An adaptive robust controller based on nonlinear observer friction compensation was designed aiming at the problem of obvious friction nonlinearity and other interferences in the electromechanical actuation system at low speed, which may easily lead to the decrease of system tracking accuracy and stability. The LuGre friction model was used to describe the friction phenomenon of the system for the nonlinearity of friction, and a nonlinear observer was proposed to observe the internal friction state of the model. The parameter adaptive law was designed for uncertain parameters such as system friction coefficient and rotational inertia. The friction nonlinearity and parameter uncertainty were compensated by feed-forward compensation, and robust terms were designed to overcome other disturbances in the system. Lyapunov stability theory was used to prove that the proposed controller can achieve the bounded stability of the system in the presence of disturbances. The experimental results show that the proposed controller has higher control accuracy and stronger robustness, which is an order of magnitude higher than the tracking accuracy of traditional PID controller.