Trajectory tracking control of a class of lower mobility parallel manipulators using ANFIS

摘要

In engineering applications, very precise motion of parallel manipulators to track the desired trajectory has received a great deal of attention. This paper presents a new methodology to investigate this problem based on the Adaptive Neuro-Fuzzy Inference System (ANFIS) controller for a particular family of lower mobility parallel kinematic manipulators. This novel structure has three degrees of pure translational freedom by employing architecture of a modified version of Delta parallel mechanism. The proposed mechanism like ordinary Delta parallel mechanism has three degrees of pure translational freedom, but the position of robot's three active joints relative to each other is the main difference between this mechanism and Delta parallel mechanism, which has caused the change in geometry of platforms, and it shapes the asymmetrical structure in the robot mechanism and its workspace. Inverse dynamic modeling is performed based on principle of virtual work. The conventional PD controllers of Computed Torque Control (C-T) method usually need manual retuning to make a successful task, particularly in the presence of disturbance. In the present paper, we study feasibility of applying ANFIS controller instead of conventional controller used in C-T method to cope with above problem for this nonlinear complex system. Several computer simulations imply that proposed method is encouraging compared with ordinary C-T method implemented with conventional controller.