Dynamic space reconfiguration for Bayesian search and tracking with moving targets

Abstract

This paper presents a technique for dynamically reconfiguring search spaces in order to enable Bayesian autonomous search and tracking missions with moving targets. In particular, marine search and rescue scenarios are considered, highlighting the need for space reconfiguration in situations where moving targets are involved. The proposed technique improves the search space configuration by maintaining the validity of the recursive Bayesian estimation. The advantage of the technique is that autonomous search and tracking can be performed indefinitely, without loss of information. Numerical results first show the effectiveness of the technique with a single search vehicle and a single moving target. The efficacy of the approach for coordinated autonomous search and tracking is shown through simulation, incorporating multiple search vehicles and multiple targets. The examples also highlight the added benefit to human mission planners resulting from the technique's simplification of the search space allocation task.

Searching for multiple targets using Probabilistic Quadtrees

“…From a robotic point of view the line of work most similar to our approach was performed by Furukawa and colleagues in a series of papers appeared in the last years [1], [2], [10], [19]. Therein the authors cast the search problem as a Baysian framework accounting for faulty sensors.…”

“…Moreover, their implementation focuses on UAVs flying at a constant elevation, and therefore sensor accuracy does not vary during the mission. In [10] the authors present one of the few examples of search architectures based on a spatial representation that is reconfigured during the search. However their model does not feature any hierarchical layering, but rather consists of planar shapes whose boundary varies as the search evolves.…”

Cost-Aware Bayesian Sequential Decision-Making for Search and Classification

“…In [9] the authors investigate search-and-tracking using recursive Bayesian filtering with foreknown targets' positions. The results are extended in [10] for dynamic search spaces, where a target might not be within a static search space at the next time step. In [11] the author proposes a Bayesian-based multisensor-multitarget sensor management scheme.…”

Cost-aware Bayesian sequential decision-making for domain search and object classification

“…In [4], the authors investigate the search-and-tracking problem using recursive Bayesian filtering with foreknown targets' positions with noise. The results are extended in [5] for dynamic search spaces based on forward reachable set analysis. In [6], the author proposes a Bayesianbased multisensor-multitarget sensor management scheme.…”