Underwater Passive Target Tracking from a Stationary Observer Using Modified Gain Extended Kalman Filter
Ashok Kumar.N, K Raja Rajeswari, Dr.S.Koteswara Rao · 2013
Target tracking in underwater ,for a stationary observer, observability is less compared to moving observer. Modified Gain Extended Kalman Filter (MGEKF) developed by Song and Speyer (2) was proven to be suitable algorithm for angles only passive target tracking applications in air. In this paper, this improved MGEKF algorithm is explored for underwater applications with some modifications. In underwater, the noise in the measurements is very high, turning rate of the platforms is low and speed of the platforms is also low when compared with the missiles in air. These characteristics of the platform are studied in detail and the algorithm is modified suitably for tracking applications in underwater. Monte- Carlo simulated results for one typical scenario is presented for the purpose of explanation. From the results it is observed that this algorithm is suitable for stationary observer in underwater passive target tracking using angles only measurements. 1.INTRODUCTION: In the ocean environment, an observer monitors noisy sonar bearings and elevations from a radiating target. The measurements are extracted from a single stationary observer and the observer processes these measurements to find out target motion parameters-Viz., range, course, bearing, elevation and speed of the target. Here the measurements are nonlinear; making the whole process nonlinear. However, the modified gain extended kalman filter (MGEKF) developed by Song and Speyer (2), was the successful contribution for angles only passive target tracking applications in air. This MGEKF algorithm was further improved by P.J. Galkowiski and M.A. Eslam (4). In this paper, this improved MGEKF algorithm is explored for underwater applications with some modifications. In underwater, the noise in the measurements is very high, turning rate of the platforms is low and speed of the platforms is also low when compared with the missiles in air. These characteristics of the platform are studied in detail and the algorithm is modified suitably for tracking applications in underwater. Section 2 deals with mathematical modelling of bearing and elevation measurements. Section 3 describes the implementation of the filter and section 4 is about the the results obtained in simulation. 2.MATHEMATICAL MODELING: Let a target be at a point P and the observer be at the origin, as shown in Fig-1. The measurement vector ,Z, is written as