This paper presents a cooperative framework of the driver and his/her semi-autonomous vehicle in order to achieve desired steering performance. In particular, a co-pilot controller and the driver operate together and control the vehicle simultaneously. Exploiting the classical small-gain theory, our proposed cooperative steering controller is developed independent of the unmeasurable internal states of human driver, and only relies on his/her steering torque. Furthermore, by adopting data-driven adaptive dynamic programming and an iterative learning scheme, the cooperative controller is learned from measurable data of the driver and the vehicle. Meanwhile, the accurate knowledge of the driver and the vehicle dynamics is unnecessary, which settles the problem of their potential parametric variations in practice. The effectiveness of the proposed method is validated by rigorous analysis and demonstrated by numerical simulations.