A constant-time efficient stereo SLAM system
Continuous, real-time mapping of an environment using a camera requires a constanttime estimation engine. This rules out optimal global solving such as bundle adjustment. In this article, we investigate the precision that can be achieved with only local estimation of motion and structure provided by...
| Main Authors: | , , , , |
|---|---|
| Format: | Journal article |
| Published: |
BMVA Press
2009
|
| _version_ | 1797088429035487232 |
|---|---|
| author | Mei, C Sibley, G Cummins, M Newman, P Reid, I |
| author_facet | Mei, C Sibley, G Cummins, M Newman, P Reid, I |
| author_sort | Mei, C |
| collection | OXFORD |
| description | Continuous, real-time mapping of an environment using a camera requires a constanttime estimation engine. This rules out optimal global solving such as bundle adjustment. In this article, we investigate the precision that can be achieved with only local estimation of motion and structure provided by a stereo pair. We introduce a simple but novel representation of the environment in terms of a sequence of relative locations. We demonstrate precise local mapping and easy navigation using the relative map, and importantly show that this can be done without requiring a global minimisation after loop closure. We discuss some of the issues that arise from using a relative representation, and evaluate our system on long sequences processed at a constant 30-45 Hz, obtaining precisions down to a few metres over distances of a few kilometres. |
| first_indexed | 2024-03-07T02:49:57Z |
| format | Journal article |
| id | oxford-uuid:ad591a5a-2021-4c39-be16-3e1611049bae |
| institution | University of Oxford |
| last_indexed | 2024-03-07T02:49:57Z |
| publishDate | 2009 |
| publisher | BMVA Press |
| record_format | dspace |
| spelling | oxford-uuid:ad591a5a-2021-4c39-be16-3e1611049bae2022-03-27T03:34:59ZA constant-time efficient stereo SLAM systemJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:ad591a5a-2021-4c39-be16-3e1611049baeSymplectic Elements at OxfordBMVA Press2009Mei, CSibley, GCummins, MNewman, PReid, IContinuous, real-time mapping of an environment using a camera requires a constanttime estimation engine. This rules out optimal global solving such as bundle adjustment. In this article, we investigate the precision that can be achieved with only local estimation of motion and structure provided by a stereo pair. We introduce a simple but novel representation of the environment in terms of a sequence of relative locations. We demonstrate precise local mapping and easy navigation using the relative map, and importantly show that this can be done without requiring a global minimisation after loop closure. We discuss some of the issues that arise from using a relative representation, and evaluate our system on long sequences processed at a constant 30-45 Hz, obtaining precisions down to a few metres over distances of a few kilometres. |
| spellingShingle | Mei, C Sibley, G Cummins, M Newman, P Reid, I A constant-time efficient stereo SLAM system |
| title | A constant-time efficient stereo SLAM system |
| title_full | A constant-time efficient stereo SLAM system |
| title_fullStr | A constant-time efficient stereo SLAM system |
| title_full_unstemmed | A constant-time efficient stereo SLAM system |
| title_short | A constant-time efficient stereo SLAM system |
| title_sort | constant time efficient stereo slam system |
| work_keys_str_mv | AT meic aconstanttimeefficientstereoslamsystem AT sibleyg aconstanttimeefficientstereoslamsystem AT cumminsm aconstanttimeefficientstereoslamsystem AT newmanp aconstanttimeefficientstereoslamsystem AT reidi aconstanttimeefficientstereoslamsystem AT meic constanttimeefficientstereoslamsystem AT sibleyg constanttimeefficientstereoslamsystem AT cumminsm constanttimeefficientstereoslamsystem AT newmanp constanttimeefficientstereoslamsystem AT reidi constanttimeefficientstereoslamsystem |