A Natural Language Programming Application for Lego Mindstorms EV3

Abstract

In this paper, a controlled natural language (CNL) based program synthesis system for the Lego Mindstorms EV3 (EV3) is introduced. The system is developed with the intention of helping middle and high school Lego robotics enthusiasts and non-programmers to learn the necessary skills for programming and engineering the robot with less effort. The system generates the resulting code in Microsoft Small Basic that controls the EV3 Intelligent Brick with supports for all EV3 sensors and motors. Preliminary results show that our approach is capable of generating functional, executable code based on the users' controlled natural language specifications. Detailed error messages are also given when confronted with unimplementable sentences.