![]() Reinforcing this issue, a case study and consequent results are presented and discussed on the perspective of the simulation usage for the MAS validation process. Moreover, and based on all the study of both agent modeling and simulation, it is foreseen that this integration approach can be a step forward on the study of shop-floor dynamics where a replication for the inclusion of simulation on well-known Agent Oriented Software Engineering methodologies, avoiding the MAS validation process only in the deployment phase. The concept of agent-in-the-loop is introduced as a way of leveraging the use of complex and consistent agent models as a benefit for the simulation process. I've released a fix for this in version 0.5.2 such that both NetLogo 6.0 and NetLogo 6.1 should be supported now. More information on this issue can be found here. There are two versions available: NetLogo Desktop, a downloadable version that runs on Linux, MacOS, and Windows, and NetLogo Web. In version 6 a new syntax was introduced using 'anonymous' procedures. ![]() For that purpose, a set of simple and easy to understand API was created to implement a simple simulation model using NetLogo and a MAS implemented using JADE, framed in the scope of industrial domain. There was a change in the NetLogo Controlling API with NetLogo 6.1.0 to the HeadlessWorkspace.open (.) method, which caused a compatibility issue with NL4Py. NetLogo runs on a Java virtual machine and uses a modified version of the Logo programming language, so a lot of people will be familiar with its terminologies, such as turtles and lists. In NetLogo before version 6, the symbols 1 1 and so on were used as placeholders in some primitives, especially those that processed list items. Based on the fact that NetLogo Extensions and JADE use Java programming language, Java Sockets were used to establish a simple and reliable communication means between the platforms, using FIPA-ACL specification as messaging content. Therefore, an integration of the Net Logo simulation platform and JADE agent-modeling framework is presented within an industrial agent-based approach. Nowadays, there's a notorious gap between simulation and agent-based modeling, not being either effective to use a simulation platform to model a complex Multi-Agent System, and agent-based modeling tools to simulate a certain system dynamics.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |