Conducting polymer actuators (CPAs) are promising candidates for replacing conventional actuators due to their advantageous properties such as low cost, low weight, small actuation voltage and biocompatibility. One of the obstacles for these actuators to become widespread in real world applications is the difficulty of controlling their position or force output as these actuators represent time varying and nonlinear dynamics due to various effects such as synthesis process and conditions, changes in ambient conditions, etc. Linear models are of limited use to design controllers for them since the performance of these controllers may not be sufficient due to model mismatches. Especially, due to their time varying behaviour, a pre-designed controller based on an identified model may show performance deterioration in time. In this study, a model-free control framework is proposed to control the tip displacement of a trilayer conducting polymer actuator with polypyrrole electrodes. The proposed control strategy eliminates the requirement of identification of the dynamics of CPAs for each application and is based on fast derivative estimation of noisy signals. Another advantage of the method is that it is very simple to design and implement. Experimental results are obtained for the model free control method and compared with those of classical PI control.