Robotic arm trajectory tracking control based on RBF neural network estimation of dynamic parameters

In robotic arm operations, after planning the desired motion trajectory, accurate and stable trajectory tracking control is essential. However, in actual control, there are issues such as the uncertainty of the dynamic parameters and susceptibility to external disturbances during motion. Addressing these issues, this paper employs RBF neural networks to estimate dynamic parameters and utilizes an adaptive controller to achieve online trajectory tracking. The dynamic approximation capability and adaptability of this method enhance the real-time performance and disturbance rejection of robotic arm trajectory tracking control. Finally, the availability of this method is verified by experiments in the simulation environment.