A Novel Global Set-Membership Filtering Approach for Localization of Automatic Guided Vehicles
Hao Yang, Yilian Zhang, Wei Gu, Huaicheng Yan, Fuwen Yang
Abstract
This article investigates the localization problem of the automatic guided vehicle (AGV) system. In order to improve the reliability and flexibility of the localization process, a distributed sensor network structure is introduced to realize the localization of the AGV. Moreover, considering the influence of unknown-but-bounded noise and the accuracy requirements of the localization, a novel global set-membership filtering approach is proposed to obtain accurate localization results including a distributed set-membership filtering (DSMF) strategy and a circumscribed rectangle method. First, a DSMF strategy is designed to obtain local state estimation ellipsoids. Sufficient conditions for the existence of the state estimation ellipsoids are derived and a convex optimization process is developed to obtain the optimal local estimation ellipsoids. Then, a circumscribed rectangle method is proposed to fuse all local state estimation ellipsoids and obtain global set-membership filtering results. The proposed fusion method does not have a complicated optimization process, and can obtain more accurate estimation ellipsoids than local estimation results. Performance analysis verifies the effectiveness of the proposed global set-membership filtering approach.