Abstract: For a typical unstable， high-order， multi-variable， strong-coupling， nonlinear seesaw system， an adaptive decoupling sliding mode control (SMC) algorithm based on RBF network(RBF-SMC) is proposed considering the effect of the environment on the seesaw. The decoupling algorithm is used to decouple the model， and the RBF neural network is used to compensate for the interference and uncertainty of the model， which realizes the seesaw balance control under large interference with small switching gain. The algorithm is simulated in Matlab and Matlab/Adams and the results indicate that compared with the existing SMC methods， using the RBF network in an uncertain environment to learn and evaluate the external interference， modeling error， model simplification， external excitation， friction damping and so on， not only improves the anti-interference ability of the system effectively， but also reduces the switching gain of the system and realizes the balance control of the seesaw in a limited time. Through the comparison of simulation results， the validity and practicality of the proposed algorithm are proved.

Key words: seesaw system; sliding mode control (SMC); RBF network; decoupling algorithm; interference compensation