Exjobbsförslag från företag

Detta är ett uppsatsförslag hämtat från Nationella Exjobb-poolen. Klicka här för att komma tillbaka till samtliga exjobbsförslag.

Förslaget inkom 2010-03-16

Implementation of a Robot Localization and Mapping Strategy within a Cognitive Architecture

An autonomous mobile agent needs to represent its surroundings in order to reason an plan actions within it. The typical spatial knowledge representations used in mobile robotics are purely metrical and rely on information extracted from simple, but accurate metric sensors. However, as the robots are designed to perform human-like tasks in more and more complex and dynamic environments, metrical global maps become harder to control and observe. Moreover, it is not clear that the level of detail offered by such maps is necessary, or even desirable, when the agent is a cognitive system intended to interact with the world in a human-like way. It is commonly accepted that the spatial knowledge of a cognitive agent should be abstracted in order to make it robust to dynamic variations, easier to maintain and useful for spatial reasoning. At the same time, the agent should be able to exploit sensory information that might be complex and non-metric, yet reflects crucial aspects of the environment. To address those issues, a new localization and mapping strategy was proposed based on a cognitive mapping framework that builds on the concepts of discrete places and scenes expressed in terms of arbitrary, possibly complex features as well as local spatial relations.

This project aims at implementing the localization and mapping algorithms in C++ within an integrated cognitive architecture based on the CoSy Architecture Schema Toolkit (http://www.cs.bham.ac.uk/research/projects/cosy/cast/index.php) and integrate the algorithms with other components of an integrated cognitive system. During the project work, the student will work as part of a dynamic group (http://www.cas.kth.se/) performing research on spatial understanding for mobile robots and will closely collaborate with researchers developing the cognitive mapping framework.

Requirements: background in computer science, engineering or equivalent. Strong scientific programming skills (primarily in C++). Experience with Linux-based computing environments. Fundamental knowledge in pattern recognition, computer vision, digital image processing or signal processing is a plus.

The project is intended to start on May 1st and is expected to finish by the end of September; however, students will also be accepted for shorter periods of time (e.g. May until July). Applications should be made as soon as possible. Potential applicants are invited to email a CV, study report, and a very brief email summarizing the experience, personal interests and skills related to:
- programming, computing environments
- computer vision / robotics

- Patric Jensfelt (http://www.csc.kth.se/~patric/)
- Andrzej Pronobis (http://www.pronobis.pro/)

Contact: Andrzej Pronobis (http://www.pronobis.pro/)


Informationen om uppsatsförslag är hämtad från Nationella Exjobb-poolen.