Multi-agent probabilistic search in a sequential decision-theoretic framework

Abstract

Abstract-Consider the task of searching a region for the presence or absence of a target using a team of multiple searchers. This paper formulates this search problem as a sequential probabilistic decision, which enables analysis and design of efficient and robust search control strategies. Imperfect detections of the target's possible locations are made by each search agent and shared with teammates. This information is used to update the evolving decision variable which represents the belief that the target is present in the region. The sequential decision-theoretic formulation presented in this paper provides an analytic framework to evaluate team search systems, as it includes a performance metric (time until decision), a measure of uncertainty (decision confidence thresholds) and imperfect information gathering (detection error). Strategies for cooperative search are evaluated in this context, and comparisons between homogeneous and hybrid search strategies are investigated in numerical studies.

On Probabilistic Search Decisions under Searcher Motion Constraints

“…In this manner, the temporal evolution of the decision can be examined, with the decision performance measured by the time till the search decision is made. The main contributions of this paper include further theoretical and algorithmic developments of the search-as-a-decision approach posed in [7,8], including lower bounds on the expected decision time found by examining the special case of perfect detections. However, whereas this previous work considered the search strategy employed by a mobile searcher to determine where it should search, this paper examines the influence of different degrees of mobility, i.e., how the searcher reaches its goal, on the search decision performance.…”

An Efficient Framework for Probabilistic Search with Multi-UAVs

“…Under different scenes, the model can be largely different, what we focus is the scene described in section 1. So the search in the last two kinds is in our attention [11], [12].…”

“…About the movement, a large number researches have a pre-defined assumption that the UAV can not re-planned its movement no matter how the UAVs are distributed [13], [14], [15]. And most of them is associated with the first kind of search such as lane base search, random search, grid based search [16].…”

Coordinated Search with a Swarm of UAVs

“…To account for this dependence, we update all the cells in the belief map after each observation in a similar way as described in [3] and [5]. In addition, we also account for the fact that for each measurement, the k th UAV can observe a set of M h k cells simultaneously, where h k is the altitude of the UAV.…”