CRII: CHS: Training and Feedback Systems to Improve Vehicle Cybersecurity

University Of Massachusetts Amherst, Amherst MA

Investigators

Abstract

Although modern vehicles have many computing components, they are not designed with cybersecurity in mind. Consequently, they are susceptible to cyberattacks. For example, in 2015 two hackers remotely exploited a vulnerability in an SUV's entertainment system to send the vehicle into a ditch. Though many technological solutions to improve vehicle cybersecurity have been proposed, the PI argues real progress toward achieving that goal will require including the human (driver) in the loop. Thus, this research will combine findings from cybersecurity and transportation safety to develop training and in-vehicle messaging systems to improve drivers' awareness of and responses to cyberattacks, which ultimately should help reduce crash-related costs and potentially save lives. The effectiveness of the new systems will be tested in a driving simulator study. Project outcomes will yield quantitative evidence as to how drivers respond to unintended events that occur in their vehicles. More importantly, the work will create systems that can warn drivers of dangers associated with vehicle hacking, and provide information on how these dangerous situations can be mitigated. Similar systems will be applicable in other transportation safety contexts (e.g., automated driving). This research will proceed in two phases. Phase I will focus on the iterative development of the training and in-vehicle message systems using methods essential to the human-centered design process (e.g., interviews and co-design sessions). Phase II will evaluate the training and new in-vehicle messaging systems though a driving simulator study. During the experiment, drivers will be exposed to various situations that are indicative of a vehicle cyberattack while their driving behavior is recorded. Drivers' responses to the scenarios according to quantitative measures of driving performance along with their subjective evaluation of the systems' effectiveness will be used to determine the extent to which the training and/or in-vehicle messages have a positive impact on driver performance. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.