Survey: mobile sensor networks for target searching and tracking

Abstract

Mobile sensor networks can be employed in multiple applications, such as search and rescue, border patrol, battle scenarios, and environmental monitoring. In this survey, we review the literature utilizing mobile sensor networks in applications classified as target searching and/or tracking. Our contribution is threefold. First, we focus on the diverse types of filters applied to estimating the state of the targets. Second, we present the most common approaches to high-level trajectory planning for the sensors in the network to do target searching and/or tracking. Finally, we classify the literature based on the problem formulation used and solution characteristics. At the end of the survey, we discuss the current state of the literature and possible directions for future research efforts.

An Overview on Optimal Flocking

“…Flocking, as defined by Reynolds, will seemingly be driven by advances in decentralized control, robust control, and longduration autonomy in the future. However, some application areas, such as mobile sensor networks, have criticized Reynolds flocking as a novelty that does not necessarily have advantages in terms of performance or ease of implementation; see Albert and Imsland (2018).…”

“…There have been several surveys and tutorials on decentralized control that include flocking; see Barve and Nene (2013); Bayindir (2016); Ferrari et al (2016); Zhu et al (2016Zhu et al ( , 2017; Albert and Imsland (2018). In one motivating example, Fine and Shell (2013) discuss various flocking controllers without considering optimality.…”

Distributed moving horizon state estimation for sensor networks with low computation capabilities

“…In recent decades, the interest on distributed state estimation increased tremendously due to its variety of applications over sensor networks such as target tracking [1]- [2], exploration [3], monitoring [4], etc. The estimation algorithms used in these applications are mainly formulated as centralized sensor fusion architectures, where all the sensors transmit their local measurements to a central unit which processes the provided data to update the estimation [5].…”

Distributed moving horizon estimation with pre-estimating observer

“…In recent decades, there has been an increasing interest in research on distributed state estimation due to its variety of applications over sensor networks such as target tracking [1]- [2], exploration [3], surveillance [4], etc. The estimation algorithms used in these applications are mainly formulated as centralized fusion architectures, where all the sensors transmit their local measurements to a central unit that processes the provided measurements to update the estimate.…”