A deep reinforcement learning based searching method for source localization

Zhao, Yong; Chen, Bin; Wang, Xianghan; Zhu, Zhengqiu; Wang, Yiduo; Cheng, Guangquan; Wang, Rui; Wang, Rongxiao; He, Miao; Liu, Yu

doi:10.1016/j.ins.2021.12.041

Cited by 27 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A deep CNN was presented in [32] to predict the focused image from a regular migration image that contains a quasi-symmetric pattern in both space and time. While the authors in [28], [33], and [34] established different DRL localization models by leveraging the deep Q-learning network (DQN) learning framework. However, the source localization efficiency is not considered in these works, making the works could not be directly used in time-sensitive commitments.…”

Section: The Deep Learning-based Source Localization Methodsmentioning

confidence: 99%

Securing Fast and High-Precision Localization for Shallow Underground Explosive Source: A Curiosity-Driven Deep Reinforcement Learning Approach

Wu,

Wang,

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Shallow underground explosive source localization technology is a key technology in the field of underground space localization. The existing approaches mainly aim to improve the localization accuracy, but need to deploy enormous sensors in the monitoring area, and rely on a large number of back-end workstations to solve. These methods have the defects of considerable calculation and high time cost, and are hard to satisfy the precise and real-time requirements of onsite testing, ultimately resulting in slow localization speed and accurate localization failure. Fortunately, emerging deep reinforcement learning (DRL) can effectively solve the problem of slow search policy by modeling the source localication as a Markov decision process (MDP). Therefore, a Curiosity-Driven Deep Dueling Double Q-learning Network (C-D3QN) is subsequently proposed to solve the above MDP. The overestimation problem is solved by decoupling selection and evaluation of the bootstrap action, and the action difference is effectively increased by introducing the dueling network that separately represents state values and action advantages. Meanwhile, the exploration is jointly reinforced by an intrinsic reward outputted from the curiosity module and an extrinsic reward supplied by the environment, guaranteeing the convergence to global optimal. Finally, extensive simulation results based on the outfield experiment data show that compared with other algorithms, the proposed scheme can significantly improve exploration ability and learning speed as well as generalization and robustness. Additionally, compared to the baseline algorithm DQN, the C-D3QN algorithm can offer an improved localization accuracy as high as 99.62% and an increased localization speed of 66.23%.

show abstract

Section: The Deep Learning-based Source Localization Methodsmentioning

confidence: 99%

Securing Fast and High-Precision Localization for Shallow Underground Explosive Source: A Curiosity-Driven Deep Reinforcement Learning Approach

Wu,

Wang,

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In general, source search is a kind of problem that aims to determine the location of the source (gas or signal) in the shortest possible time, as it is of vital importance for both nature and mankind [24][25][26] , for example, the search for preys [27] , submarines [28] , survivors [29] , and pollution sources [30] . As a classical kind of source search algorithm, the bio-inspired algorithm typically leverages the gradient ascent strategy to approach the source based on a reasonable assumption that the signal emitted by the source has a greater intensity near the source [31,32] .…”

Section: Source Searchmentioning

confidence: 99%

Leveraging Human-AI Collaboration in Crowd-Powered Source Search: A Preliminary Study

Zhao,

Zhu,

Chen

et al. 2023

J. Soc. Comput.

Self Cite

View full text Add to dashboard Cite

Source search is an important problem in our society, relating to finding fire sources, gas sources, or signal sources. Particularly, in an unexplored and potentially dangerous environment, an autonomous source search algorithm that employs robotic searchers is usually applied to address the problem. Such environments could be completely unknown and highly complex. Therefore, novel search algorithms have been designed, combining heuristic methods and intelligent optimization, to tackle search problems in large and complex search spaces. However, these intelligent search algorithms were not designed to address completeness and optimality, and therefore commonly suffer from the problems such as local optimums or endless loops. Recent studies have used crowd-powered systems to address the complex problems that cannot be solved by machines on their own. While leveraging human intelligence in an AI system has been shown to be effective in making the system more reliable, whether using the power of the crowd can improve autonomous source search algorithms remains unanswered. To this end, we propose a crowd-powered source search approach enabling human-AI collaboration, which uses human intelligence as external supports to improve existing search algorithms and meanwhile reduces human efforts using AI predictions. Furthermore, we designed a crowd-powered prototype system and carried out an experiment with both experts and nonexperts, to complete 200 source search scenarios (704 crowdsourcing tasks). Quantitative and qualitative analysis showed that the sourcing search algorithm enhanced by crowd could achieve both high effectiveness and efficiency. Our work provides valuable insights in human-AI collaborative system design.

show abstract

“…They employed a k-center method to identify multiple sources and the corresponding infection areas in general networks. [14] Other methods have also been proposed for source localization, including maximum likelihood estimation, [11,[15][16][17] statistical physics, [18][19][20][21] reverse propagation, [22][23][24][25][26][27][28][29][30][31][32][33][34] machine learning, [35,36] etc. [37][38][39] However, existing source localization research overlooks the signed nature of node connections, which is frequently encountered in signed networks, such as social networks that incorporate friend and enemy relationships.…”

Section: Introductionmentioning

confidence: 99%

Source localization in signed networks with effective distance

Ma 马,

Sun 孙,

Ding 丁

et al. 2024

Chinese Phys. B

View full text Add to dashboard Cite

While progress has been made in information source localization, it has overlooked the prevalent friend and adversarial relationships in social networks. This paper addresses this gap by focusing on source localization in signed network models. Leveraging the topological characteristics of signed networks and transforming the propagation probability into effective distance, we propose an optimization method for observer selection. Additionally, by using the reverse propagation algorithm we present a method for information source localization in signed networks. Extensive experimental results demonstrate that a higher proportion of positive edges within signed networks contributes to more favorable source localization, and the higher the ratio of propagation rates between positive and negative edges, the more accurate the source localization becomes. Interestingly, this aligns with our observation that, in reality, the number of friends tends to be greater than the number of adversaries, and the likelihood of information propagation among friends is often higher than among adversaries. In addition, the source located at the periphery of the network is not easy to identify. Furthermore, our proposed observer selection method based on effective distance achieves higher operational effciency and exhibits higher accuracy in information source localization, compared with three strategies for observer selection based on the classical Full-order neighbor coverage.

show abstract

A deep reinforcement learning based searching method for source localization

Cited by 27 publications

References 27 publications

Securing Fast and High-Precision Localization for Shallow Underground Explosive Source: A Curiosity-Driven Deep Reinforcement Learning Approach

Securing Fast and High-Precision Localization for Shallow Underground Explosive Source: A Curiosity-Driven Deep Reinforcement Learning Approach

Leveraging Human-AI Collaboration in Crowd-Powered Source Search: A Preliminary Study

Source localization in signed networks with effective distance

Contact Info

Product

Resources

About