Breadcrumb: An indoor simultaneous localization and mapping system for mobile devices
GPS as a localization system performs poorly indoors. Other methods were developed for indoor navigation, many of which required external infrastructure making them location and environment dependent. One approach that does not require external infrastructure is dead reckoning. Given users with smar...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | text |
| Published: |
Animo Repository
2016
|
| Subjects: | |
| Online Access: | https://animorepository.dlsu.edu.ph/faculty_research/3507 https://animorepository.dlsu.edu.ph/context/faculty_research/article/4509/type/native/viewcontent/SAS.2016.7479900 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | De La Salle University |
| Summary: | GPS as a localization system performs poorly indoors. Other methods were developed for indoor navigation, many of which required external infrastructure making them location and environment dependent. One approach that does not require external infrastructure is dead reckoning. Given users with smartphones, dead reckoning helps users find their way by tracking their position and mapping their environment, achieving Simultaneous Localization and Mapping (SLAM). SLAM can be achieved through Inertial Navigation Systems (INSs). However, conventional INSs alone are inherently erroneous due to sensor drift and error accumulation, necessitating modification to compensate. One modification utilizes cameras to aid estimation, creating Vision-Aided Inertial Navigation Systems (V-INS). Typically, conventional V-INSs achieve motion estimation by integrating accelerometer readings, whereas conventional step-based INSs detect steps and estimate stride length. Thus, this research modified a V-INS by using a step-based approach for motion estimation. Results on distance estimation, final displacement, and total position error show that the modified system generally performed better than the basis systems. For error as percentage of total distance travelled, INS had an average of 5.21%, V-INs had 10.53%, and Breadcrumb had 3.52%. For percentage of error in final displacement, INS had an average of 16.09%, V-INS had 12.40%, and Breadcrumb had 5.75%. © 2016 IEEE. |
|---|