Open Collections

Real-time 3D motion tracking for a mobile robot

University of British Columbia

This thesis presents a vision-based tracking system suitable for autonomous robot vehicle guidance. The system includes a head with three on-board CCD cameras, called Triclops®, which can be mounted anywhere on a mobile vehicle. By processing consecutive trinocular sets of precisely aligned and rectified images, the local 37J trajectory of the vehicle in an unstructured environment can be tracked. The use of three cameras enables the system to construct an accurate representation of its environment and therefore results in accurate motion estimation. First, a 3D representation of stable corners in the image scene is generated using a stereo algorithm. Second, motion is estimated by tracking matched features over time. The motion equation with 6 degrees of freedom is then solved using an iterative least squares fit algorithm. Finally, a Kalman filter implementation is used to optimize the 3D representation of scene features in world coordinates. The system has been implemented on a Pentium processor in conjunction with a Triclops and a Matrox Meteor® frame grabber. The system is capable of detecting rotations with 3% error over a 90 degree rotation and translations with 13% average error over a 6m long path.

Thesis/Dissertation

application/pdf

2009-07-03

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/10024

Electrical and Computer Engineering

University of British Columbia

UBCV

DSpace