Cooperative searching for stochastic targets

Entropy

The problem is a two-dimensional area-restricted search for a target using a coordinated team of autonomous mobile sensing platforms (agents). Sensing is characterised by a range-dependent probability of detection, with a non-zero probability of false alarms. The context is underwater surveillance using a swarm of amphibious drones equipped with active sonars. The paper develops an intermittent information-driven search strategy, which alternates between two phases: the fast and non-receptive displacement phase (called the ballistic phase) with a slow displacement and sensing phase (called the diffusive phase). The proposed multi-agent search strategy is carried out in a decentralised manner, which means that all computations (estimation and motion control) are done locally. Coordination of agents is achieved by exchanging the data with the neighbours only, in a manner which does not require global knowledge of the communication network topology.

CSI Transactions on ICT, 2017

In this article, we present a unique search game model to search/hide an immobile object by/from a mobile sensor in a two-dimensional bounded space. In the proposed model, the mobile sensor is a searcher, the immobile object is a target and the search space is a square/rectangular region. The proposed model is suitable for the case where the searcher has no prior knowledge about the probability distribution of the target location in the region. The game model helps the players (searcher and target) to choose their best response strategies considering all possible strategies of their respective opponents and computes the expected payoffs and Nash Equilibrium of the game. The novelty of the proposed model is to guide both the players to choose their best response strategies. The proposed model is set up as follows: Initially, the search space which is a square/rectangular region is divided into square blocks of equal size to represent it as a grid and the distance is measured from the starting block of the searcher to all other blocks using shortest path followed by which the transition probabilities of each block is determined. Once the payoff matrix is obtained, we use lrslib tool to compute the mixed strategies, Nash equilibria and expected payoffs. The proposed model is applicable in real-time scenarios which involve large square/rectangular grids where the number of blocks is large.

Oper. Res., 2020

An object is hidden in one of several discrete locations according to some known probability distribution, and the goal is to discover the object in minimum expected time by successive searches of individual locations. If there is only one way to search each location, this search problem is solved using Gittins indices. Motivated by modern search technology, we extend earlier work to allow two modes—fast and slow—to search each location. The fast mode takes less time, but the slow mode is more likely to find the object. An optimal policy is difficult to obtain in general, because it requires an optimal sequence of search modes for each location, in addition to a set of sequence-dependent Gittins indices for choosing between locations. Our analysis begins by—for each mode—identifying a sufficient condition for a location to use only that search mode in an optimal policy. For locations meeting neither sufficient condition, an optimal choice of search mode is extremely complicated, dep...

IEEE Transactions on Aerospace and Electronic Systems, 2000

This paper presents a receding-horizon cooperative search algorithm that jointly optimizes routes and sensor orientations for a team of autonomous agents searching for a mobile target in a closed and bounded region. By sampling this region at locations with high target probability at each time step, we reduce the continuous search problem to a sequence of optimizations on a finite, dynamically updated graph whose vertices represent waypoints for the searchers and whose edges indicate potential connections between the waypoints. Paths are computed on this graph using a receding horizon approach, in which the horizon is a fixed number of graph vertices. To facilitate a fair comparison between paths of varying length on non-uniform graphs, the optimization criterion measures the probability of finding the target per unit travel time. Using this algorithm, we show that the team discovers the target in finite time with probability one. Simulations verify that this algorithm makes effective use of agents and outperforms previously proposed search algorithms. We have successfully hardware tested this algorithm in two small UAVs with gimbaled video cameras. DRAFT 3 for a stationary target. In [5], Eagle and Yee were able to solve somewhat larger problems than what had previously been possible by formulating the problem as a nonlinear integer program and using a branch and bound algorithm to solve it. However, the size of computationally feasible problems was still severely limited. DasGupta et al. presented an approximate solution for the continuous stationary target search based on an aggregation of the search space using a graph partition . In [21], we used a similar approach, but with a "fractal" formulation, which allowed the partitioning process to be implemented on multiple levels. All of the results mentioned so far are for a single search agent.

