SUM Demonstration Results

This document is the “Demonstration Results” of the Surveillance in an Urban environment using Mobile sensors Project, corresponding to the Contract A-0827-RT-GC[AD. 2] of the “Defence R&T Joint Investment Programme on Force Protection” (A-0120-RT-GC) under the Call “Data Analysis, including Data Fusion from various Sources” (A-0444-RT-GC). Situational awareness (SA) refers to the perception of environmental elements with respect to time and space, the comprehension of their meaning, and the projection of their status. It is concerned with perception of the environment critical to decision-makers in complex, dynamic areas. The SUM system’s primary objective is augmenting the operator’s situational awareness with respect to potential threats, in other words to produce decision support information via the man-machine-interface (MMI). Identifying these potential dangers requires a minimum understanding of the environment. The latter remains a key challenge for the signal processing community. Via auxiliary information and other global intelligence, a preliminary SA can be built. This can be verified to the actual situation and/or extended with local intelligence. Moreover, it can aid the extraction of local intelligence. In this project local intelligence embodies possible threats to the SUM vehicle. For this purpose, the SUM vehicle has a dedicated sensor array which aims at detecting several types of threats under different circumstances. A fusion engine augments the detection performance. It consists of several fusion processes. Initial fusion mechanisms take the raw sensor data, combine these with their respective meta-data and fuse local and global information respectively. Recall that local information stems from the vehicle’s sensor array and comprises a queue of candidate threats with a set of features that describe well their local characteristics. The global information comprises the risk-maps created via the fusion module for UAV and satellite information. It may also include global information from other external sources. The final fusion step aims at differentiating between true threats and false alerts. In the context of this project, the MMI allows an operator to monitor the unprocessed outputs of each sensor and a map of the area on which detected threats are displayed. This document compiles the different processing results of the tests conducted with the SUM demonstrator, as defined in [RD. 1], during the Test Campaign described in [RD. 2]. The objective includes the assessment of the adopted the technological lines and approaches. It also explores the performance of the individual systems in a quasi-real circumstance, and their interdependence. Note that an essential part of the Test Campaign comprises the recording of all relevant data so that offline processing becomes possible