Three more reasons to attend my talk (except eating pizza :)

• I'll try to unite the areas which are used to be Artificial Intelligence and which are no longer conventially (and unfortunately ?) considered to be it: Computer Vision, Robotics and Uncertainty.
• The robot "Boticelli" will be introduced to the audience. Live Demo!
• Frusrated of making slides in "Power"point or LaTeX. Come and see how you can make them in HTML!

Abstract:

We consider an occupancy-based approach for fusing visual range data and identify three major problems of this approach. The first problem concerns the visual range sensor, the sensor model of which has to be developed in order to be used in fusing. The second problem deals with the combination rule which in many cases assumes the independence of range data, contrary to the usual situation. Finally, the third problem concerns the redundancy of stored and processed data, which results from using the grid representation of the occupancy function and which, in fact, is the main obstacle to building 3D occupancy world models. We propose a solution to these problems by proposing a new range data fusion technique based on regression. This technique uses the evidence theory in assigning occupancy values and builds the occupancy function by fitting the sample data provided by a sensor with a piecewise linear function. Having developed a general framework for our approach, we apply it to building 3D occupancy models to be used in autonomous robot exploration. Data obtained by running a single-camera robot will be presented.

See my WorldModeling page for more details on the presented topics.