Achieving Educational and Research Goals with Small, Low-Cost Robot Platforms

R ecently, there has been much interest in achieving educational [1]-[6] and research [7] objectives through the use of small, low-cost, configurable mobile robot kits. One such article [1] argues that robotics provides students with needed experience dealing with integrated systems building, real-world issues, teamwork, multidiscipline information, and critical thinking. Recent excitement with the initial success of robot-based courses has led to ambitious plans incorporating robotics into computer science (CS) curriculums [2]. Universities and high schools are employing these kits in artificial intelligence (AI), CS, engineering, and physics courses where they provide students with a new perspective on building integrated systems, allowing hands-on education and experimentation at low cost. While our initial experiences with these platforms were similarly positive, we questioned whether these platforms could be pushed beyond their early uses and transitioned towards achieving substantial educational and research goals. This article reports initial results of this investigation—the construction and implementation of a series of detailed lab assignments using these platforms to tackle basic CS, AI, robotics, and engineering problems. These labs focus on dead reckoning, understanding sensors, real-time scheduling, and sensor sampling as well as feature detection and map building with sonar.We additionally report on our efforts to supplement these lab assignments with research and toolbuilding projects. We assembled mobile robot kits similar to those used elsewhere, centering on the Handy Board microcontroller board [8], LEGO construction pieces, and sensors built with parts from various vendors. The microcontrollers are programmed in the Interactive C (IC) programming language. A detailed parts list is available from our Web site at http:// itcsl.cs.drexel.edu.A second mobile robot kit option,based on the LEGO Mindstorms RCX microcontroller, is discussed later in this article. Our primary goal was to see how far we could push the limits of this technology. In particular, we designed labs with a variety of sensors (e.g., servo-mounted sonar,wheel encoders), complex programming and CS concepts (e.g., multitasking, scheduling, resource limitations), and potential for careful experimentation and analysis. Additionally, we investigated the potential of using these platforms to teach more advanced AI and robotics concepts, such as Bayesian representation and reasoning, map-based path planning, resource-bounded reasoning, and real-time control. Finally, we developed research projects to use these platforms to explore infrared (IR) communications networks, teleoperation, and real-time control of robots over public networks. In this article, we first provide detailed descriptions of the labs we have developed and then discuss the robot platforms, including the progression of hardware issues encountered. Finally, we share what we have learned from this endeavor. We believe our efforts and experiences can be of benefit to a wide range of CS and AI educators and researchers in assessing the potential of achieving educational and research goals with these platforms.