Advisor Information
Azad Azadmanesh
Location
UNO Criss Library, Room 225
Presentation Type
Oral Presentation
Start Date
6-3-2015 1:30 PM
End Date
6-3-2015 1:45 PM
Abstract
This study presents a hybrid approach to indoor object detection and tracking. The test area is partitioned into regions that are covered by sensors placed on the ceilings. The approximate location of an object is determined by these sensors using RSSI. The region determination is then followed by localizing the exact location of the object using Building Information Modeling (BIM) and the 3D stereo image measurements. To determine the coverage ratio of a region for better placement of sensors, the 3D space is partitioned into 2D planes whose coverage points are determined by a newly developed software product called LOCOPac. The package is not only able to simulate the 2D coverage under various distribution algorithms and parameters such as sensor failures and energy efficiency, it also assists in the visualization of an object roaming in different regions of the test space. Other than indoor localization, some simulation results of sensor distribution algorithms that are more suitable for outdoor applications will also be presented. Based on the experimental and simulation results, the proposed approach to indoor localization and LOCOPac has the potential to be used for real world applications.
A Hybrid Approach to Indoor Sensor Area Localization and Coverage
UNO Criss Library, Room 225
This study presents a hybrid approach to indoor object detection and tracking. The test area is partitioned into regions that are covered by sensors placed on the ceilings. The approximate location of an object is determined by these sensors using RSSI. The region determination is then followed by localizing the exact location of the object using Building Information Modeling (BIM) and the 3D stereo image measurements. To determine the coverage ratio of a region for better placement of sensors, the 3D space is partitioned into 2D planes whose coverage points are determined by a newly developed software product called LOCOPac. The package is not only able to simulate the 2D coverage under various distribution algorithms and parameters such as sensor failures and energy efficiency, it also assists in the visualization of an object roaming in different regions of the test space. Other than indoor localization, some simulation results of sensor distribution algorithms that are more suitable for outdoor applications will also be presented. Based on the experimental and simulation results, the proposed approach to indoor localization and LOCOPac has the potential to be used for real world applications.