Multiple Target Tracking for Air Surveillance System

Abstract

This thesis is concerned with three important components in target tracking, namely, data association, multiple-model filtering, and sensor resolution modeling for maneuvering and overlapping targets. The thesis proposes Classification Information based Multiple Hypotheses Tracking (CIMHT) scheme to address the data association problem, Interacting Multiple Model (IMM) scheme with velocity, acceleration, and turn model filters to perform tracking in Multiple Sensor Multiple Target (MSMT) scenarios with target maneuvers, and Combined IMM based Probabilistic Data Association Filter (IMMPDAF) scheme for tracking a large number of overlapping targets. newline