Michele Micheli

Scalable Adaptive Multitarget Tracking Using Multiple Sensors

In networked mobile multitarget tracking systems, parameters such as detection probabilities, clutter rates, and motion model parameters are often unknown and time-varying. Such parameter variability can seriously degrade the performance of a multitarget tracking system. Here, we propose a Bayesian tracking framework in which the multisensor-multitarget tracking problem is formulated according to the measurement origin uncertainty paradigm and the unknown parameters-in the present case, the detection probabilities at the individual sensors-are modeled as Markov chains. The resulting Bayesian estimation problem is then solved using the belief propagation scheme. This approach results in a multisensor-multitarget tracking method that is able to adapt to the time variations of the detection probabilities. Moreover, the method has a low complexity that scales very well in all relevant system parameters. The performance of the method is assessed using data collected by a mobile underwater wireless sensor network.

Improving Automatic Target Recognition with Forward Looking Sonar Mosaics

Abstract Automatic Target Recognition (ATR) is a key element needed to make Mine Countermeasure missions using robots entirely autonomous. While there has been much progress in applying ATR algorithms on high-resolution Synthetic Aperture Sonar (SAS) and sidescan sonar data, performing ATR with a low cost Forward Looking Sonar (FLS) is much more challenging. An algorithm for the detection of underwater man-made objects in FLS previously developed can work in real-time although it suffers considerably from typical noise in sonar images and false alarms. The work presented here shows that ATR algorithms can be exercised on sonar mosaics built also in real-time instead of raw data coming from the FLS. The use of mosaics can help the detection of the targets by reducing some noise (including harmonics from other acoustic devices mounted on the robot) and giving a better contrast to the images to be processed. Moreover, mosaic images can be useful for post-processing and data analysis. The mosaicking algorithm also runs in real-time to maintain the performance of the system and to be useful in real missions. It was tested both on data previously collected and in real experiments with different set-ups and with different sonars. The wide range of results obtained with different surface vehicles and in different situations demonstrate the usefulness of the method.

Adaptive underwater sonar surveys in the presence of strong currents

We consider the task of conducting underwater surveys with a sonar-equipped autonomous underwater vehicle (AUV) in environments with strong currents. More specifically, this topic is addressed in the context of mine countermeasure operations employing synthetic aperture sonar (SAS) sensors. Two complementary algorithms that allow the AUV to autonomously adapt its survey route based on sophisticated sensor data it collects in situ, while respecting the unique constraints imposed by the problem, are proposed. The algorithms allow the AUV to (i) adapt its survey heading based on the presence of currents to ensure quality data is collected, and (ii) adapt its survey route to reinspect the most suspicious objects at additional aspects. The flexibility to immediately react in situ to the environmental and tactical conditions sensed during the mission allow the most useful data for object recognition purposes to be collected efficiently. By obviating the recovery and redeployment of the AUV, as well as laboratory-based data-processing during the interregnum, the overall mission timeline can be greatly compressed and operational costs can be reduced. Experimental results illustrating the real-time execution of the proposed algorithms on an AUV are shown for a completely autonomous mission conducted in the North Sea.

Forward looking sonar mosaicing for Mine Countermeasures

Abstract Forward looking sonars (FLS) are nowadays popular for many different applications. In particular, they can be used for Automatic Target Recognition (ATR) in the context of Mine Countermeasures. Currently, ATR techniques are applied to raw data which generates many false positives and the need for human supervision. Mosaicing FLS data increases target contrast and thus reduces false positive rate. Moreover, it implies a considerable data size reduction which is important if one thinks of exchange of data in real time through an acoustic channel with very limited bandwidth. Results of applying a real-time mosaicing algorithm to FLS data generated during Mine Countermeasures missions are shown and discussed thoroughly in this article.

Interfaces between acoustic prediction, embedded signal processing, and behaviors at NATO Undersea Research Centre

The use of acoustic sensing systems for ASW in heterogeneous sensor networks utilizing marine robots has been a subject of research at the NATO Undersea Research Centre for the past several years. In this talk, we discuss the unique challenges of implementing ASW on autonomous, collaborative networks of AUVs, including the challenges of embedding the active sonar signal processing, implementing effective underwater messaging, and designing adaptive behaviors to optimize system performance. Theoretical studies, simulations, and results from the recent GLINT series of sea trials are shown and the way forward for autonomous sensor system studies at NURC is discussed.

Real-time sonar signal processing on-board an autonomous underwater vehicle

An active, bistatic signal processing system has been implemented for use on-board autonomous underwater vehicles (AUVs) using towed arrays. The NATO Undersea Research Centres (NURC) AUVs are programmed to maneuver in order to best track a target. To perform this action autonomously, the vehicle must be able to fully process and track targets via its towed array data in real time or faster. The processor implemented includes processing from the array hydrophone data, through beamforming, matched filtering, Doppler processing, and tracking. This system adapts many previously developed algorithms to function in real time. Research and algorithm adaptations in the areas of normalization, CFAR detection, and array navigation, have also been developed to deal with practical issues which have arisen in sea trials. The implementation, practical issues, and steps taken to address them will be presented. This research is supported by the NURC Consolidated Programme of Work, Cooperative ASW program.