Section: Software and Platforms

Fuzzy logic tool

Participant : Joshué Pérez Rastelli [correspondant] .

A fuzzy logic module has been implemented to translate human knowledge to driverless control processes, considering risk/warning situation. Fuzzy logic techniques have been widely implemented in different industrial process in the last decade. For this reason, many libraries, mainly developed in C++, are easily found in the literature. The goal is to achieve the autonomous driving of the vehicle using simple sentences defined in a rule base. Then, it is just necessary to define the input and output membership functions. Two modules based on fuzzy logic libraries were created. One of them was developed in order to compare the classic controller of a previous work with a fuzzy controller to improve the lateral control tracking previously developed. Moreover, another module to warn speed references at intersections with traffic lights was done in the framework of the project CoDrive. The idea is that the vehicle is able to know at which speed it must travel to avoid abrupt braking and save fuel.