Colloquium "Sensor placement planning and exploration strategies with mobile robot"

Abstract:

The aim is to find a method for environment exploration that minimizes the experimentation time. Sensor placement planning is needed for goal directed acquisition of 3D data. The task of the sensor placement planning algorithm is to find a set of sensor configurations needed for obtaining a detailed environment model. Since a typical 3D laser scan takes about 3 minutes for one position, depending on resolution, it is desirable to minimize the number of scanning positions to acquire a complete model of indoor environment. This leads to an optimization problem similar to the Art Gallery Problem (AGP) (where to place guards such that the entire gallery is guarded). The developed strategies are divided into 2D and 3D exploration strategies and combining them it is possible to minimize the number of the sensor positions taking into account the exploration time.

CV:

Ivan Maurović received his Master of Science in Electrical Engineering and Information Technology degree with honors (cum laude), in 2010 from UNIZG-FER. In 2011, he joined the Department of Control and Computer Engineering as a research assistant on SEE-ERA.NET PLUS project ThermalMapper and he enrolled into the graduate (doctoral) programme. He is a member of AMOR research group. Since 2012, Ivan has been employed by the ACROSS project. His main research activities are in the area of autonomous mobile robot control and exploration strategies in complex environments including 3D exploration problem.