| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 2.18 MB | Adobe PDF |
Advisor(s)
Abstract(s)
This paper addresses the three-dimensional modelling of large scale underground galleries, such as traffic tunnels and mines. This work employs techniques from mobile robotics to achieve an autonomous mobile modelling system, adapted to general underground environments. So far, the state-of-the-art methods in underground modelling remain restricted to environments in which pronounced geometric features are abundant.
This limitation is a consequence of the scan matching algorithms used to solve the localization and registration problems.
This work aims to extend the modeling capability to structures characterized by uniform geometry and smooth surfaces, as is the case of road and train tunnels.
A visual monocular Simultaneous Localization and Mapping (MonoSLAM) approach based on the Extended Kalman Filter (EKF) and complemented by the introduction of inertial measurements in the prediction step, allows our system to build threedimensional models and localize himself over long distances, using exclusively sensors carried on board a mobile platform.
By feeding the Extended Kalman Filter with inertial data we were able to overcome the major problem related with MonoSLAM implementations, known as scale factor ambiguity, which emerges from the absence of metric measurements in monocular images. The monocular visual features used in MonoSLAM were extracted by the SIFT algorithm, and inserted directly in the EKF mechanism according to the Inverse Depth Parametrization. Through the 1-Point RANSAC (Random Sample Consensus) wrong frame-to-frame feature matches were rejected.
To build the model, vertical cross-sections of the gallery, acquired by a laser range finder sensor, are placed on a common reference frame using the estimated localization.
The system was tested based on a dataset acquired inside a real road tunnel. Results from the localization strategy and the modelling process are presented.
Description
Robotica 2012: 12th International Conference on Autonomous Robot Systems and Competitions
April 11, 2012, Guimarães, Portugal