LiDAR-based localization has been used in many applications,but it is difficult to achieve global localization with LiDAR only for its low dimension feedback,especially in environments with repetitive geometric features.A senso r fusion-based localization system is introduced,which has the capability of a d dressing global localization.Both LiDAR and vision sensors are integrated,maki ng use of the rich information from vision sensor and the robustness from LiDAR. A hybrid grid-map is built,and visual global descriptor is applied to speedin g up the localization convergence,combined with the pose refining pipeline for i mproving the localization accuracy.Also,a trigger mechanism is introduced to s olve kidnapped problem.The experiments are designed to evaluate the proposed ap proach,as well as a comparison with the existing localization systems.Accordin g to the experimental results,our system is able to solve the global localizati on problem,and has an improved performance.