This paper propses a robust adaptive controller based on Radial Basis Function Neural Network (RARBFNN) for cleaning and detecting robot manipulators (CDRM) in uncertain dynamical environments. By combining the sliding mode technique with Radial Basis Function neural network and adaptive controller the tracking control was improved. The RARBFNN works well with the defined problems because of its simple structure, faster training update laws and better approximation for the unknown dynamic of CDRM. The adaptive turning rules for the network parameters were derived using the Lyapunov stability theorem. In this control scheme, a robust compensator was constructed as an auxiliary controller to guarantee the stability and robustness under various environments’ conditions such as the mass variation, the external disturbances, and modeling uncertainties. Therefore, the stability, the robustness and the desired tracking performance of RARBFNN for CDRM were guaranteed. Finally, the simulation results in comparison with adaptive Radial Basis Function Neural Network controller are provided to verify the effectiveness of the proposed controller.
Năm 2024 
