On May 5th, the Department gathered to celebrate the opening of the new Autonomous Systems Laboratory. The new lab, housed in Communications 1110, is an important resource, equipped with hands-on learning opportunities, including a smart-city, a driving simulator, and a net for drone flying.
The lab was generously funded in part by LAS, serving as a physical reminder of the college's dedication to providing industry-leading educational resources to its students. LAS Dean Schmittmann and all who attended got to see lab demonstrations from the faculty and students using the lab.
Hridesh Rajan, Professor and Chair of the Department of Computer Science said of the lab, "This project is an important step in furthering the Department’s commitment to providing strong research programs needed to address the grand challenges of tomorrow. Autonomous Systems is a field with a bright future ahead, and this lab will facilitate research, educational and outreach projects to help meet the demands of the moment,"
Last fall, the Department launched the state's first Master's program in Artificial Intelligence, a program that will greatly benefit from this lab. The hands-on education will strengthen the rigorous coursework in AI, machine learning, robotics, and autonomous systems. In particular, the smart city will be utilized in COMS 476 (Motion Strategy Algorithms and Applications) to provide undergraduate students a hands-on experience in implementing algorithms taught in the course on a physical system, reinforce the core concepts, and demonstrate challenges related to the complex interaction between the computational, physical, and communication components in most robotics and autonomous systems.
On the research side, the lab provides the necessary platforms to study various aspects of autonomous and robotics systems. This will start at the component level (e.g., perception, localization, planning, and control) and continue to system-level design, development, and analysis of autonomous systems to interactions between multiple autonomous systems and between autonomous systems and humans. This new facility will allow the PIs in Computer Science to lead competitive proposals to the NSF Cyberphysical System program, NSF National Robotics Initiative, etc.
The smart city features a modular, inexpensive research platform for studying autonomy in complex systems. Classic control algorithms do not translate well between tasks since most of them have to be tuned to specific driving conditions such as lighting, road type, camera position, etc. To overcome this, the smart city offers a dataset compiled from different sources to offer more robustness to control models that can perform in diverse environments. The driving simulator provides a complimentary setup for studying the interactions between humans and autonomous systems. Its design features allow for the attachment of wheels, yokes, Joysticks, keyboards, mice, LCD screens, etc. Having a driving simulator will allow the student to become the “driver” of an autonomous racing simulator and learn expertise in robust perception, planning, and control at high speeds, overtaking algorithms, mapping, localization, head-to-head racing, and end-to-end autonomous driving.
"I recognize that none of this would be possible without the Principal Investigators of this project, Nok Wongpiromsarn and Ali Jannesari as well as the contributions from LAS. Their diligence and dedication are what made this idea a reality," Rajan said.