 | Pandian Arunkumar |
   |
| Project: |
| |
| Supervisor: |
| |
| Host University: |
| |
| Personal introduction | top |
Hailing from Puducherry (then Pondicherry), a peaceful and scenic coastal city situated in southeast India, I obtained my undergraduate degree (B.Tech) in the field of Electronics and Communication Engineering from Pondicherry Engineering College in 2006. My passion for technology motivated me to continue my education further and I, now realise that I could not have made a better decision than to join the Erasmus Mundus Master in Vision and Robotics (Vibot). As an innovative and enthusiastic fresh graduate with an outstanding academic record throughout my career, I am currently looking for a career in the industry that would offer the satisfaction of applying my technical knowledge to solve the engineering challenges of today and also offers scope to shape the technology of tomorrow. In addition to my indepth technical knowledge confirmed by the achievement section of my CV, I consider the soft skills that I have gained through the wide range of roles played in different team projects to be my biggest asset. The exposure to the different domains such as Electronics, Software Engineering, Telecommunicaiton, Computer Vision and Robotics that I have gained through my degrees would offer the flexibility of employing me in a wide range of interdisciplinary projects.
My hobbies include listening to music, playing cricket, pencil sketching, riding my Hero Honda Splendor and playing badminton. I am highly interested in travelling and my heartfelt gratitude to Vibot for having enabled me to explore Scotland, England, Spain, France, Switzerland, Belgium and Italy.
| |
| |
| Autonomous off-road navigation is a highly complicated task for a robot owing to the different kinds of obstacle it could encounter. Inparticular, water hazards such as puddles and ponds are very common in outdoor environments and are hard to characterise even by the state-of-the-art sensing technologies. This work employs concepts from stereo vision and polarization imaging to detect water hazards for robot navigation.
|
|
| |
| Though, a wide range of techniques such as colour image segmentation, multi-feature based schemeand thermal infrared imagery have been attempted, a system capable of detecting any water hazard camouflaged by various terrain covers is yet to be developed. The potential for applying polarization imaging setups starting from a two camera stereo to a four camera setup are experimented in this work and their relative merits and demerits are analysed. | |
| |
| Contributions of this work | top | | The important contributions of this work include the demonstration of the sensory augmentation capabilities of polarization imaging, an investigation into the stereo matching score suited to polarization images (zero mean normalised cross correlation and the Hamming distance from census transformed images) and finally, the establishment of guidelines for future research in this field. | |
| |
| I take this opportunity to thank my supervisors, the coordinating universities of the vibot master's course, Vibot staff members, my friends and family for their valuable support. | |
| |
Dr. Olivier Morel
Laboratoire Le2i UMR-CNRS 5158
IUT Le Creusot
France
Olivier.Morel(anti-spam-at)u-bourgogne.fr (replace (anti-spam-at) by @)
Tel : +33 (0)3.85.73.10.90
Dr. Simon Lacroix
Research scientist in mobile robotics
LAAS/CNRS
France
Simon.Lacroix@laas.fr
Tel: +33 561 33 62 66
| |
| |
| |
| |
| |
| |
