Neuro-fuzzy motion control for mobile robot

Abstract

This paper focuses on autonomous motion control of a nonholonomic platform with a robotic arm. We focus on two components of the motion control architecture and the strong interactions binding them: motion planning and execution. We suggest the use of dynamic path planning to solve the problem of locomotion of a mobile platform, and reactive approaches to solve the motion planning of the arm manipulator. The execution part is designed as an adaptive neuro-fuzzy controller. In real world systems, sensor-based motion control becomes essential to deal with model uncertainties and unexpected obstacles.