An autonomous radiation source detection policy based on deep reinforcement learning with generalized ability in unknown environments
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Autonomous radiation source detection has long been studied for radiation emergencies. Compared to conventional data-driven or path planning methods, deep reinforcement learning shows a strong capacity in source detection while still lacking the generalized ability to the geometry in unknown environments. In this work, the detection task is decomposed into two subtasks: exploration and localization. A hierarchical control policy (HC) is proposed to perform the subtasks at different stages. The low-level controller learns how to execute the individual subtasks by deep reinforcement learning, and the highlevel controller determines which subtasks should be executed at the current stage. In experimental tests under different geometrical conditions, HC achieves the best performance among the autonomous decision policies. The robustness and generalized ability of the hierarchy have been demonstrated
Abstract
Autonomous radiation source detection has long been studied for radiation emergencies. Compared to conventional data-driven or path planning methods, deep reinforcement learning shows a strong capacity in source detection while still lacking the generalized ability to the geometry in unknown environments. In this work, the detection task is decomposed into two subtasks: exploration and localization. A hierarchical control policy (HC) is proposed to perform the subtasks at different stages. The low-level controller learns how to execute the individual subtasks by deep reinforcement learning, and the highlevel controller determines which subtasks should be executed at the current stage. In experimental tests under different geometrical conditions, HC achieves the best performance among the autonomous decision policies. The robustness and generalized ability of the hierarchy have been demonstrated
참고문헌 (28)
- Dorit S. Hochbaum, "Nuclear threat detection with mobile distributed sensor networks", Annals of Operations Research, 187(1), pp.45~63 (2009), Springer Science and Business Media LLC[DOI]
- R. Vilim, "RadTrac: A System for Detecting, Localizing, and Tracking Radioactive Sources in Real Time", Nuclear Technology, 168(1), pp.61~73 (2009), Informa UK Limited[DOI]
- Paul Kump, "Detection of shielded radionuclides from weak and poorly resolved spectra using group positive RIVAL", Radiation Measurements, 48, pp.18~28 (2013), Elsevier BV[DOI]
- D. Connor, "Airborne radiation mapping: overview and application of current and future aerial systems", International Journal of Remote Sensing, 37(24), pp.5953~5987 (2016), Informa UK Limited[DOI]
- Ioannis Tsitsimpelis, "A review of ground-based robotic systems for the characterization of nuclear environments", Progress in Nuclear Energy, 111, pp.109~124 (2019), Elsevier BV[DOI]
- Nageswara S.V. Rao, "Network algorithms for detection of radiation sources", Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 784, pp.326~331 (2015), Elsevier BV[DOI]
- E. W. Bai, "The maximum likelihood estimate for radiation source localization: initializing an iterative search", Proc. IEEE Conf. Decis. Control. 2015-Febru, pp.277~282 (2014)
- G. Cordone, "Improved multi-resolution method for MLE-based localization of radiation sources", 20th Int. Conf. Inf. Fusion, Fusion 2017 - Proc (2017)
- M.R. Morelande, "Radiological Source Detection and Localisation Using Bayesian Techniques", IEEE Transactions on Signal Processing, 57(11), pp.4220~4231 (2009), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Robert Blake Anderson, "Mobile Robotic Radiation Surveying Using Recursive Bayesian Estimation", 2019 IEEE 15th International Conference on Automation Science and Engineering (CASE), pp.1187~1192 (2019), IEEE
- Prateek Tandon, "Detection of radioactive sources in urban scenes using Bayesian Aggregation of data from mobile spectrometers", Information Systems, 57, pp.195~206 (2016), Elsevier BV[DOI]
- Tomas Lazna, "Cooperation between an unmanned aerial vehicle and an unmanned ground vehicle in highly accurate localization of gamma radiation hotspots", International Journal of Advanced Robotic Systems, 15(1), pp.172988141775078~ (2018), SAGE Publications[DOI]
- Michael Hutchinson, "Adaptive Bayesian Sensor Motion Planning for Hazardous Source Term Reconstruction", IFAC-PapersOnLine, 50(1), pp.2812~2817 (2017), Elsevier BV[DOI]
- R.A. Cortez, "Experimental implementation of robotic sequential nuclear search", 2007 Mediterr, Conf. Control. Autom. MED (2007), IEEE
- Manish K. Sharma, "Three-dimensional localization of low activity gamma-ray sources in real-time scenarios", Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 813, pp.132~138 (2016), Elsevier BV[DOI]
- R. S. Sutton, "Reinforcement Learning: an Introduction" (2018), MIT press
- David Silver, "Mastering the game of Go with deep neural networks and tree search", Nature, 529(7587), pp.484~489 (2016), Springer Science and Business Media LLC[DOI]
- Zheng Liu, "Double Q-Learning for Radiation Source Detection", Sensors, 19(4), pp.960~ (2019), MDPI AG[DOI]
- Gregory R. Romanchek, "Stopping criteria for ending autonomous, single detector radiological source searches", PLOS ONE, 16(6), pp.e0253211~ (2021), Public Library of Science (PLoS)[DOI]
- Philippe Proctor, "Proximal Policy Optimization for Radiation Source Search", Journal of Nuclear Engineering, 2(4), pp.368~397 (2021), MDPI AG[DOI]
- J. Berkson, "Do radioactive decay events follow a random Poisson-Exponential?", Int. J. Appl. Radiat. Isot, 26, pp.543~549 (1975)
- G. E. Monahan, "State of the artda survey of partially observable Markov decision processes: theory, models, and algorithms", Manage. Sci, 28, pp.1~16 (1982)
- P. Cunningham, "Mach. Learn. Tech. Multimed", pp.21~49 (2008), Springer
- Y. Yu, "A review of recurrent neural networks : LSTM cells and network architectures", Neural Comput, 31, pp.1235~1270 (2019)
- "Proximal Policy Optimization Algorithms" (2017)[URL]
- V. Konda, "Adv. Neural Inf. Process. Syst" (1999), MIT Press
- S. Amari, "Backpropagation and stochastic gradient descent method", Neurocomputing, 5, pp.185~196 (1993)
- N. Heess, "Emergence of Locomotion Behaviours in Rich Environments", ArXiv, ~ (2017)