Building agent teams using an explicit teamwork model and learning
Artificial Intelligence - Special issue on Robocop: the first step
Random Evolution in Massive Graphs
FOCS '01 Proceedings of the 42nd IEEE symposium on Foundations of Computer Science
| Hi-index | 0.01 |
A directed agent implies an agent with high constraints in both recognition and motion. Because of the embodied restrictions, the directed agent perceives a sense of subjective distance according to its position and direction, which is not asymmetry unlike the physical distance. The directed agent can take advantage of this asymmetry of this sense of distance, synthesizing Interaction Network, which is introduced to represent a cooperative and competitive form based on the directed graph network. Each node corresponding to the agent has a variable number of links to the neighboring nodes depending on the internal state and local interactions characterized by activation and inhibition. Simulation results illustrate an efficient team play and analysis.