Proposed projects for Multi-Agent systems course
|
|
Description:
|
Based on the ideas of custumized commercials such as in the movie "Minority
Report" create a set of applications which are delivering customized commercials
to users based feedback from proximity sensors, GPS and identification systems.
|
| Team members |
Majid Khan ma680109@pegasus.cc.ucf.edu,
Linus Luotsinen linusluotsinen@hotmail.com
|
|
|
Description:
|
Based on the ideas of custumized commercials such as in the movie "Minority
Report" create a set of applications which are delivering customized commercials
to users based feedback from proximity sensors, GPS and identification systems.
|
| Team members |
Angela Salva amalban@cfl.rr.com, Shane Lewis srldiver@yahoo.com
|
|
|
Description:
|
Create an agent which can take as an input a set of goods to be obtained and
a deadline. By participating in one-to-one negotiations, or multi-player auctions
(english or dutch) the goal of the agent is to buy the goods as presented for
the smallest price possible.
|
|
Team members
|
Santhosh Kumar santhu79@hotmail.com,
Shankar Karthik Vaithianathan svaithia@cs.ucf.edu
|
|
|
Description:
|
Create a collection of swarming agents which are moving in a simulated
martian environment. The goal of the agents is to gather information about the
interesting features of the land and send it to the mother ship. The agents do
not have the ability to communicate over long distances, so they are
communicating using pheromon tracers (similar to what ants and bees are using in
the terran environment).
|
|
Team members
|
Rishan Chandarana ri512384@pegasus.cc.ucf.edu
Jingfei Kong (???)
|
|
|
Description:
|
Create a set of agents for coordinating a distributed scientific computing
application run on a set of heterogeneous computers. You will need to create at
least three types of agents:
- wrapper agents which wrap legacy applications,
- coordination agents which are coordinating the execution and manage the data
flow
- user interface agents for computation steering
|
Team members
|
Nick Paladino paladino@cs.ucf.edu
|
|
|
Description:
|
Develop an intelligent refrigerator which is able to shop various stores, and
maintain a healthy diet of the user, constrained by certain spending limits.
You need to develop the following agents:
- Refrigerator agent
- Merchant agents
- User agent
For extra credit:
- develop user interfaces
- keep track of the expiration dates of the products
- provide support for exceptional events e.g. vacation, parties etc.
|
|
Team members
|
Shawn Hillis shillis@mac.com,
|
|
|
Description:
|
Develop an intelligent refrigerator which is able to shop various stores, and
maintain a healthy diet of the user, constrained by certain spending limits.
You need to develop the following agents:
- Refrigerator agent
- Merchant agents
- User agent
For extra credit:
- develop user interfaces
- keep track of the expiration dates of the products
- provide support for exceptional events e.g. vacation, parties etc.
|
|
Team members
|
Guoqiang Wang gwang@cs.ucf.edu,
Han Yu hyu@cs.ucf.edu,
Xin Bai xbai@hector.cs.ucf.edu
|
|
|
Description:
|
The project will consist of a number of unmannel aerial vehicles (UAVs). The
UAVs will coordinate with one another in order to meet mission parameters.
Mission parameters will be set forth by an end user at a console, but user
participation can be minimal. Mission parameters may include target
acquisition and tracking, attack, and communication relay. The UAVs will be
simulated based on real-world aircraft flight characteristics. The UAVs will
communicate using a message-passing protocol over serial radio with limited
range. The user interface will consist of a map displaying a United States
Geographical Survey (USGS) map and the constantly updating relative positions
of the deployed UAVs. The map will also display approximate location of
targets and other areas of interest identified by the UAVs.
|
|
Team members
|
Paul DeJong dejongp@celebration.saic.com,
Scott VanderWeide vanderweides@celebration.saic.com
|
|
|
Description:
|
What I want to do is to create a very simple agent, pretty much
it be two sticks connected by a joint. It will have the capability of
bending. If many of these were connected together, they will have to
coordinate with one another achieve something. When they are connected
they will have created a multi-joint arm. One end will be stationary,
while the other end will have a task of moving its end somewhere. The end
with the task of moving somewhere cannot move there on its own, since it's
arm is not long enough, so it will need to send a message to the agent
that it is connected that it needs to move to X,Y position (I am sticking
to two coordinates for simplicity). Now the agent will try to move it
self in that direction, but may need assistance from its connected agent
on the other end, and so on. These type of cases will happen. You
mentioned to me to use a Virtual Reality enforcer. The VR enforcer will
actually move the agents in the 2 D space. When it gets a message from an
agent that it is going to bend, it will bend that agent and it will rotate
the connected agents.
|
|
Team members
|
Jason Schutz jschutz@cs.ucf.edu
|
|
|
Description:
|
I propose doing a project for the EEL 5937 Multiagent Systems class
using the standard game of chess. Since the rules and basic strategy of
chess have been understood (by most) for centuries, developing a project
based on chess would allow the student to focus more on learning about
agents and how to create them rather than spending time creating a new
game or world from scratch. Chess also has the advantage of having Open
Source GUIs available that could be used to display the progress of the
game and information about the program.
I think every chess piece can be an agent with characteristics of color,
type, current location, legal moves, alive or not, etc. Each piece
agent can evaluate its current situation and determine the best move to
make on an individual basis and report it to a master agent. The master
agent can select among the moves reported by the piece agents based on
the current strategy and future prospects. The master agent would then
instruct the piece agent to make its best move. The piece agent would
make the move and broadcast its new position to all the other piece and
master agents. If the new position represents the location of another
piece, that piece would recognize it had been captured and set its
current status to "dead", broadcasting this information to all the other
piece and master agents.
This paradigm may not produce a very capable chess game (particularly if
the master agent algorithms are weak) but because of the different types
of agents and agent interaction, I think it would make an effective
project for learning about multiagent systems. |
|
Team members
|
Charles H. Chapman ccmail@bellsouth.net
|
Student-proposed projects (idea factory)
Nick Paladino:
As a suggestion for another possible project, I was thinking of a
possibly malicious type of agent.? Something that could probe the Bond
system for vulnerabilities and weaknesses.? It had occurred to me that
none of the projects?listed were of the malicious type.? If it has not
been done, I believe that an analysis of this could be useful to the
overall Bond project.
Han Yu:
I also have another idea in mind which is based a project that I did
last year. The main topic of it is to formulate a group of sensors to
detect enemy radars. Every sensor, regarded as an agent, can work
individually or cooperate by communication methods.