Enhanced Trajectory Tracking of a 6-DOF Robotic Manipulator Using GA–PID and ANN–PID Controllers

Abstract

Accurate trajectory tracking of multi–degree-of-freedom robotic manipulators remains a challenging control problem due to nonlinear dynamics, parameter uncertainties, and external disturbances. This paper presents an enhanced trajectory tracking approach for a 6-degree-of-freedom (6-DOF) robotic manipulator using hybrid Genetic Algorithm–PID (GA–PID) and Artificial Neural Network–PID (ANN–PID) controllers. In the proposed framework, the GA is employed to optimally tune the PID controller gains by minimizing a multi-objective fitness function based on tracking error, overshoot, and settling time, thereby improving control robustness. Additionally, an ANN-PID controller is developed to adaptively compensate for system nonlinearities and dynamic variations by learning the inverse dynamics of the manipulator in real time. The dynamic model of the 6-DOF manipulator is derived using the Euler–Lagrange formulation and is utilized for controller design and simulation. Extensive simulation studies are conducted under various trajectory profiles and external disturbance conditions to evaluate tracking accuracy, control effort, and robustness. The results demonstrate that both GA–PID and ANN–PID controllers significantly outperform the conventional PID controller, with the ANN-PID scheme achieving superior tracking precision and faster convergence. The proposed control strategies offer an effective and intelligent solution for high-performance robotic manipulation in industrial and autonomous applications.