engleski

Automatic assessment of programming solutions for educational robots Lego Mindstorms EV3

In this paper we present a method for automatic assessment of the correct-ness of a solution for a task solved in the visual programming language for Lego Mindstorms EV3. The first step in the proposed method is to translate the solution into its textual representation using image pattern matching and optical character recognition. The second step is to grade a textual rep- resentation of the solution. The method is tested on 20 students and all so-lutions are correctly translated into their textual presentations. Since some of the solutions given by the students were incorrect, deeper analysis about the grading is given.

template matching, image processing analysis, visual programming language, Lego Mindstorms EV3

nije evidentirano