Faculty Host: Dinesh Pai

Abstract:

Multisensory data acquisition is a newly addressed research held in robotics. The first part of the talk will deal with the description of the basic principles of multisensory data acquisition, e.g. time- synchronization of dierent sensors. A system realizing these principle is presented and applications such as photo-realistic model generation are introduced. Besides robotic applications, these models can e.g. be used for the conservation of cultural goods in virtual museums or in reverse engineering. A special emphasis is put on on the dierent sensor types and their spatial calibration. The second part will deal with the multisensory robotic application of the knowledge gained in the previous work. The basic ideas of combining C-space (robot conguration space) exploration with automated multisensory inspection tasks to occur in grasp planning and similar applications at DLR is presented. This work will be part of a PhD thesis and takes into account the previous work at DLR and SFU[1].

Speaker Bio:

Michael received his master's degree in electrical engineering, emphasis on automatization and mechatronics, University of Hanover, Germany in 2000, graduating second best of all year 2000 graduates. Since then he has been working for the German Aerospace Center (DLR), Institute of Robotics and Mechatronics, situated near Munich. He is mainly involved in the design of vision based systems and its multisensory integration for robotic applications as well as digitization purposes. Michael is pursuing his PhD at the University of Hanover, Germany in conjunction with DLR. He just visited the SFU Robotics Lab (Dr Kamal Gupta's research group) for a 3 month period.

References

1. http://www.dlr.de
2. http://www.robotic.de/mechatronics/scanner/
3. http://www.ensc.sfu.ca/research/kamal/

[1] Simon Fraser University (SFU), School of Engineering Science, Burnaby, Canada