Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random Fields
In this paper, we present algorithms for predicting a spatio-temporal random field measured by mobile robotic sensors under uncertainties in localization and measurements. The spatio-temporal field of interest is modeled by a sum of a time-varying mean function and a Gaussian Markov random field (GM...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/18/9/2866 |
| _version_ | 1811213485952991232 |
|---|---|
| author | Mahdi Jadaliha Jinho Jeong Yunfei Xu Jongeun Choi Junghoon Kim |
| author_facet | Mahdi Jadaliha Jinho Jeong Yunfei Xu Jongeun Choi Junghoon Kim |
| author_sort | Mahdi Jadaliha |
| collection | DOAJ |
| description | In this paper, we present algorithms for predicting a spatio-temporal random field measured by mobile robotic sensors under uncertainties in localization and measurements. The spatio-temporal field of interest is modeled by a sum of a time-varying mean function and a Gaussian Markov random field (GMRF) with unknown hyperparameters. We first derive the exact Bayesian solution to the problem of computing the predictive inference of the random field, taking into account observations, uncertain hyperparameters, measurement noise, and uncertain localization in a fully Bayesian point of view. We show that the exact solution for uncertain localization is not scalable as the number of observations increases. To cope with this exponentially increasing complexity and to be usable for mobile sensor networks with limited resources, we propose a scalable approximation with a controllable trade-off between approximation error and complexity to the exact solution. The effectiveness of the proposed algorithms is demonstrated by simulation and experimental results. |
| first_indexed | 2024-04-12T05:46:05Z |
| format | Article |
| id | doaj.art-3c4d06ffd9d8456aaa70bed399575041 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-12T05:46:05Z |
| publishDate | 2018-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-3c4d06ffd9d8456aaa70bed3995750412022-12-22T03:45:26ZengMDPI AGSensors1424-82202018-08-01189286610.3390/s18092866s18092866Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random FieldsMahdi Jadaliha0Jinho Jeong1Yunfei Xu2Jongeun Choi3Junghoon Kim4Monsanto, St. Louis, MO 63146, USASchool of Mechanical Engineering, Yonsei University, Seoul 03722, KoreaDenso International America, Inc., San Jose, CA 95110, USASchool of Mechanical Engineering, Yonsei University, Seoul 03722, KoreaDepartment of Civil and Environmental Engineering, Yonsei University, Seoul 03722, KoreaIn this paper, we present algorithms for predicting a spatio-temporal random field measured by mobile robotic sensors under uncertainties in localization and measurements. The spatio-temporal field of interest is modeled by a sum of a time-varying mean function and a Gaussian Markov random field (GMRF) with unknown hyperparameters. We first derive the exact Bayesian solution to the problem of computing the predictive inference of the random field, taking into account observations, uncertain hyperparameters, measurement noise, and uncertain localization in a fully Bayesian point of view. We show that the exact solution for uncertain localization is not scalable as the number of observations increases. To cope with this exponentially increasing complexity and to be usable for mobile sensor networks with limited resources, we propose a scalable approximation with a controllable trade-off between approximation error and complexity to the exact solution. The effectiveness of the proposed algorithms is demonstrated by simulation and experimental results.http://www.mdpi.com/1424-8220/18/9/2866Gaussian markov random fieldfully Bayesianmobile sensor networklocalization uncertainty |
| spellingShingle | Mahdi Jadaliha Jinho Jeong Yunfei Xu Jongeun Choi Junghoon Kim Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random Fields Sensors Gaussian markov random field fully Bayesian mobile sensor network localization uncertainty |
| title | Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random Fields |
| title_full | Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random Fields |
| title_fullStr | Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random Fields |
| title_full_unstemmed | Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random Fields |
| title_short | Fully Bayesian Prediction Algorithms for Mobile Robotic Sensors under Uncertain Localization Using Gaussian Markov Random Fields |
| title_sort | fully bayesian prediction algorithms for mobile robotic sensors under uncertain localization using gaussian markov random fields |
| topic | Gaussian markov random field fully Bayesian mobile sensor network localization uncertainty |
| url | http://www.mdpi.com/1424-8220/18/9/2866 |
| work_keys_str_mv | AT mahdijadaliha fullybayesianpredictionalgorithmsformobileroboticsensorsunderuncertainlocalizationusinggaussianmarkovrandomfields AT jinhojeong fullybayesianpredictionalgorithmsformobileroboticsensorsunderuncertainlocalizationusinggaussianmarkovrandomfields AT yunfeixu fullybayesianpredictionalgorithmsformobileroboticsensorsunderuncertainlocalizationusinggaussianmarkovrandomfields AT jongeunchoi fullybayesianpredictionalgorithmsformobileroboticsensorsunderuncertainlocalizationusinggaussianmarkovrandomfields AT junghoonkim fullybayesianpredictionalgorithmsformobileroboticsensorsunderuncertainlocalizationusinggaussianmarkovrandomfields |