Jason Embedded: A Spin-off Version of Jason for Programming Embedded Multiagent Systems
Multi-Agent Systems (MAS) are composed of multiple autonomous and proactive agents situated in a physical or simulated environment. Agents and MAS can be encapsulated in physical devices to control sensors and actuators and provide autonomy and cognition to these devices. However, the development process of an Embedded MAS needs to reunite concepts from several areas of knowledge, such as Electronics, firmware, and reasoning programming. Besides, the existing agent-oriented programming languages do not offer mechanisms for controlling hardware, communicating with other devices, and agent mobility. Some solutions provide ad-hoc extensions for hardware interfacing, communicability, and mobility. But there is a lack of solutions including these three characteristics, even more using BDI. This work presents an extended version of the Jason framework named Jason Embedded. It comprises two types of agents: one for dealing with sensors and actuators and another for communicating with other prototypes and moving agents from one system to another. A case study demonstrates the usage of the extended framework in a scenario where two embedded MAS running in different prototypes need to cooperate and coordinate. It shows that Jason Embedded supports the process of creating autonomous devices.