-Intelligent lighting system is the core component of the smart home. In order to achieve user-centric intelligent control, improve indoor visual comfort level and saving energy at the same time, we have designed a multi-agent intelligent lighting system based on multi-source information fusion. Multi-agent systems able to improve the interactivity of the system, so that the subsystem concurrent processing of data and information for decision making distributions to achieve a closed-loop control. An intelligent data fusion model is proposed to fuse ambient light, occupancy, energy limitations, user preferences and other information for decision making. We use ANFIS modeling algorithm for environmental control of occupancy estimated. Finally, through FIS algorithm, we can implements the entire lighting control subsystem.