Terrain navigation

Some of the typical sensors carried by an AUV provide bathymetric measurements; either as a main product or by-product. Among these sensors are DVLs, multi beam echosounders, altimeters, and interferometric side scan sonars. The bathymetric measurements from these sensors can be correlated with a pre-obtained digital terrain model (DTM) of the seafloor, and as a result the AUV position within this DTM may be estimated. This technique is called terrain navigation. The algorithms performing the correlation can conceptually be divided into global search algorithms and tightly integrated algorithms.

Examples of global search algorithms are the Terrain Contour Matching (TERCOM) (Golden, 1980), the Point Mass Filter (PMF), and particle filters (Bergman et. al, 1999). There are different degrees of sophistication to each of these algorithms, but in essence they take input from the INS, a bathymetric sensor and a DTM, and estimate a global position measurement to be integrated back with the INS. These algorithms handle highly non-linear bathymetry with great results, but have convergence problems in terrain with little variation.

The tightly integrated algorithms, such as Terrain Referenced Integrated Navigation (TRIN) (Hagen & Hagen, 2000), and Sandia Integrated Terrain Aided Navigation (SITAN) (Hostetler, 1978), integrate the bathymetric range measurements tightly with the INS. These algorithms handle linear and weakly non-linear terrain very well, but may have some problems with highly non-linear terrain.

Terrain navigation is a fully autonomous technique, but it requires a DTM of the mission area. This information is unfortunately not always available. A solution to this problem is called Concurrent Mapping and Navigation (CMN), where the DTM is made by the AUV in-situ, and used concurrently by the terrain navigation algorithms to bind the position error drift.

Point Mass PDF Contours alter 9 pings Particle Filter cloud alter 9 pings

Point Mass PDF Contours alter 9 pings Particle Filter cloud alter 9 pings

East oflset[m] East oflset[m]

Fig. 8. The contour lines of the PMF's probability density function (left) and the particle filter cloud (right) after processing 9 pings from a DVL with the tool TerrLab (Hagen, 2006). Red line is a priori navigation solution, black line is true position and cyan marks position estimates from the terrain navigation algorithms.

East oflset[m] East oflset[m]

Fig. 8. The contour lines of the PMF's probability density function (left) and the particle filter cloud (right) after processing 9 pings from a DVL with the tool TerrLab (Hagen, 2006). Red line is a priori navigation solution, black line is true position and cyan marks position estimates from the terrain navigation algorithms.

Learn Photoshop Now

Learn Photoshop Now

This first volume will guide you through the basics of Photoshop. Well start at the beginning and slowly be working our way through to the more advanced stuff but dont worry its all aimed at the total newbie.

Get My Free Ebook


Post a comment