Control of Robotic Arm using Neuro-fuzzy Systems

Περίληψη: Στόχος της παρούσης εργασίας είναι η ανάπτυξη αξιόπιστου συστήματος νευροασαφούς ελέγχου σε περιβάλλον MATLAB(Simulink) και η σύνδεσή του με μηχανικό σύστημα ρομποτικού βραχίονα 2 βαθμών ελευθερίας. Οι τεχνικές που θα αναπτυχθούν στο πλαίσιο της παρούσας εργασίας είναι κατάλληλες για επίλυση γραμμικών και μη γραμμικών συστημάτων και μπορούν να εφαρμοστούν σε πλή-θος διαφορετικών εφαρμογών.Summarization: Intelligent mechanical systems, which are equipped with sensors and exciters made of piezoelectric materials, interact with the environment through an active control mechanism. One such example is the case of a robotic arm. However, due to imperfections in the mechanical model, but also in the sensory organs, a certain degree of uncertainty always enters the system. Classic control can operate under conditions of uncertainty, but with reduced efficiency. Designing nonlinear feedback control rules using classical methods is a difficult task. It is therefore recommended to use intelligent control techniques, which are based on neurofuzzy systems. The aim of this work is the development of a reliable neurofuzzy control system in MATLAB (Simulink) environment and its connection to a mechanical robotic arm system of 2 degrees of freedom. The techniques that will be developed in the context of this work are suitable for solving linear and non-linear systems and can be applied in a numeral of different applications