Discovering Planes and Collapsing the State Space in Visual SLAM

AP Gee; Chekhlov D; WW Mayol-Cuevas; AD Calway

Discovering Planes and Collapsing the State Space in Visual SLAM

AP Gee, Chekhlov D, WW Mayol-Cuevas, AD Calway

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

41 Citations (Scopus)

Abstract

Recent advances in real-time visual SLAM have been based primarily on mapping isolated 3-D points. This presents difficulties when seeking to extend operation to wide areas, as the system state becomes large, requiring increasing computational effort. In this paper we present a novel approach to this problem in which planar structural components are embedded within the state to represent mapped points lying on a common plane. This collapses the state size, reducing computation and improving scalability, as well as giving a higher level scene description. Critically, the plane parameters are augmented into the SLAM state in a proper fashion, maintaining inherent uncertainties via a full covariance representation. Results for simulated data and for real-time operation demonstrate that the approach is effective.

Translated title of the contribution	Discovering Planes and Collapsing the State Space in Visual SLAM
Original language	English
Title of host publication	18th British Machine Vision Conference
Publication status	Published - Sept 2007

Bibliographical note

Conference Proceedings/Title of Journal: Proceedings of the 18th British Machine Vision Conference
Conference Organiser: British Machine Vision Association

Access to Document

Handle.net

Persistent link

Cite this

@inproceedings{9eaa9382fd95484b897975ea750f70ec,

title = "Discovering Planes and Collapsing the State Space in Visual SLAM",

abstract = "Recent advances in real-time visual SLAM have been based primarily on mapping isolated 3-D points. This presents difficulties when seeking to extend operation to wide areas, as the system state becomes large, requiring increasing computational effort. In this paper we present a novel approach to this problem in which planar structural components are embedded within the state to represent mapped points lying on a common plane. This collapses the state size, reducing computation and improving scalability, as well as giving a higher level scene description. Critically, the plane parameters are augmented into the SLAM state in a proper fashion, maintaining inherent uncertainties via a full covariance representation. Results for simulated data and for real-time operation demonstrate that the approach is effective.",

author = "AP Gee and Chekhlov D and WW Mayol-Cuevas and AD Calway",

note = "Conference Proceedings/Title of Journal: Proceedings of the 18th British Machine Vision Conference Conference Organiser: British Machine Vision Association",

year = "2007",

month = sep,

language = "English",

booktitle = "18th British Machine Vision Conference",

}

TY - GEN

T1 - Discovering Planes and Collapsing the State Space in Visual SLAM

AU - Gee, AP

AU - D, Chekhlov

AU - Mayol-Cuevas, WW

AU - Calway, AD

N1 - Conference Proceedings/Title of Journal: Proceedings of the 18th British Machine Vision Conference Conference Organiser: British Machine Vision Association

PY - 2007/9

Y1 - 2007/9

N2 - Recent advances in real-time visual SLAM have been based primarily on mapping isolated 3-D points. This presents difficulties when seeking to extend operation to wide areas, as the system state becomes large, requiring increasing computational effort. In this paper we present a novel approach to this problem in which planar structural components are embedded within the state to represent mapped points lying on a common plane. This collapses the state size, reducing computation and improving scalability, as well as giving a higher level scene description. Critically, the plane parameters are augmented into the SLAM state in a proper fashion, maintaining inherent uncertainties via a full covariance representation. Results for simulated data and for real-time operation demonstrate that the approach is effective.

AB - Recent advances in real-time visual SLAM have been based primarily on mapping isolated 3-D points. This presents difficulties when seeking to extend operation to wide areas, as the system state becomes large, requiring increasing computational effort. In this paper we present a novel approach to this problem in which planar structural components are embedded within the state to represent mapped points lying on a common plane. This collapses the state size, reducing computation and improving scalability, as well as giving a higher level scene description. Critically, the plane parameters are augmented into the SLAM state in a proper fashion, maintaining inherent uncertainties via a full covariance representation. Results for simulated data and for real-time operation demonstrate that the approach is effective.

M3 - Conference Contribution (Conference Proceeding)

BT - 18th British Machine Vision Conference

ER -

Discovering Planes and Collapsing the State Space in Visual SLAM

Abstract

Bibliographical note

Access to Document

Handle.net

Fingerprint

Cite this