Adaptive Neural Network Control for Load-Varying Two-Link Robots Using Honey Badger Optimization

This paper illustrates a proportional-integral-derivative based neural network (PID-NN) controller to manipulate the angular position of the two-link robot considering the load variation on the system. The two-link robot system's dynamic equations were derived using the Lagrange method. To improve the tuning process of the design coefficients of the controller, the learning process was framed as an optimization task. Subsequently, to determine the optimal weight values, the honey badger algorithm (HBA) was introduced. To analyze how well the proposed controller performs, Simulations in MATLAB were carried out to compare the PID-NN controller against a PI-PD controller. The findings revealed superior performance of the PID-NN controller in standard conditions. Furthermore, the PID-NN demonstrated a substantial enhancement when a load variation was augmented.