IEEE Transactions on Robotics, 2010

A robust satisficing approach based on info-gap theory is suggested as a solution for a spatial search planning problem with imprecise probabilistic data. A group of agents are searching predefined patches of land for stationary targets, given an a priori probability map of the targets' locations. This prior probabilistic information is assumed to be severely uncertain and may contain large errors. An analysis of a simplified case shows that in some situations one might prefer a different strategy than the expected-utility maximizing one, in terms of robustness to uncertainty. Deterministic numeric results confirm the theoretical predictions for more complex cases. Finally, stochastic numeric analysis of robust satisficing solutions on a large group of much more complex, randomly generated cases, reveals an interesting behavior of a consolidation of effort in specific cells, and implies the potential of robust satisficing in more realistic scenarios. As the robustness to uncertainty comes at the expense of the expected utility, one must choose its decisions carefully. However, it is shown that, in various circumstances, one obtains results which are superior to the expected-utility maximizing strategy in terms of robustness, while sacrificing almost no expected utility. He is an associate fellow of AIAA, and a senior member of IEEE.

International Journal of Systems, Control and Communications, 2018

Target searching is widely accepted as a significant area of study by various research communities. This paper addresses four target searching scenarios in a two-dimensional grid with obstacles, where multiple mobile sensors aim to search a single mobile target in a minimal time. The reachability condition of the target is checked before modelling the problem. The proposed work classifies the scenarios based on information-set available to the mobile sensors and the target. The scenarios are modelled as games that involve two adversary players: mobile sensor and target. The search strategies for the mobile sensors are formulated under different circumstances, and the strategic differences between cooperative and non-cooperative strategies are analysed. Later, the proposed work is extended in a new dimension, where information gain for the mobile sensors is determined by information refreshment interval. The proposed work helps the decision makers by facilitating the search operation in different scenarios.

Artificial Intelligence, 2013

Computer systems science and engineering, 2022

Concepts in search theory have developed since World War II. The study of search plans has found considerable interest among searchers due to its wide applications in our life. Searching for lost targets either located or moved is often a time-critical issue, especially when the target is very important. In many commercial and scientific missions at sea, it is of crucial importance to find lost targets underwater. We illustrate a technique known as coordinated search, that completely characterizes the search for a randomly located target on a plane. The idea is to avoid wasting time looking for a missing target. Two searchers or robots start from the center of a circle to search out a lost target, the first searcher looks for the target on the right side of the circular area, and the second one looks for it on the left side. The time taken to detect the target is obtained by assuming the target's position has a symmetric distribution. The procedures to facilitate the detection of the target are presented as an algorithm and as a flowchart. An application demonstrates the applicability of this search technique and the associated decrease in search cost. Its effectiveness is illustrated by numerical results, which indicates considerable promise.

An info-gap approach to spatial search planning under severe uncertainty is presented. In the problem of interest a single agent (e.g., an uninhabited aerial vehicle (UAV)) is searching for stationary ground targets, given an a priori probability map of the targets' locations. Due to its nature, this type of information is prone to severe uncertainties. An analytical comparison of a robust satisficing strategy to an expected utility optimizing one in a simplified case was conducted. It is shown that in some situations one might prefer a different strategy than the expected utility maximizing one, in terms of robustness to uncertainty. Simulation results confirm the theoretical predictions for more complex cases.

Proceedings of the 2010 American Control Conference, 2010

This paper outlines a strategy for tracking evasive objects in discrete space using game theory to allocate sensor resources. One or more searchers have to allocate the effort among the discrete cells to maximize the object detection probability within a finite time horizon or minimize the expected search time to achieve the desired detection probability under a false alarm constraint. We review the standard formulations under a sequential decision setting for finding stationary objects. Then we consider both robust and optimal search strategies and extend the standard search problem to a two-person zerosum search allocation game where the object wants to hide from the searcher and the object has incomplete information about the searcher's remaining search time. We discuss how the results affect the sensor management and mission planning for cooperative unmanned aerial vehicle (UAV) search tasks and provide simulation examples to show the effectiveness of the proposed method compared with random search strategy.