Automated Vehicle Behavior Monitoring for Vulnerability Management
项目名称: Automated Vehicle Behavior Monitoring for Vulnerability Management
摘要: It is clear that autonomous vehicles will penetrate the marketplace in the next few years. It is unclear how prepared these systems will be to withstand cyber attacks that pose a serious threat to the safety of vehicle occupants and other road users. Previous studies have listed ways in which a hacker could cause harm to autonomous vehicle occupants. Other studies have demonstrated the feasibility of attacking vulnerable automotive systems. Safety threats might be mitigated if one could quickly identify attacks, but it is not at all clear that traditional cybersecurity threat detection approaches are well-suited to connected and autonomous vehicles. This project seeks to develop algorithms for identifying when a vehicle has been compromised in a cybersecurity attack, and new approaches to designing and evaluating such techniques. There is virtually no possibility of obtaining historical data of autonomous driving while under cybersecurity attack which is also predictive of future attacks, so reference data must be created. Well-motivated cyber attack models will be defined which characterize attacks on vehicle-internal data used by driver automations. Algorithm development will take place in a cycle of increasing sophistication of both cyber attack models in the form of a synthetic dataset and behavior monitoring algorithms for attack identification. For each iteration, a synthetic dataset of attacks models will be updated in an effort to fool the current algorithm. This cyclic method parallels real world scenarios in which cyber attack models become more sophisticated, resulting in the need for more robust counter methods.
状态: Active
资金: 73229
资助组织: Office of the Assistant Secretary for Research and Technology
管理组织: Safety through Disruption University Transportation Center
项目负责人: Harwood, Leslie C
执行机构: Virginia Tech Transportation Institute
主要研究人员: Gorman, Thomas
开始时间: 20180110
预计完成日期: 20190109
实际结束时间: 0
检索历史
应用推荐