A system of sensors, cameras, computer systems and communication devices installed along Perimeter Road on the southside of Clemson’s campus are tracking pedestrian and vehicular traffic. The system communicates with connected vehicles through a network protected with an innovative cybersecurity blanket.

“Right now, traffic signals can talk to vehicles,” says Mashrur “Ronnie” Chowdhury, director of the U.S. Department of Transportation Center for Connected Multimodal Mobility (C2M2) at Clemson. “A camera sees pedestrians, and we have developed an algorithm so the system can instantly calculate the pedestrian’s safety risk at intersections and, if needed, communicate a warning to prevent an accident.”

Chowdhury sees great opportunity for connected cyberphysical systems to reduce accidents, move freight more efficiently, reduce traffic congestion and even reduce fuel consumption. The Connected and Automated Vehicles Testbed along Perimeter Road provides a platform for testing connected vehicles and related technologies, but Chowdhury is looking beyond the functionality of the car, the sensors, the software or the communication devices.

“We want to develop technologies to protect the system,” Chowdhury says. “If we see that these systems are vulnerable to cyberattacks, then autonomous vehicles will never be mainstreamed. To build public trust, we must make sure systems are secure.”

That requires developing resilient cyberphysical systems that detect and disarm attacks instantly.

Chowdhury, the Eugene Douglas Mays Professor of Transportation in the Glenn Department of Civil Engineering, with joint appointments in computer science and automotive engineering, is bringing together experts from diverse disciplines to answer questions that could support the development of safe, efficient, cost-effective, connected, cyberphysical multimodal transportation systems.

Social and behavioral science experts are helping to understand if people will accept autonomous, connected vehicles. “In the U.S., more than 36,000 people die in roadway crashes each year,” Chowdhury says. More than 90 percent of these, he says, are due to human factors. But incidents involving automated vehicles get a lot of publicity, which influences public opinion even as automated vehicles are reducing accidents overall.

“There are major ethical questions as well,” Chowdhury says. “Let’s say a vehicle needs to swerve to avoid a pedestrian but doing so could seriously injure the occupant of the vehicle. How will such decisions be made when programming vehicles? I believe we need a national standard, a public policy to guide these decisions.”

There is also the question of how connected and automated vehicles will be incorporated into society. “Will this technology be accessible to all people or just people with resources?” Chowdury asks. “These are expensive vehicles. So are we just opening it up to the private market? Will we first deploy these for public transit? What are the economic consequences or impacts of vehicle automation? We need to look at social and economic interests.”

Chowdhury has also assembled a team of social and behavioral scientists, engineers and computer scientists, including collaborators at Virginia Tech, Florida International University and the University of Texas at Dallas, to continue development of secure technologies through a planned establishment of an engineering research center. The team includes experts in information assurance, battlespace coordination, game theory, behavior pattern detection, software and system security, malware program analysis, secure cloud computing, data mining, human-centered computing, social-network analysis, and more.

“Clemson’s autonomous vehicle is an open system, so we can program the robot and see how it operates,” Chowdury says. “We will test how easily we can hack the system, so we can improve the cybersecurity. We can also evaluate sensors to see if we can reduce the cost. We are looking at the software and  communications infrastructure. There is so much opportunity.”