Jason Leezer worked for a summer on exploring negative diffusion in perturbed planetary rings. The following summer he worked on a distributed, load-balanced, fault tolerant simulation framework in Java.
Amy West is working on ringlet formation in high M-number resonances.
Glenn Kavanagh is working on GPGPU applied to numerical simulations for an Honors Thesis. He also worked for a summer as part of an NSF grant on parallelizing different codes including adding multithreading to SwiftVis.
Brent Peckham worked on optimizing gravitational N-body simulations for the Cell processor. This was the topic of his Hornors Thesis.
Keller O'Hara developed an API for doing Multi-Agent simulations. This was the topic of his Hornors Thesis.
Andrew Krausnick explored agent behavior in a simple economy with human avatars. This was the topic of his Hornors Thesis.
Tom Dietzel looked at economic modeling in Multi-Agent systems. This was the topic of his Thesis.
Domingo Lara worked on comparing different spatial data structures to determine the optical data structures for doing collision processing in N-body simulations.
Eric Garza worked on algorithmic generation and layouts of cities. This work began with the original offering of Effective Object-Oriented Programming and continued into his Honors Thesis.
Mike McBryde worked on algorithmic generation of commercial buildings for his Honors Thesis.
Jess Martin worked on algorithmic generation of residential buildings. This work started as part of the original offering of Effective Object-Oriented Programming and continued on through his Honor Thesis.
Paul West worked on an honors thesis that explores the use of templates and expression templates to produce efficient bignum implementations in C++. He preented some of his early work on this at NCUR 2004.
Victoria Winbow worked with me for a semester on trying to produce a simple streamline model for rings. Full simulations, in research listed below, has shown that collisions prevent streamlines from shearing through one another. In theory a simple model based only on this can exhibit many of the behaviors seen in large scale simulations much more cheaply. Unfortunately, we haven't foudn the right blend of simplicity and complexity yet.
Nick Wing did his honors thesis with me on the topic of parallelizing collision detection algorithms. This work became two papers that were published in the annual proceedings of the PDPTA conference 2002 and 2003.
At CU I began working with Glen Stewart on a project to simulate gravitational
wakes in rings. The primary physical example of this phenomenon is at the edges
of the Encke Gap in Saturn's A-ring. This 325km wide gap is caused by the presence
of a small moonlet named Pan. When particles in the ring pass by Pan they are
perturbed into more eccentric orbits. Because the particles are basically on
Keplerian orbits, the differential rotation produces shear which produces the
wakes. The easiest way to find papers and abstracts by me on this is to go to
the ADS search site
and search for Lewis, M.C. as an author with the keyword ring.
More recently this work has included gravitational effects and I have also looked at narrow rings systems like Saturn's F ring.
This project looked at the evolution of planetary atmospheres. It includes applets that illustrate the way in which different atmospheric loss processes work.
In this project I explored how the nature of a surrogate Hamiltonian that is used in a symplectic mapping can impact the onset of chaos in a system. In particular, the way that KAM tori form. This includes some applets that let you play with different integrators.
This project looked at anomalous absorption of light in clouds. It includes an applet that uses a Monte-Carlo simulation to bounce photons through different cloud configurations with different scattering rules to measure how much the path length increases.
For this project I looked at how L-systems might be used for parsing natural languages instead of Chompsky grammars. While there didn't seem to be much of an advantage to L-systems for this, it does include some applets that let you play with L-systems.