On Krause’s Multi-agent Consensus Model With Weight

In recent years, the multi-agent System(MAS, Multi- Agent System) is widely studied, Multi-agent system has the properties of autonomy, coordination and distribution, and we can use it to organize, learn something and ratiocinate.Multi-agent model often assume that the state of the system and the connection relations can accurate measurement and description, but in fact many internal or external system will have uncertainty, especially the problems in the field of social science. For example, mentioned in the introduction, the model is assumed observations can be measured accurately, especially, can be regarded as certainly real numbers. In fact, however, we cannot measure accurately to the size of one’s point of view, for Cucker- Smale model, we also have similar problems. Because the model is intended to make an agent according to the size of the relative position witch is decided by the accelerated speed, but the location of the relativity in the actual circumstances is difficult to give an accurate evaluation. Classical Cucker- Smale model only reflects an ideal condition. In order to be more close to our real life, it is necessary to use weight function to replace a specific numerical evaluation.In the paper, we estimate Krause model through weight function, discuss the existence of stochastic multi-agent model’s solution and the state of the synchronization. Of course, this paper proves that any two agents’ point of view will either eventually converge to the same point of view, or they will never influence each other after a period of time. The conclusion are consistent with the Krause model of using accurate numerical. It is a traditional extension and supplement of the classical model.