US 12,189,781 B1
Artificial intelligence-based fuzzing of cyber-physical systems
Matthew Osamu Williams, Honolulu, HI (US); and David Siu, Honolulu, HI (US)
Assigned to Oceanit Laboratories, Inc., Honolulu, HI (US)
Filed by Oceanit Laboratories, Inc., Honolulu, HI (US)
Filed on Sep. 21, 2021, as Appl. No. 17/481,275.
Claims priority of provisional application 63/082,703, filed on Sep. 24, 2020.
Int. Cl. G06F 21/57 (2013.01); G06F 11/34 (2006.01); G06F 16/901 (2019.01); G06N 5/01 (2023.01)
CPC G06F 21/577 (2013.01) [G06F 11/3476 (2013.01); G06F 16/9024 (2019.01); G06N 5/01 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A method for detecting vulnerabilities in a cyber-physical system, the method comprising:
defining, by at least one processor, a default state of a cyber-physical system, wherein the default state comprises a plurality of messages, and wherein the default state represents expected functioning of the cyber-physical system;
identifying, by the at least one processor, one or more dependency relationships between one or more of the plurality of messages;
generating, by the at least one processor, an expected response message to an incoming message sent by the cyber-physical system based on the one or more dependency relationships;
fuzzing, by the at least one processor, the expected response message and transmitting the expected response message to the cyber-physical system; and
recording, by the at least one processor, any response from the cyber-physical system to the fuzzed expected response message.