Modeling and Tracking of Maneuvering Extended Object With Random Hypersurface

Abstract

Extended object tracking, an important and emerging research field, has attracted attention for applications in civilian and military fields. However, the modeling and tracking of maneuvering extended objects still face challenges that need to be solved explicitly. Specifically, it is usually difficult to describe the irregular or complex extension dynamics (e.g., the object extension changes in shape and orientation over time) of a maneuvering extended object with random hypersurface. Further, it is tightly coupled to the centroid kinematics during the object maneuvers. Hence, this paper proposes a general approach for modeling a maneuvering extended object with random hypersurface. In this approach, both the object centroid and the extension dynamics for describing the turn maneuvers with different turn rates can be characterized accurately, in a concise linear form. This facilitates the design of an effective tracking algorithm to deal with different turn maneuvers of a maneuvering extended object with random hypersurface. The merits of this proposed model are illustrated by simulation results, where the kinematic state and object extension are estimated jointly and effectively.