A neurodynamics based neuron-PID controller and its application to inverted pendulum

A novel neuron-PID controller that has some excellent characteristics of nonlinear filtering and auto gain-regulation is developed for nonlinear systems in this paper. The biological neuron described by the shunting model is used to construct a nonlinear controller, which is based on the frame of a typical PID controller. The neural activity of the biological neuron model is stable, bounded and smooth so that the output of the neuron-PID controller is bounded and smooth. The proposed controller can be employed to design a class of flexible and safe control systems. The effectiveness and efficiency of the proposed control strategy have been demonstrated by applying it to the stabilization control of an inverted pendulum with uncertain dynamics. The simulations show that the dynamic responses of the control system can be effectively improved and the robustness of the proposed controller is better than that of the PID controller.