Multi-sensor information fusion based on local least squares support vector machines（LLSSVM）technology is used to realize height information fusion technology research.Aiming at height data of miniature unmanned helicopter（MUH） which could not accurately measured.LLSSVM is used to establish a data-driven black-box model

and this multi-sensor information fusion model is adopted to predict height information.Compared simulation results of models based on least squares support vector machines（LSSVM）and LLSSVM.The results show that the latter is better than the former

and show that the information fusion model based on LLSSVM is more reasonable and effective.The problems of prediction delay and error accumulation enlarges of LSSVM model are solved.