This work explores the application of reinforcement learning (RL) for advanced control of linear actuators in a simulated environment. We present the development of an RL agent using Python libraries to control the position of a linear actuator modelled with a specific dynamic system. The agent interacts with the simulated environment, receiving rewards based on its performance in achieving desired positions. Through continuous learning and exploration, the agent refines its control strategy, surpassing traditional methods in terms of improved accuracy and tuning effort. This approach offers a data-driven solution for complex control problems, particularly beneficial for actuators with non-linearities or uncertainties.