Cloud-model PID control of double inverted pendulum based on information fusion

Abstract

Aimed at the characteristics of nonlinear, strong coupling, multi-variable and natural unstable for inverted pendulum, based on information fusion technology, the state feedback gain matrix with quadratic optimal performance index is adopted as fusion function to reduce dimension of the outputs vector for double inverted pendulum, and decrease the difficulty of controller design. On this basis, a cloud model based PID controller is designed by combining clouds model with conventional PID and applied to the stabilization control of double inverted pendulum, the cloud inference mode is used to adjust PID control parameters adaptively. The simulation results show that the control method has excellent dynamic performance and strong robustness, and reflect the superiority of the cloud model as an uncertain conversion between qualitative concept and quantitative description in dealing with uncertainty problems.