In this article, robust non-linear control design to an ionic polymer metal composite (IPMC) with hysteresis, uncertainties and input constraints is studied. The IPMC is a novel smart polymer material, and many potential applications for low mass high displacement actuators in biomedical and robotic systems have been shown. In general, the IPMC has highly non-linear property and hysteretic behaviour, and the control input is subject to some constraints to ensure safety and longer service life of IPMC. Moreover, there exist uncertainties caused by identifying some physical parameters and approximate calculation in dynamic model. As a result, considering measurement error of parameters and model error, a practical non-linear model is obtained, and a non-linear robust control design with hysteresis, uncertainties and input constraints using operator-based robust right coprime factorisation is proposed for an IPMC setup, where, the Prandtl-Ishlinskii (PI) model is used to describe the hysteresis, and which is identified using experimental data. The effectiveness of the proposed control method based on obtained non-linear model is confirmed by simulation and experimental results.