Robust Control for Robotic Manipulators Base on Adaptive Neural Network

Abstract

As robotic manipulators are increasingly applied in industrial production, higher precision control methods are being studied by researchers. But robotic manipulators are a coupled system with a lot of uncertainties; higher precision is difficult to obtain by traditional control methods. A novel adaptive robust control method based on neural network is proposed by the paper. Neural network controller has been designed for adaptive learning and compensate for the unknown system and approach errors as disturbance is eliminated by robust controller. The weight adaptive laws on-line based on Lyapunov theory are designed. Robust controller is proposed based on H theory. These can assure the stability of the whole system, and L 2 gain also is less than the index value. Simulation studies show that the proposed control strategy is able to achieve higher control precision and has important engineering applications value.