The model-free learning enhanced motion control of DC motor

This paper presents an approach towards learning enhanced motion control of DC motor, suitable for applications involving repeated iterations of motion trajectories. The overall structure of the control consists of a feedback and a feed-forward components. The model-free learning adaptive feedback control (MFLAC) provides for the main system stabilization and an iterative learning control (ILC) algorithm is proposed to serve as a feedforward compensation to nonlinear and unknown dynamics and disturbances, thereby enhancing the improvement achievable with PID or MFLAC alone. It serves as the basis for simulation study of the proposed control scheme. A comparison of the performance achieved with traditional PID and MFLAC is also provided to highlight the advantages of the additional intelligent feedforward mode.