Land use change research has achieved remarkable results through rapid development in recent three decades. Various models have been constructed based on different research objectives of land use change. These models have been developed from quantitative simulation to spatial pattern simulation, from single model to multiple models. Cellular Automata model and the CLUE-S (Conversion of Land Use and its Effect at Small region extent) model are combined with other models or methods to improve the performance. Markov Chain, System Dynamics (SD), Logistic regression, and Random Forest are all able to calculate the land use requirements in the CA (Cellular Automata) model and CLUE-S model. Multi-criteria evaluation, geographically weighted regression, multi-agent model and Artificial Neural Network are also used to optimize the CA model, and the improvement of CLUE-S includes the upgrade of the CLUE series. These combined models are widely used to the case study of land use change in various regions, and to develop some software systems and datasets. The driving force analysis is mainly carried out from two aspects:natural factors and human factors, and the latter is the main factor that causes land use changes. In the current research, there are still problems of spatiotemporal scale, data error, uncertainties in data integration. In the future, land use change simulation research should further utilize the technological advantages of the big data and promote it to become more refined and diversified. In addition, future research should combine relevant knowledge in the field of ecological environment and dig deeper into the mutual feedback mechanism between land use change and its ecological environment effects. The research perspective can shift from exploring the impact of human activities on land use changes to the interaction between land use change and human activities for promoting the coordinated development of human-land relationships.