IMM-IPDA Based Maneuvering Target Tracking with Mobile Robot in Unknown Clutter Environments

This paper presents a novel approach for maneuvering target tracking with mobile robot in unknown clutter environments. This issue involves both simultaneous localization and mapping (SLAM) in dynamic environments and maneuvering objects tracking (MOT), and it is referred to as the SLAMMOT problem. Our approach derived by Integrating the full covariance extend Kalman filter based SLAM(EKFSLAM) algorithm and interacting multiple model (IMM) estimation algorithm with integrated probabilistic data association (IPDA) filters. The IMM portion of our algorithm is consisted of several filters based on different object motion models to handle maneuvering objects tracking with mobile robot in unknown environment, and each filter is an IPDA to deal with the problem of clutter in object observations. Both real robot and simulation experiments are presented to show the effectiveness of this approach.