Tracking multiple targets using binary proximity sensors

Abstract

Recent work has shown that, despite the minimal information provided by a binary proximity sensor, a network of such sensors can provide remarkably good target tracking performance. In this paper, we examine the performance of such a sensor network for tracking multiple targets. We begin with geometric arguments that address the problem of counting the number of distinct targets, given a snapshot of the sensor readings. We provide necessary and sufficient criteria for an accurate target count in a one-dimensional setting, and provide a greedy algorithm that determines the minimum number of targets that is consistent with the sensor readings. While these combinatorial arguments bring out the difficulty of target counting based on sensor readings at a given time, they leave open the possibility of accurate counting and tracking by exploiting the evolution of the sensor readings across time. To this end, we develop a particle filtering algorithm based on a cost function that penalizes changes in velocity. An extensive set of simulations, as well as experiments with passive infrared sensors, are reported. We conclude that, despite the combinatorial complexity of target counting, probabilistic approaches based on fairly generic models for the trajectories yield respectable tracking performance.

Minimalist multiple target tracking using directional sensor beams

“…This effectively gives a lower bound on the number of agents. This result is analogous to the one found by [21] in the case of tracking in the real line. Thus, we present the following proposition:…”

“…Algorithms using binary proximity sensors have been proposed for tracking a single agent [5], [13], [20]. For the multiple agent tracking case, extensions have been proposed which use classification [17] or constrain the tracking to the real line [21]. The binary proximity sensors are usually modeled as circular disk regions that must cover the tracking region of interest.…”

On Non-Cooperative Multiple-Target Tracking With Wireless Sensor Networks

“…The initial work [25]- [27] on binary proximity sensors focuses on tracking single target. Singh et al [29] extended the approach to track multiple indistinguishable targets by applying particle filtering algorithms. Approaches based on binary proximity sensors have two obvious advantages: (a) The sensors are very simple since they only report binary information.…”

SHOT: Scenario-Type Hypothesis Object Tracking with Indoor Sensor Networks

“…They also proposed an object tracking method that achieves these limits for binary sensor networks. In a previous study [4], the authors applied the particle filter algorithm for tracking multiple objects with binary sensor networks. In this method, the hypothesis in the near future is estimated from current sensor information.…”

“…We then consider the situation in which a real object departs from the monitoring region through gate g h , which is placed on a boundary between the outside of the monitoring region and micro-cell m z ∈ M at time t 4 . When the tracking server receives the departure event information from sensor node s h on gate g h , the tracking server calculates the minimum value of the estimated velocity square error ε , is selected as the corresponding real object k p , and the trajectory of virtual object b p is chosen as the trajectory of real object k p .…”