Software for the Learning of Monte Carlo Localization Strategies using omnidirectional Images.
Computing the localization of a robot in an environment is an essential task when the robot has to carry out an autonomous task in that environment. In the scientific literature we can find many methods, and one of the most extended currently is Monte Carlo Localization algorithm. On the other hand in this kind of applications, the appearance-based approach has attracted the interest of researchers recently due to its simplicity and robustness. In this work, we present a tool we developed to be used in a robotics subject. It provides the students an intuitive platform to test the Monte Carlo Localization algorithms when working with some set of omnidirectional images captured in real environments and with an appearance-based descriptor. This tool is very useful to provide real data and to facilitate some mechanisms to understand this probabilistic localization algorithm.