The six-degrees-of-freedom (6-dof) robotic manipulator control using genetic algorithm and elman neural network implemented generalized predictive control

In this thesis study, Generalized Predictive Control (GPC), Neural Generalized Predictive Control (NGPC), Simple Genetic Algorithm implemented GPC (SGA-GPC) and Recurrent Elman Neural Network implemented NGPC (ENGPC) algorithms belong to the class of Model Based Predictive Control (MBPC) were investigated and each of them was applied to a 6-DOF (Degrees-Of-Freedom) robotic manipulator as SISO (Single Input Single Output) and MIMO (Multiple Inputs Multiple Outputs) for the trajectory control based joint. Dynamics modeling of the robotic manipulator was made by using the Lagrange-Euler equations. The frictional effects, the state of carrying and falling load were also added to dynamics model. In addition, the random distortions between and were added to the torques applied to the joints in every control step, and the effect to the performance of the distortions was investigated. Dynamics model was transformed into robotic arm simulator by using the fourth-order Runge-Kutta integration method. The trajectory planning for the joints of the robotic arm was designated according to the sinusoidal trajectories principles. The control algorithms were compared with themselves for different examples and cases.