By simon | February 11, 2010
I mentioned so-called Dutch (or reverse) auctions last week before the seminar and have just come across an example of how Dutch-auctions occur in the wild. From Charlie’s Diary:
Pricing … we sell books by reverse auction, most expensive editions first, then cheaper editions, then mass market, until we get to the remainder shelves.
[Amazon, Macmillan: an outsider's guide to the fight]
I hadn’t thought about book selling in this way before but that is exactly what it is. Or rather it has characteristics of a Dutch auction but also plays off the price against availability. If you want the good sooner then you pay more, even though there are many books available at this higher price, and if you wait then the price drops, although you get an inferior version in my opinion if you go for the paperback versus the hardback, but I digress. I wonder how this version affect the attributes of the auction in comparison to a true Dutch auction.
Anybody know of any other versions of the Dutch auction happening in the real world (not necessarily agent related)?
By simon | January 21, 2010
On their website, AOS has a number of descriptions of agent projects that either they have developed or that have been developed using their agent framework:
- Intelligence, Surveillance and Reconnaissance Broker – A broker designed to assist a human (e.g. pilot or weapons systems operator) to source critical ISR sensor data and deliver such data to the necessary onboard systems in order to successfully prosecute a mission.
- Rules of Engagement – This project adds sophisticated modelling of ROE (Rules of Engagement) to simulation environments. This application is described in more detail here: BRIMS 2007 Best Paper
- Oil Trading & Operations - Norwegian-based Statoil, one of the world’s largest suppliers of crude oil and natural gas, has developed software to support oil trading and operations management, using JACK.
- Behaviour Recognition Agents in a Land Logistics Environment (BRAILLE) – Assist humans to recognise “interesting” behaviours from a surveillance picture and provide visual feedback.
- Intelligent Prognostic Health Manager (PHM) – A project to enable systems to intelligently react to, and accommodate hardware failures.
- Human Behaviour Representation – An MOD project to provide individual agents with realistically based models of behaviour based upon human psychology.
By simon | January 21, 2010
I found a nice example of the use of an agent-oriented approach over at the AOS website. The example describes an agent-based aircraft control system. Firstly, the problems are depicted, identifying:
- Chaos from many aircraft converging on an airport
- The need to hold aircraft on approach to airports
Even in a medium-sized airport, there will be congestion at certain times of the day, with more aircraft arriving than can be accommodated by the runway(s). Resolving this competition for the runway resource requires that the flow manager take many factors into account, and can lead to overload:
- Too many unexpected dynamic changes in the situation
- sudden change in wind
- aircraft diverging from expected behaviour
- one of the runways unavailable due to weather
- Too many aircraft in a given time frame
- too many ETAs (Estimated Time of Arrival) and pilot instructions to calculate/issue
- Optimisation factors
- potential for simultaneous landing on crossing runways
- heavy aircraft can land close behind light aircraft (but not vice versa)
- Leads to
- communications bottlenecks with air traffic controllers
- aircraft unnecessarily forced to enter holding pattern
- sub-optimal use of the runway resource
It is suggested that too much complexity and interaction leads to a need for delegation, one of the trends that Wooldridge identified as leading towards agent-oriented approaches. Furthermore
AOS identify that traditional software techniques are inappropriate because they are unwieldy. Although there is a single algorithm that could satisfy the problem, it is inefficient and doesn’t deal nicely with the dynamism of the real world, hence it is a fragile approach:
The key point is that there is a single locus of control. The main loop grows with the complexity of the problem, and reactivity has to be manually incorporated into each step of the loop. Because of the inherent complexity of interactions, traditional software approaches lead to a monolithic code-base that is error-prone and difficult to alter without breaking the carefully coded handling of the dynamic aspects of the problem.
The agent-oriented approach is to delegate various functions to autonomous agents, essentially distributing functionality throughout the system and avoiding the centralised control loop. More on the proposed AOS solution, showing the types of agents required and relationships between them, is
here.
By simon | January 21, 2010
Dundee university is a member of AgentLink III which is the premier Co-ordination Action for Agent Based Computing, funded by the European Commission’s 6th Framework Program.
One of the AgentLink publications, the AgentLink Roadmap [ local pdf mirror ] is a very useful way to get an overview of agents and multiagent systems, the technologies, and their applications.
Launched on 1st January, 2004, it provides support for the network of European researchers and developers with a common interest in agent technology through events aimed at industry outreach, and standardisation issues, as well as providing support for academic events and providing resources through the AgentLink Portal.
M. Luck, P. McBurney, and O. Shehory and S. Willmott, Agent Technology: Computing as Interaction (A Roadmap for Agent Based Computing), AgentLink, 2005. ISBN 085432 845 9
Bibtex entry for the Roadmap :
@book{al3roadmap,
author = {Luck, M. and McBurney, P. and Shehory, O. and Willmott, S.},
title = {Agent Technology: Computing as Interaction
(A Roadmap for Agent Based Computing)},
publisher = {AgentLink},
year = {2005}
}
By simon | January 21, 2010
I have posted the reading for next thursday’s seminar onto the papers and reading resources page. You should read the shoham paper “Agent Oriented Programming” for next weeks seminar. It will be quite difficult for many of you as it is written in an academic style that is generally hard to read until you are used to it. However, learning to read and comprehend academic papers is a key skill to acquire if you plan a long and cutting edge career in computing. If you still have time then reading the Jennings paper “On Agent Based Software Engineering” will give you a decent background to the agent oriented approach to developing complex software systems. Finally, useful supplemental reading to this weeks lecture can be found in the first chapter of Wooldridge “An Introduction to MultiAgent Systems” first edition which should be available from the library.
By simon | January 20, 2010
Welcome to all of the students taking Research Frontiers (AC42001) in the School of Computing @ the University of Dundee. Information about the module is available from the pages indexed along the right hand side. In addition to core module content, such as lecture slides and labsheets, I shall be using this site to post interesting resources related to the module content which I hope that you will find interesting and useful and will also contribute to. By utilising new resources, for example, from news stories as they happen, we can hopefully ground our learning on this module in real world applications of A.I.