Emergent command and control architecture for dynamic agent communities

Dynamic multi-agent systems require some accountability mechanism to ensure goal distribution and completion. Accountability can be maintained through proper delineation of power. An organisation is well delineated when each member has an unambiguous power relationship with its peers. Since responsibility and authority follow fully de-conflicted patterns, a delineated organisation is well suited for solving problems. This paper describes a non-deterministic agent level algorithm that emergently simulates various organisational structures and maintains a well-delineated power structure. The appropriateness of this algorithm is demonstrated in a highly dynamic, poor communication network system. A metric is provided which quantifies the appropriateness of the emerging organisational structures. A structure is considered appropriate if an unambiguous chain of accountability exists leading to the human overseer of the system.     

Trust-based protocols for regulating online,friend-of-a-friend communities

An online community is a group whose members are connected by means of information technologies, typically the Internet rather than face to face. Online communities allow social and cultural barriers to be spanned for communication. Social communities rely on interpersonal trust to regulate cohesion in their communities. After a general discussion, we offer protocols that extend viability of friend-of-a-friend framework along with experimental results.     

Modelling social norms in multiagent systems

Social norms are cultural phenomena that naturally emerge in human societies and help to prescribe and proscribe normative patterns of behaviour. In recent times, the discipline of multi-agent systems has been used to model social norms in an artificial society of agents. In this paper we review norms in multi-agent systems and then explore a series of norms in a simulated urban traffic setting. Using game-theoretic concepts we define and offer an account of norm stability. Particularly in small groups, a relatively small number of individuals with cooperative attitude are needed for the norm of cooperation to evolve and be stable. In contrast, in larger populations, a larger proportion of cooperating individuals are required to achieve stability.     

Natural human role in supervising complex control systems

Large collections of communities such as those found in complex control systems necessitate sophisticated techniques for high-level human supervision due to their requirements of influence spanning over individuals, communities, and global system behaviours. We have developed psycho-socio-cultural models for mediation of system-level behaviours and interactions. The ensuing lowered human cognitive load will enable supervisors to effectively guide large systems with competing objectives. This model paves the way for developing the ‘Man On The Loop’ (MOTL) paradigm, a phrase we are proposing for a novel human supervision role that contrasts with typical micromanagement. We have highlighted and validated MOTL parameters through the implementation of a domain-neutral simulation and compared our results with those found in natural systems.     

Stages of autonomy determination

We discuss stages of autonomy determination for software agents that manage and manipulate knowledge in organizations that house other software agents and human knowledge workers. We suggest recognition of potential autonomies in the belief-desire-intention (BDI) paradigm and actual reasoning about autonomy choice decision theoretically. We show how agents might revise their autonomies in light of one another's autonomy and might also experience new, derived autonomies. We discuss the conditions under which an entire group of agents might have a collective autonomy attitude toward agents outside their group. We believe group attitudes are a novel concept and form a strong basis for developing theories of dynamic organizational structure. We briefly sketch an outline of a case study that motivates reasoning about autonomies     

Improving behavior of computer game bots using fictitious play

In modern computer games, "bots" - intelligent realistic agents play a prominent role in the popularity of a game in the market. Typically, bots are modeled using finite-state machine and then programmed via simple conditional statements which are hard-coded in bots logic. Since these bots have become quite predictable to an experienced games player, a player might lose interest in the game. We propose the use of a game theoretic based learning rule called fictitious play for improving behavior of these computer game bots which will make them less predictable and hence, more a enjoyable game.