Location dependent variable information gain approach to multi robot coverage problem

In this paper, we propose a new approach for selecting and updating information gains for multi robot exploration problem. The exploration problem, also known as coverage problem, is converted into an optimization problem in which solution of a distributed objective function has to be found. In this work, information gain part of the objective function consists of two different parts, a static and a dynamic part. The dynamic part provides weighting the dense areas in sense of information and it is also updated at each optimization step. In order to find a control law, objective function is solved using frontiers and dynamic Voronoi partitions. Additionally, a case study of mine clearance is given with simulation results.