SMA(Shape Memory Alloy)actuators,a class of smart material actuators with large deformation,high strength and nanometres,have the ability to deform back to its original shape when heated or cooled.SMA actuators were used in many applications such as aerospace,medical surgical instruments and robotics and received a lot of attention in the last decades because of their unique properties.The main advantages of SMA actuators include potentially simplified construction,High strength,Easy fabrication,Easy to sterilize,High Power/weight ratio,Light weight and so on.These attributes potentially allow developing simpler,more reliable and cost-effective implementation actuating systems with a significant reduction in mechanical complexity and size.However,the driving performance of the shape memory alloy actuators(SMAs)is limited due to their intrinsic hysteresis nonlinearities with a local memory,resulting from the influence of a previous input on the subsequent behaviour.To improve the activation accuracy of SMA actuators,it is necessary to characterize the input-output behaviour and design the appropriate controller to maintain system performance.In this thesis,the properties of SMA actuators using as artificial muscle platform are discussed.Addressing on the properties of saturated hysteresis in SMA actuators,a modified Jiles-Atherton model is adopted to model the input-output properties of SMA actuators.The parameters of the modified JilesAtherton model are identified on the basis of the measured data acquired for a pair of SMA actuators in a wide frequency range.The effectiveness of the modified model is testified by comparing the model output with the measured data.