APPLICATION OF SELF-ORGANIZING WIDE LEARNING SYSTEM FOR TWO LINK MANIPULATOR

Abstract

In this paper, the authors present an intelligent control structure for nonlinear systems based on the combination of sliding mode control and a broad learning
system (BLS) with a two-loop feedback design. The main feature of BLS is its self-organizing (SO) capability, which allows the network to adapt its size to the
requirements automatically. The SO method uses two dynamic thresholds to automatically decide whether to keep, remove, or add neurons, optimizing computing
resources and improving operational efficiency. The double feedback loop design significantly improves the network dynamics while ensuring the system's stability
thanks to the Lyapunov function. To verify the effectiveness and stability, the authors simulated the application of the wide learning network to monitor and control
a robot arm with two degrees of freedom on the Matlab-Simulink software platform. The simulation results show that the system ensures stability and performs well
in monitoring control. Based on this result, many potential research directions open up in the field of intelligent control and practical applications.