Iterative Query-Based Approach to Efficient Task Decomposition and Resource Allocation

Intelligent coordination in complex multi-agent environments requires sophisticated mechanisms for suboptimal task decomposition and efficient resource allocation provided by the the agents. Besides the quality of coordination (i.e. efficiency of decomposition and resource allocation) we need to handle also computational efficiency restriction such as fast response time and limited communication traffic among the agents as well as optimization of the amount of private knowledge disclosure, during collaboration patterns negotiation among the semi-collaborative agents. We present a novel contracting mechanism based on the use of the approximated acquaintance model, a structure where the agents store the information about the states, capabilities and resources of possible collaborators. We suggest an approach of iterative construction of the partially-linear acquaintance models that is beneficial mainly in complex agent communities.