Published January 4, 2017
Industry is racing to bring self-driving cars to the masses. Despite much progress, questions remain regarding safety, efficiency, environmental impact, user acceptance, policy and other issues.
To help study these matters, the National Science Foundation has awarded UB and its partners a $1.2 million grant to create a multifaceted research facility for self-driving and connected cars. UB will contribute $500,000 in matching funds toward the project.
“This significant award is a statement of confidence in the innovative work happening right here at the University at Buffalo,” says Rep. Brian Higgins.
The heart of the facility will be a research platform dubbed iCAVE2 — short for Instrument for Connected and Autonomous Vehicle Evaluation and Experimentation. The platform will sync UB’s existing driving, traffic and wireless networking simulators to a gadget-equipped vehicle, and sensors and other instruments to be installed along roads on the North Campus.
“The grant builds upon the University at Buffalo’s robust and interdisciplinary research enterprise that works to make traffic systems smarter, safer and more sustainable,” says Liesl Folks, dean of the School of Engineering and Applied Sciences. “It also gives us additional tools to engage with industry, nonprofits and other organizations.”
The project is led by Chunming Qiao, professor and chair of the Department of Computer Science and Engineering (CSE) at UB. Partner organizations include Carnegie Mellon University, Cisco Systems and Southwest Research Institute.
The iCAVE2 platform will integrate UB’s driving simulators (these devices resemble high-tech arcade games, allowing users to interact with virtual environments), its traffic simulator (to generate virtual environments that mimic driving conditions) and its network simulator (to model wireless networks that cars use to communicate with each other and their surroundings).
Together, the simulators will create virtual, yet realistic, traffic scenarios to evaluate connected and self-driving car technology. For example, they will mimic everything from how people respond to safety messages delivered via their connected car to rush-hour traffic, including both human-driven cars and self-driving cars.
On campus, the research team plans to equip about one mile of North Campus roads with road sensors, wireless access points and other equipment. These instruments will communicate with the connected car and provide real-time data that will help researchers improve the simulators, algorithms and other back-end infrastructure needed to safely operate self-driving cars and improve connected vehicle technology.
“Traditionally, driving simulators and road-testing facilities have operated independently of each other” Qiao explains. “With iCAVE2, we are bridging that gap and creating a space where academia, information technology companies, automakers and other industries can evaluate and validate their products.”
The platform, Qiao says, is designed to answer questions that relate to the safety, efficiency and sustainability of traffic systems where, increasingly, humans allow machines to do the work.
The project will enhance UB’s existing facilities and expertise in transportation studies, including the Institute for Sustainable Transportation and Logistics (ISTL) and the UB-headquartered Transportation Informatics Tier 1 University Transportation Center. Both organizations — led by Adel Sadek, professor of civil, structural and environmental engineering (CSEE) — conduct cutting-edge research, especially as it relates to big data analysis.
Sadek and the following UB researchers are co-principal investigators on grant: Qing He, the Stephen Still Assistant Professor in Transportation Engineering and Logistics, affiliated with both CSEE and the Department of Industrial and Systems Engineering (ISE); Kevin Hulme, senior research associate at UB’s Motion Simulation Laboratory; Dimitrios Koutsonikolas, CSE associate professor; Changxu Wu, ISE associate professor and CSE adjunct associate professor; and Lukasz Ziarek, CSE assistant professor.