Using hands-on project with Lego Mindstorms in a graduate course

The aim of this paper is to propose an educational hands-on project using inexpensive subsystems for learning guidance and control. A Lego Mindstorms NXT, a low-cost web camera, and tractable tools are used for searching for and mapping of an obstacle in an indoor environment. In order to provide the Lego robot with navigation information indoors, visual tracking is implemented by using color marker detection and an extended Kalman filter. Furthermore, spiral-like search, command-to-line-of-sight guidance, and motor control are applied to sensing and mapping of an unknown obstacle. The experimental results from a short-term group project verify that the proposed hands-on work is an efficient educational tool for learning vision processing and estimation as well as guidance and control with a low-level burden of time and cost.