Vehicle localization by sensor fusion of LRS measurement and odometry information based on moving horizon estimation

In this study, we propose a localization method based on the fusion of the laser range sensor (LRS) measurements and the odometry information of a vehicle using moving horizon estimation (MHE). LRS measurement includes outliers and suffers from the intermittent observation; alleviation of their effect is required in order to localize a vehicle position with high accuracy. Proposed localization method merges multi-sampling data by exploiting MHE, which greatly reduces the effect of outliers and intermittent observation on localization using the data of other sampling. We show the efficacy of proposed localization by numerical simulations and experiments.