Multiple quadcopters exploring an unknown environment

Abstract: In this project a simulation of multiple quadcopters is created to solvethe problem of exploring and mapping an unknown environment. Thetwo major challengers are allocation of the route of the quadcopters tooptimize the mapping, as well as the merging of the three-dimensionalsub maps to get a final result. Occupancy grid maps will be used forthe allocation task to calculate the probability that a cell is occupiedby an obstacle. A cost and an utility function will also be calculatedfor every cell next to the unexplored areas to make a decisions for theroutes of the quadcopters. The cost makes the quadcopter travel tothe closest unexplored cell that is not occupied with some probabilitywhile the utility function works in a way that several quadcopters willnot travel to the same area. The combination of these two calculatesthe optimal path. The optimal merging of the sub maps is then to findan optimal rigid transformation that optimally aligns two sets of pointsin R3 in a least square sent. To do this an optimal translation andan optimal rotation using Singular Value Decomposition is calculated.Finally, algorithms which can be implemented in any suitable softwarefor simulation and demonstration of how this things can be handledin an practical environment, is described.