Dynamic dandelion optimization-based Model Predictive Control for accurate tracking of a 6-dof robot manipulator

This paper proposes an optimized control strategies for a dynamics 6-link robot manipulator using Model Predictive Control (MPC) and traditional Computed Torque Control (CTC). Two optimization methods are used for tuning of MPC controller, namely MPCbased Dandelion Optimization (DO) and MPCbased Genetic Algorithm (GA). The strategies are assessed in terms of performance indices based timedomain, settling time, rise time, overshoot percentage, and steady-state error (SSE). Simulation results depicted that traditional CTC suffers from higher control effort and slower transient response, mainly for links with high order. In contrast, MPC-based control strategy significantly improve dynamic performance and tracking accuracy. Among the tested approaches, MPCbased DO achieves minimal overshoot, smoother torque profiles, fastest convergence, and the lowest SSE across all six robotics links. The attained MATLAB/SIMULINK results confirm and prove that optimized MPC controller, especially MPCbased DO, offers an efficient and robust control solution for high DOF robotic manipulators.