Neural Network Sliding Mode Control of Robotic Manipulator Based on Exponential-Trigonometric Optimization Algorithm

The sliding mode control based on the radial basis function (RBF) neural network (NNSM) can effectively address model uncertainties in the motion control of robotic manipulators. However, such controllers exhibit high parameter sensitivity, which poses significant challenges in parameter adjustment during implementation. To overcome this problem, an NNSM controller parameter adjustment method based on the Exponential-Trigonometric Optimization (ETO) algorithm is proposed. First, the RBF neural network approximates the model uncertainties of rigid robotic manipulators, while a robust term compensates for the effects of time-varying disturbances on system tracking performance. Then ETO algorithm optimizes the controller parameters by minimizing a designed fitness function. The simulation results show that under model uncertainty and bounded external disturbances, the optimized NNSM controller achieves a higher control accuracy for the reference trajectory.