Federated Information Mode-matched Filter in an Imm Algorithm

In this paper, a tracking algorithm for autonomous navigation of automated guided vehicles (AGVs) operating in container terminals is presented. The developed tracking algorithm is an interacting multiple model algorithm used to detect other AGVs using fused information from multiple sensors. In order to detect other AGVs, two kinematic models were derived: A constant velocity model for linear motion, and a constant-speed turn model for curvilinear motion. In the constant-speed turn model, a nonlinear information filter is used in place of the extended Kalman filter in nonlinear systems. The model-matched filter used in multi-sensor environments takes the form of a federated nonlinear information filter. In this paper, the structural features and information sharing principle of the federated information filter are discussed. The performance of the proposed filter using a Monte Carlo simulation under various patterns is evaluated. Copyright © 2005 IFAC