The proliferation of agent-based and IoT technologies has enabled the development of novel applications involving hundreds of agents. Considering that autonomous devices that can control several aspects of our daily life are going to be available en mass in just a few years it will not be long until massive systems of autonomous agents, each acting upon its own knowledge and beliefs to achieve its own (or group) goals, become available and widely deployed.

The projects initially aims to study and define techniques for creating plans in multi-agent domains where aspects like dishonesty, deception, and incomplete knowledge are taken into account and the goal is related to both the agents’ knowledge and beliefs.

Moreover, as second step of the project, we will exploit the newly developed system to explore the concept of etichal-AI. The ability of reasoning in an epistemic environment will allow to define, through logical formalism, the idea of ethic and consequently to introduce a planning system that will search for solutions taking into consideration morality when needed.

The creation of such a system requires the definition of extra features for a multi-agent language (i.e. mA*) and the development of a declarative-based planner that also exploits high parallelization and logic programming.