In this study, three different nature-based algorithms were used to find the best combinations of the prediction horizon and control horizon coefficients of the model-predictive-controller for a two-degree-of-freedom ball and plate system. These algorithms are Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Bacterial Foraging Optimization (BFO). During the calculations, the factors are treated as two-dimensional in terms of numerical values of the prediction horizon and the control horizon and these values are called variable pairs. Each pair of variables imposed a cost according to the system outputs in line with various control conditions such as percentage overshoot, settling time, etc. These operations were performed on MATLAB Simulink and MATLAB. In addition to the effect of the variable values contained in the algorithms themselves, the results of the algorithms compared to each other were also revealed within the scope of the study. All of the algorithms were able to find the controller pair values with the lowest cost, however, there was a difference between them in terms of generation/iteration at which they found this point.