Backstepping design, neural network technology, robust adaptive control and terminal sliding-mode control is combined. Firstly, neural networks are used to estimate the virtual control laws of the controlled system. Secondly, terminal sliding-mode control is utilized to improve the convergence speed and robustness of the controlled plant. Thirdly, robust adaptive control method is employed to estimate the unknown upper boundary of uncertainties. Lastly, backstepping design is incorporated to design the virtual control laws and actual control law of the controlled system. Based on Lyapunov stability theorem, all of error signals are bounded and exponentially converge to a bound neighbor of the origin. Simulation results show the effectiveness of the proposed control method.
