An Effective Intelligent Control Strategy for the Inverted Pendulum System on a Cart Applies a Hierarchical Sliding Mode Control Technique Combined with an Adaptive RBF Neural Network

In this paper, an effective intelligent control method, constructed based on the connection between the Hierarchical Sliding Mode Control (HSMC) technique and the adaptive radial basis function neural network (ARBF), is applied to the inverted pendulum system on a cart. Developed control strategy exploited the strengths of HSMC in controlling underactuated robotic systems. The versatility of the scheme, as well as its adaption which is established when combining proposed HSMC technique and the ARBF neural network is enhanced. Moreover, the proposed ARBF is particularly designed with automatical updating laws for the centroid and standard deviation in the activation function online. This improvement brings the integrated ARBF a ability to flexibly weigh changing. The stability of the closed-loop system has been rigorously proven based on the Lyapunov theory. All the simulation results are implemented in the Matlab/Simulink platform to demonstrate that the inverted pendulum system on a cart can operate accurately and sustainably, even when affected by external noises and uncertainties in model parameters.