BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Computer Science - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Computer Science
X-ORIGINAL-URL:https://csc.ncsu.edu
X-WR-CALDESC:Events for Computer Science
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20240310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20241103T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20250309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20251102T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20260308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20261101T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20250922T100000
DTEND;TZID=America/New_York:20250922T110000
DTSTAMP:20260424T082641
CREATED:20250911T140114Z
LAST-MODIFIED:20250911T140331Z
UID:10000028-1758535200-1758538800@csc.ncsu.edu
SUMMARY:Requiem for a Drone: Stealthy Attacks on Autonomous Systems
DESCRIPTION:Speaker\nSibin Mohan\, George Washington University \nAbstract\nAutonomous Systems (AVs) require an accurate sensing and modeling of the real-world in order to carry out their missions. They use the observations from sensors to reason about the vehicle’s state and can correct for deviations\, even attacks. In practice\, even the most meticulously designed control systems always operate under a certain amount of noise because of the unavoidable errors involved in both sensor measurement as well as the modeling of complex vehicular dynamics. \nAll of this inherently creates a space that can be exploited by an adversary. In this talk\, I will present novel methods that can exploit this space\, using a software-only attack. Our system\, Requiem\, presents a blackbox attack -— i.e.\, there is no knowledge required about the internal details of the system —- the only requirement is that the state estimation function be “learnable” from observation of the inputs and outputs. Requiem constructs two sets of deep models (a ‘surrogate’ and a ‘spoofer’) to evaluate the system and generate the attack. The final result of a Requiem-based attack causes significant deviations in the physical system’s trajectory (sometimes by tens or hundreds of metres!). Meanwhile the system believes it is following the original mission parameters. Hence\, this is an attack that is hard to detect or defend against. \nIn this talk\, I will also present some initial ideas on how to detect these stealthy attacks using optical-flow based motion-verification. \nBio\nSibin Mohan is an Associate Professor in the Department of Computer Science at The George Washington University. He also holds an adjunct faculty appointment in the Siebel School of Computing and Data Science at the University of Illinois at Urbana-Champaign. \nSibin completed his Ph.D. and M.S. in Computer Science from North Carolina State University. His undergraduate degree was in Computer Science and Engineering from Bangalore University\, India. In the past\, he worked at Hewlett Packard. His research has won multiple best paper awards and he is the recipient of the NSF CAREER award. \nSibin’s research interests are in the area of systems\, security\, networking and autonomous systems. Sibin has pioneered research to improve the resiliency and security of real-time\, cyber-physical and autonomous systems. Current research efforts include resiliency and security for autonomous and cyber-physical style systems\, securing operating systems via code debloating\, resiliency for safety-critical networks\, security for V2X systems and understanding the behavior of UAV swarms.
URL:https://csc.ncsu.edu/event/requiem-for-a-drone-stealthy-attacks-on-autonomous-systems/
LOCATION:EB2 3211 Seminar Room\, 890 Oval Dr.\, Raleigh\, NC\, 27695\, United States
CATEGORIES:Lecture/Seminar
ORGANIZER;CN="Frank Mueller":MAILTO:fmuelle@ncsu.edu
END:VEVENT
END:VCALENDAR