Modelling uneven terrain for geo-location of mines detected via vehicular mounted sensors

Abstract

When searching for landmines using vehicular-mounted sensors, it is important that the ground locations of the detected mines be accurately determined. This is useful for data association when one has multiple looks at a mine by a single sensor or if one uses multiple sensors. This knowledge is necessary for neutralizing the detected mines or at least marking them for avoidance. Factors that contribute to errors in geo-location include sensor error, inaccurate knowledge of the vehicle position and/or attitude, and incomplete knowledge about the terrain being searched. This paper addresses the problem of incomplete terrain knowledge and its effects on geo-location accuracy. It presents different techniques for modelling the terrain and reducing the geo-location errors. A variety of uneven surfaces have been simulated. Random noise has been added to the basic surfaces to simulate the terrain roughness. A terrain model is then derived based on measurements one could obtain from a coarse scan of the field of view using a ranging laser. Two types of sensors are considered for the landmine detection: Synthetic Aperture Forward-Looking Radar (SAFLR) and Camera Type Sensors such as Infrared (IR). Relationships between the terrain unevenness and resulting geo-location errors are presented. The geo-location errors are worse for the Camera Type Sensor, but they can be significant for SAFLR also. The results of this analysis demonstrate that there may be significant geo-location errors for situations where the terrain is not so smooth, e.g. off-road searches, and that the problem can be alleviated via better knowledge of the terrain. If the field of view is planar or almost planar, even coarse-range scanning can improve the geo-location accuracy. For more complex surfaces, finer scanning may be required (preliminary results on this topic were presented at the SPIE (International Society for Optical Engineering) Conference, Orlando, FL, April 2001 (G. Cook, S. Jakkidi, S. Kansal, C. Amazeen and K. Sherbondy, “Impact of uneven terrain on geo-locations errors for mines detected via vehicular mounted sensors”, in Proceedings of SPIE, 2001, pp. 594–605.)). The computed geo-location errors, and the conclusions drawn as to the effectiveness of the different models are presented in the paper.