A Comparison of Task Driven and Information Driven Sensor Management for Target Tracking

Abstract

Abstract-Several authors have proposed sensor scheduling methods that are driven by information theoretic measures. In the information driven approach, the relative merit of different sensing actions is measured by the corresponding expected gain in information. Information driven approaches stand in stark contrast to task driven methods, i.e., methods that select some physical performance criteria and explicitly manage the sensor based on that criteria. This paper investigates the difference between a particular information driven approach, one that maximizes an alpha-Rényi measure of information gain, and task driven methods with a combination of theory and simulation. First, we give a mathematical relation that shows that when the decision error depends only weakly on the target state a certain type of marginalized information gain is a close approximation to the Bayes risk associated with performing a specific task. Second, we perform an empirical comparison between information driven and task driven approaches that maximize information gain or minimize risk, respectively. In particular, we give a task driven method that uses the sensor in a manner that is expected to maximize the probability the target is correctly located after the next measurement. We find as expected that the task driven method outperforms the information driven method when the performance is measured by risk, i.e., probability of localization error. However, the performance difference between the two methods is very small, suggesting that the information gain is a good surrogate for risk for this application.

Network sensor management for tracking and localization

“…In this sense the information gain can be considered as a proxy for performance for any of these tasks. The fundamental role of information gain as a near universal proxy has been demonstrated both by simulation and by analysis in [6]. The key result is a bound that shows any bounded risk function is sandwiched between two weighted alpha divergences.…”

An Information-Based Approach to Sensor Management in Large Dynamic Networks

“…The material in the first half of this section is largely drawn from a series of previously published papers [14], [40], [37]. It provides the background and notational conventions necessary before introducing the main topic of this paper, multiplatform sensor resource allocation via maximizing information flow.…”

“…In other problems where the mode is to be estimated [36], we have used [x,ẋ, y,ẏ, m] and when the class is to be estimated [37] we have used [x,ẋ, y,ẏ, c].…”

Multi-sensor control for multi-object Bayes filters

“…Examples of such cost functions include the MAP estimate of cardinality variance [26], statistical mean of cardinality variance [27], posterior expected error of cardinality and states (PEECS) [28], [29], [30] and statistical mean of the OSPA error [31]. A general discussion and comparison between task-driven and information-driven objective functions for sensor management is presented in [32].…”