The article is devoted to the process of adaptation of the control system of a mobile robot for its movement along a line. To this end, the article proposes a fuzzy logic and structural model for robot control, combining two methods of adaptation. The first method consists in variation of the parameters of the triangular membership functions and structure of the fuzzy rules, which are used to describe the input and output variables of a fuzzy robot control system. In order to improve the reliability of the solutions proposed in the structure of the fuzzy inference, hard and soft arithmetic operations are used. Evaluation of the system is based on calculation of a standard deviation (RMSE - root mean square error). The best solution for the fuzzy system is the option, in which the RMSE is minimal. The second method envisages a change in the structure of the control code of a mobile robot. The results of the experimental studies presented in the article show that without the use of the methods of the structural and parametric adaptation the accuracy of the soft fuzzy model is 2,7 times higher than the accuracy of a rigid fuzzy model, which has a dead zone, reducing its stability. However, a combination of the above methods makes the accuracy of the rigid model higher than that of the soft model. At that, a deadband is ruled out and computational complexity is reduced. In order to explain the proposed methods of adaptation the article presents the results of the experimental studies evaluating the movement of a mobile robot along a line on an oval track.