Spacial Elliptical Model for Extended Target Tracking Using Laser Measurements

Abstract

This paper presents an approach for tracking extended targets using measurements from Light Detection and Ranging (LiDAR), where the measurement's source is distributed on the contour of the target extent. An Extended Kalman Filter (EKF) is used to estimate the target kinematic and shape simultaneously. The target shape is represented by an ellipse and the measurement's source is distributed uniformly on the contour of the ellipse. A Gaussian mixture is used to approximate the measurement likelihood. The EKF update incorporates a moment matching technique to approximate the mixture density. Data from an Ibeo LiDAR simulator is used to evaluate the performance of the proposed approach on tracking a target with a rectangular shape.