Goal-Oriented Agent System
Jump to navigation
Jump to search
A Goal-Oriented Agent System is a Multi-Agent System that is based on interaction goals.
- Example(s):
- Counter-Example(s):
- See: Agent Interaction, Goal, Agent System.
References
2019
- (Gnatyshak et al., 2019) ⇒ Dmitry Gnatyshak, Luis Oliva-Felipe, Sergio Alvarez-Napagao, Julian Padget, Javier Vazquez-Salceda, Dario Garcia-Gasulla, and Ulises Cortes (2019)."Towards a Goal-oriented Agent-based Simulation framework for High-Performance Computing". In: Preprint arXiv:1911.10055.
- QUOTE: Agent-based simulation (ABS) faces the conflicting demands of providing larger simulations (in terms of the number of agents) and providing agents with better cognitive and decision-making capabilities (which tend not to scale well in large simulations). This paper is a first attempt at addressing this friction, through the use of High-Performance Computing (HPC), with the aim of enabling scenarios in which large populations of individuals (e.g. traffic simulation, industrial and urban areas along geographical areas) have the ability to perform normative reasoning, planning or even BDI-like behaviour, in order to explore the analysis of phenomena resulting from more detailed behavioural modelling.
2005a
- (Cheong & Winikoff, 2005) ⇒ Christopher Cheong, and Michael Winikoff (2005, July). "Hermes: Implementing Goal-Oriented Agent Interactions". In: International Workshop on Programming Multi-Agent Systems (ProMAS 2005).
- QUOTE: In this paper, we focus on the implementation of goal-oriented agent interactions that have been designed using the Hermes methodology. We present a set of guidelines which can be used to implement a Hermean design, and apply them to implement a sample design in Jadex (...)
2005b
- (Khallouf & Winikoff, 2005) ⇒ Jason, Khallouf, and Michael Winikoff (2005, September). "Towards goal-oriented design of agent systems". In: Fifth International Conference on Quality Software (QSIC'05) (pp. 389-394). IEEE.
- QUOTE: Goals also play a key role in the implementation of proactive software agents. However, although some agent-oriented software engineering methodologies have incorporated (aspects of) goal-oriented requirements engineering, and although they target agent platforms that provide goals as an implementation construct, none of the methodologies provide a goal-oriented design process. We present modifications to the Prometheus methodology which make it more goal-oriented in its design phases and report on an experimental evaluation comparing the effectiveness of the original and refined methodologies.