Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian Processes
Information gathering (IG) algorithms aim to intelligently select a mobile sensor actions required to efficiently obtain an accurate reconstruction of a physical process, such as an occupancy map, or a magnetic field. Many recent works have proposed algorithms for IG that employ Gaussian processes (...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/19/5/1016 |
| _version_ | 1811263797143273472 |
|---|---|
| author | Alberto Viseras Dmitriy Shutin Luis Merino |
| author_facet | Alberto Viseras Dmitriy Shutin Luis Merino |
| author_sort | Alberto Viseras |
| collection | DOAJ |
| description | Information gathering (IG) algorithms aim to intelligently select a mobile sensor actions required to efficiently obtain an accurate reconstruction of a physical process, such as an occupancy map, or a magnetic field. Many recent works have proposed algorithms for IG that employ Gaussian processes (GPs) as underlying model of the process. However, most algorithms discretize the state space, which makes them computationally intractable for robotic systems with complex dynamics. Moreover, they are not suited for online information gathering tasks as they assume prior knowledge about GP parameters. This paper presents a novel approach that tackles the two aforementioned issues. Specifically, our approach includes two intertwined steps: (i) a Rapidly-Exploring Random Tree (RRT) search that allows a robot to identify unvisited locations, and to learn the GP parameters, and (ii) an RRT*-based informative path planning that guides the robot towards those locations by maximizing the information gathered while minimizing path cost. The combination of the two steps allows an online realization of the algorithm, while eliminating the need for discretization. We demonstrate that our proposed algorithm outperforms state-of-the-art both in simulations, and in a lab experiment in which a ground-based robot explores the magnetic field intensity within an indoor environment populated with obstacles. |
| first_indexed | 2024-04-12T19:51:41Z |
| format | Article |
| id | doaj.art-0e3c29d85d144806b6d408973d603d1c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-12T19:51:41Z |
| publishDate | 2019-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-0e3c29d85d144806b6d408973d603d1c2022-12-22T03:18:48ZengMDPI AGSensors1424-82202019-02-01195101610.3390/s19051016s19051016Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian ProcessesAlberto Viseras0Dmitriy Shutin1Luis Merino2German Aerospace Centre (DLR), Oberpfaffenhofen, 82234 Weßling, GermanyGerman Aerospace Centre (DLR), Oberpfaffenhofen, 82234 Weßling, GermanySchool of Engineering, Universidad Pablo de Olavide (UPO), 41013 Seville, SpainInformation gathering (IG) algorithms aim to intelligently select a mobile sensor actions required to efficiently obtain an accurate reconstruction of a physical process, such as an occupancy map, or a magnetic field. Many recent works have proposed algorithms for IG that employ Gaussian processes (GPs) as underlying model of the process. However, most algorithms discretize the state space, which makes them computationally intractable for robotic systems with complex dynamics. Moreover, they are not suited for online information gathering tasks as they assume prior knowledge about GP parameters. This paper presents a novel approach that tackles the two aforementioned issues. Specifically, our approach includes two intertwined steps: (i) a Rapidly-Exploring Random Tree (RRT) search that allows a robot to identify unvisited locations, and to learn the GP parameters, and (ii) an RRT*-based informative path planning that guides the robot towards those locations by maximizing the information gathered while minimizing path cost. The combination of the two steps allows an online realization of the algorithm, while eliminating the need for discretization. We demonstrate that our proposed algorithm outperforms state-of-the-art both in simulations, and in a lab experiment in which a ground-based robot explores the magnetic field intensity within an indoor environment populated with obstacles.https://www.mdpi.com/1424-8220/19/5/1016roboticsinformation gatheringGaussian processes (GPs)rapidly exploring random trees (RRT) |
| spellingShingle | Alberto Viseras Dmitriy Shutin Luis Merino Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian Processes Sensors robotics information gathering Gaussian processes (GPs) rapidly exploring random trees (RRT) |
| title | Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian Processes |
| title_full | Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian Processes |
| title_fullStr | Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian Processes |
| title_full_unstemmed | Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian Processes |
| title_short | Robotic Active Information Gathering for Spatial Field Reconstruction with Rapidly-Exploring Random Trees and Online Learning of Gaussian Processes |
| title_sort | robotic active information gathering for spatial field reconstruction with rapidly exploring random trees and online learning of gaussian processes |
| topic | robotics information gathering Gaussian processes (GPs) rapidly exploring random trees (RRT) |
| url | https://www.mdpi.com/1424-8220/19/5/1016 |
| work_keys_str_mv | AT albertoviseras roboticactiveinformationgatheringforspatialfieldreconstructionwithrapidlyexploringrandomtreesandonlinelearningofgaussianprocesses AT dmitriyshutin roboticactiveinformationgatheringforspatialfieldreconstructionwithrapidlyexploringrandomtreesandonlinelearningofgaussianprocesses AT luismerino roboticactiveinformationgatheringforspatialfieldreconstructionwithrapidlyexploringrandomtreesandonlinelearningofgaussianprocesses |