Cooperative Localization Using Fuzzy DDEIF and Broad Learning System for Uncertain Heterogeneous Omnidirectional Multi-robots

This paper presents a novel fuzzy distributed and decentralized extended information filtering (FDDEIF) method using broad learning system (BLS), called BLS-FDDEIF, for indoor cooperative localization of a group of heterogeneous omnidirectional mobile robots (HOMRs) incorporated with their dynamic effects. A new pose initialization algorithm is proposed to estimate the robots' initial poses. Once all the initial poses of the HOMRs have been roughly determined, a novel BLS-FDDEIF method is presented to fuse multisensory measurements for estimating more accurate poses of all the HOMRs. Comparative simulations and experimental results are conducted to show the effectiveness and superiority of the proposed method in finding accurate pose estimation of three cooperative HOMRs with unknown initial poses.