Enabling Graduate Engineering Students With Proficiency in Mobile Robotics

Abstract

In recent years, an aggressive expansion of research as well as commercialization efforts in autonomous vehicles can be witnessed. At the same time, many existing companies have expanded their portfolio to autonomous technologies as well (e.g. NVIDIA). This has created an already large need for autonomous-vehicle engineers who are not only proficient in single traditional engineering fields (e.g. mechanical) and old-school automotive studies, but who also have acquired the significantly different, interdisciplinary skillset for mobile robotics. Unlike students of computer science, mechanical engineering graduate students are hardly exposed to coding and robotic system integration in current traditional curricula. The new demands of the automotive industry require an automotive engineer who understands the science of autonomy as well as its impact on the design and implementation of autonomous vehicles, and is equipped with hands-on experience with the latest technology in the field.We describe a unique education program that draws content from traditional courses on mobile-robotics as well as incorporates experiential learning by hands-on training in software, specifically addressing the skill gap in traditional automotive engineering education. Geared towards engineering students with no previous training in robotic system integration, and with only basic undergraduate understanding of programming languages, the teaching experiment employed an active learning approach to introduce numerous concepts as a host of hands-on exercises on multiple robotic platforms. Beginning with simple tutorials on networked communication to demonstrate the power of ROS, the course built up to complete control system design on a student-built RC car that can avoid obstacles and navigate a racecourse by performing SLAM.A brief evaluation of the course exhibited good student performance in general with unique and creative approaches to the programming tasks in particular. Although employing different approaches, each student team was able to demonstrate comparable, efficient performance.