Enhancing the robustness, resiliency and security of autonomous vehicles.
My current research combines a digital twin framework
with adversarial search methods and rigorous software
engineering techniques in order to reveal and mitigate
"unlikely-but-possible" situations (a.k.a. corner cases)
that might arise due to unwanted feature interactions,
unanticipated environmental conditions,
human error, and novel attack vectors.
Our current experimental platform, shown below, is
a 1:5-scale autonomous vehicle with a top speed
of 60 mph.
Relevant Topics: Evolutionary robotics, high-assurance
software, autonomous systems, cybersecurity, digital twins,
learning-enabled system components,
software engineering, Robot Operating System (ROS),
self-adaptive systems, autonomic computing, cyber-physical systems, assurance cases, artificial life.
Current/Recent Research Projects: MAPE-SAC (AFRL):
Uncertainty-Aware Certification of Assurance for Autonomous Systems
Mitigating Uncertainty in High-Assurance Software Systems
Evolution Park (NSF): An Evolutionary Robotics
Habitat for the Study of Crawling, Swimming and Flying Creatures
BEACON (NSF): An NSF Center for the Study of Evolution in Action
TEAMS (NSF): Transplanting Artificial Life Behaviors to
Orchid (NSF): Harnessing Digital Evolution to Design High-Assurance Adaptive Systems
AWARE (ARO): Adaptive Software Monitoring for Critical