An optimization based method for simultaneous localization and mapping

In this study, we present an optimization based solution to the simultaneous localization and mapping (SLAM) problem. In the proposed algorithm, the SLAM problem is considered as two optimization problems. These problems are solved using forward dynamic programming. In the first problem, it is assumed that map is known perfectly and the robot path is estimated. In the second problem, the estimated robot path with their corresponding measurements is used to identify map. As optimization problem in each step of dynamic programming have high nonlinearity and also differential evolution (DE) tends to find the globally optimal solution without being trapped at local maxima, DE is developed to solve dynamic programming in each step of time. Some simulations and experiments are presented to illustrate the proposed algorithm and exhibit its performance.