We present a unifying perspective of the individual control layers of the agent architecture InteRRaP. InteRRaP aims at modeling autonomous resource-bounded agents that interact with each other in dynamic multiagent environments. InteRRaP implements a pragmatic Belief-Desire-Intention (BDI) architecture, where the agent's mental state is distributed over a set of layers. Based on the processes of situation recognition and planning and scheduling, a uniform description for each control layer — the behavior-based layer, the local planning layer, and the cooperative planning layer — is provided. We demonstrate various options for the design of interacting agents within this framework in an interacting robots application. The performance of different agent types in a multiagent environment is experimentally evaluated.
CITATION STYLE
Fischer, K., Müller, J. P., & Pischel, M. (1996). A pragmatic BDI architecture. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1037, pp. 203–218). Springer Verlag. https://doi.org/10.1007/3540608052_68
Mendeley helps you to discover research relevant for your work.