Interacting multiple model filter based moving object tracking with mobile robot in unknown environments

Abstract

This paper presents a simultaneous localization,mapping (SLAM) and object tracking(OT) method based on interacting multiple model (IMM) filter to achieve simultaneous estimation of robot and object's trajectories in unknown environments. The proposed approach integrates IMM based object tracking and EKF-based SLAM. Since the implementation of IMM, the method can help the robot to track object and estimate its motion model, which improves the accuracy of object localization. Moreover, this approach can help robot to track the object that have more ability to escape ,and improve it practicality .Simulation and real robot experiment results validate the effectiveness of the proposed method in robot and target localization and mapping. The results also show the accuracy of target motion model estimation and the system ability to track maneuvering target.