Study on data-fuzzy-based ALV lateral controller design and experiments

Abstract

Autonomous land vehicle is a complex, nonlinear and uncertain system. At present, one of the key problems in the researching field of ALV is the lack of robustness in motion control. In order to solve this problem, the kinematics model of ALV was presented at first in this paper. Then based on fuzzy logical system of "data-fuzzy", a lateral controller of Mamdani reasoning system for ALV was designed. The design of the controller utilized not only the in-out data information but also the language information generated by expertise, so the controller designed have a better performance, and can accommodate a wide range of velocity and different conditions of the road. The experiment results show that the controller designed has enough robustness, and can make ALV follow reference paths quickly and accurately.