GA-Based Optimization of Control Strategies for a 3R Nonplanar Robotic Manipulator

Robotics and control are key to automation and industry. In this work, the model of a 3R nonplanar manipulator is simulated in MATLAB. A sinusoidal trajectory was used for the robot's movement, while Gaussian noise was added to the system to simulate real-world application. The impact of the noise was handled using a moving average filter. The effect of four different control systems was evaluated, including Proportional-Integral-Derivative, Fuzzy Logic, Adaptive Neuro-Fuzzy Inference System, and Sliding Mode Control, on each joint torque. The parameters for each controller were tuned with a Genetic Algorithm. The results showed that the most accurate trajectory tracking was provided by Sliding Mode Control, especially when noise was present, while the Adaptive Neuro-Fuzzy Inference System proved to be very flexible. The study discusses the applicability of these controllers, their optimization, and their effect on manipulators under uncertainty.