Neural Network Control of Robot under Wheel Conditions based on Observer

For the complex control issues such as tire slippage, uncertain model parameters, and external disturbances in wheeled mobile robots (WMRs), a novel control approach based on observer-based fuzzy wavelet neural networks (FWNN) is proposed. To address the distortion of angular velocity information caused by tire slippage, mathematical equations for tire slippage and attitude deviation are utilized to design a sliding-mode observer for real-time estimation of angular velocity information. Considering the uncertainties in parameters and unknown model components due to external disturbances, a FWNN is designed to dynamically compensate for these uncertainties using expert knowledge from fuzzy systems and the generalization capability of wavelet neural networks (WNN). To ensure bounded control signals and system stability, a robust controller for the neural network is developed based on Hoc theory, and the asymptotic stability of the entire closed-loop system is proven using Lyapunov theory. Experimental results validate the effectiveness of the proposed control algorithm.