sort the data set in a way that dramatically cuts the number of comparisons required in order to be assured of a complete catalog-to-catalog comparison. The same general algorithm allows two logical sets to be compare to each other. Run time for this demonstration code on an 8K Connection Machine is about one second per time step, including propagation, complete catalog-to-catalog calculation of miss distances, plotting satellite positions, and recording of the miss distances to a file. Propagation of the objects is performed with an analytic propagator, using J2 only at present, though the code may easily be extended to other propagators. We demonstrate a second application of parallel computing to the problem of debris propagation resulting from a satellite breakup. The spread of such debris into n pieces is simulated by replicating the element set for the original satellite n times, then altering each to represent a distribution of velocities to the center of mass.
  • Thumbnail Image
    Item
    A Toolbox for Nonlinear Dynamics
    (Springer-Verlag, 1991) Coffey, Shannon; Deprit, André; Deprit, Eitenne; Healy, Liam; Miller, Bruce
    Using the main problem of artificial satellite theory as an illustration, we review several developments which have had a significant impact on research in nonlinear dynamics. On the mathematical front, we point to the theory of Lie transformations; in the area of computational software, we explain how massively data parallel machines open the way for symbolic solution of large problems. Finally, we show how color graphics assist in the qualitative analysis of dynamical systems.
  • Thumbnail Image
    Item
    Computation of error effects in nonlinear Hamiltonian systems using Lie algebraic methods
    (American Institute of Physics, 1992-06) Healy, Liam; Dragt, Alex; Gjaja, Ivan
    There exist Lie algebraic methods for obtaining transfer maps around any given trajectory of a Hamiltonian system. This paper describes an iterative procedure for finding transfer maps around the same trajectory when the Hamiltonian is perturbed by small linear terms. Such terms often result when an actual system deviates from an ideal one due to errors. Two examples from accelerator physics are worked out. Comparisons with numerical computations, and in simple cases exact analytical calculations, demonstrate the validity of the procedure.
  • Thumbnail Image
    Item
    Paint by number: Uncovering phase flows of an integrable dynamical system
    (American Institute of Physics, 1991-09) Healy, Liam; Deprit, Etienne
    Given an integrable dynamical system with one degree of freedom, "painting" the integral over phase space proves to be a powerful technique for uncovering both global and local behavior. This graphical technique avoids numerical integration, employing instead a nonlinear method of assigning contrasting colors to the energy values to distinguish subtle details of the flow.
  • Thumbnail Image
    Item
    The Phase Space Portrait of an Integrable Dynamical System
    (Science, 1990-02-16) Coffey, Shannon; Deprit, André; Deprit, Etienne; Healy, Liam
    For an integrable dynamical system with one degree of freedom, "painting" the integral over the phase space proves to be very effective for uncovering the global flow down to minute details. Applied to the main problem in artificial satellite theory, for instance, the technique reveals an intricate configuration of equilibria and bifurcations when the polar component of the angular momentum approaches zero.