{"title":"反馈粒子滤波及相关可控相互作用粒子系统(CIPS)综述","authors":"A. Taghvaei, P. Mehta","doi":"10.48550/arXiv.2301.00935","DOIUrl":null,"url":null,"abstract":"In this survey, we describe controlled interacting particle systems (CIPS) to approximate the solution of the optimal filtering and the optimal control problems. Part I of the survey is focussed on the feedback particle filter (FPF) algorithm, its derivation based on optimal transportation theory, and its relationship to the ensemble Kalman filter (EnKF) and the conventional sequential importance sampling-resampling (SIR) particle filters. The central numerical problem of FPF -- to approximate the solution of the Poisson equation -- is described together with the main solution approaches. An analytical and numerical comparison with the SIR particle filter is given to illustrate the advantages of the CIPS approach. Part II of the survey is focussed on adapting these algorithms for the problem of reinforcement learning. The survey includes several remarks that describe extensions as well as open problems in this subject.","PeriodicalId":50750,"journal":{"name":"Annual Reviews in Control","volume":"5 1","pages":"356-378"},"PeriodicalIF":7.3000,"publicationDate":"2023-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"A Survey of Feedback Particle Filter and related Controlled Interacting Particle Systems (CIPS)\",\"authors\":\"A. Taghvaei, P. Mehta\",\"doi\":\"10.48550/arXiv.2301.00935\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this survey, we describe controlled interacting particle systems (CIPS) to approximate the solution of the optimal filtering and the optimal control problems. Part I of the survey is focussed on the feedback particle filter (FPF) algorithm, its derivation based on optimal transportation theory, and its relationship to the ensemble Kalman filter (EnKF) and the conventional sequential importance sampling-resampling (SIR) particle filters. The central numerical problem of FPF -- to approximate the solution of the Poisson equation -- is described together with the main solution approaches. An analytical and numerical comparison with the SIR particle filter is given to illustrate the advantages of the CIPS approach. Part II of the survey is focussed on adapting these algorithms for the problem of reinforcement learning. The survey includes several remarks that describe extensions as well as open problems in this subject.\",\"PeriodicalId\":50750,\"journal\":{\"name\":\"Annual Reviews in Control\",\"volume\":\"5 1\",\"pages\":\"356-378\"},\"PeriodicalIF\":7.3000,\"publicationDate\":\"2023-01-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Annual Reviews in Control\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2301.00935\",\"RegionNum\":2,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"AUTOMATION & CONTROL SYSTEMS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annual Reviews in Control","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.48550/arXiv.2301.00935","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AUTOMATION & CONTROL SYSTEMS","Score":null,"Total":0}
A Survey of Feedback Particle Filter and related Controlled Interacting Particle Systems (CIPS)
In this survey, we describe controlled interacting particle systems (CIPS) to approximate the solution of the optimal filtering and the optimal control problems. Part I of the survey is focussed on the feedback particle filter (FPF) algorithm, its derivation based on optimal transportation theory, and its relationship to the ensemble Kalman filter (EnKF) and the conventional sequential importance sampling-resampling (SIR) particle filters. The central numerical problem of FPF -- to approximate the solution of the Poisson equation -- is described together with the main solution approaches. An analytical and numerical comparison with the SIR particle filter is given to illustrate the advantages of the CIPS approach. Part II of the survey is focussed on adapting these algorithms for the problem of reinforcement learning. The survey includes several remarks that describe extensions as well as open problems in this subject.
期刊介绍:
The field of Control is changing very fast now with technology-driven “societal grand challenges” and with the deployment of new digital technologies. The aim of Annual Reviews in Control is to provide comprehensive and visionary views of the field of Control, by publishing the following types of review articles:
Survey Article: Review papers on main methodologies or technical advances adding considerable technical value to the state of the art. Note that papers which purely rely on mechanistic searches and lack comprehensive analysis providing a clear contribution to the field will be rejected.
Vision Article: Cutting-edge and emerging topics with visionary perspective on the future of the field or how it will bridge multiple disciplines, and
Tutorial research Article: Fundamental guides for future studies.