|
Abstract:
|
Locating and identifying a radiation source with information from stationary detectors is a well-documented inverse problem, but the optimal placement of such detectors is a more challenging task. In particular, optimal sensor placement depends on the location of the source, which is generally not known a priori. Mobile radiation sensors, which can adapt to the given problem, are an attractive alternative. While most mobile sensor strategies designate a trajectory for sensor movement, we instead employ mutual information to choose the next measurement location from a discrete set of design conditions. We use mutual information, based on Shannon entropy, to determine the measurement location that will give the most information about the source location and intensity. To illustrate our sensor movement strategy, we present an example of source localization in downtown Washington D.C. based on synthetic data generated from a simplified radiation transport model.
|
