Highprecision positioning results are a prerequisite for path planning and other tasks of mobile robot. The generally used global positioning system(GPS) can obtain the global localization in an open areas, but there is a problem of low positioning accuracy in the absence of satellite signals.A combined localization method based on GPS and map matching was proposed to solve the positioning problem in complex environments without satellite signals in some region.A new motion model was established that could be used to correct the path estimation error and reduce the cumulative error.The fusion algorithm of GPS and track estimation based on the unscented Kalman filter algorithm improved the positioning accuracy of the mobile robot by 79.7%.Finally, map matching positioning was introduced, and the result of fusion of GPS and track estimation was combined to achieve accurate positioning in complex environments.Experimental results show that the combined positioning method based on GPS and map matching can solve the positioning problem of mobile robots in complex environments and the positioning accuracy of mobile robots is improved by 43% compared to traditional GPS positioning methods.