US 11,914,599 B2
Machine learning intermittent data dropout mitigation
Kirk A. Lillestolen, East Hartford, CT (US); Kanwalpreet Reen, Ellington, CT (US); and Richard A. Poisson, Avon, CT (US)
Assigned to Hamilton Sundstrand Corporation, Charlotte, NC (US)
Filed by Hamilton Sundstrand Corporation, Charlotte, NC (US)
Filed on Nov. 19, 2021, as Appl. No. 17/455,899.
Prior Publication US 2023/0161773 A1, May 25, 2023
Int. Cl. G06F 16/2455 (2019.01)
CPC G06F 16/24568 (2019.01) 19 Claims
OG exemplary drawing
 
1. A method for mitigating data-stream dropout in a serial data-stream, the method comprising:
receiving, as input, a time-sequence of messages of the serial data-stream, each of the time-sequence of messages containing a data packet communicating an operation to control an aircraft;
determining validity of each of the time-sequence of messages received, thereby determining valid messages and invalid messages of the time-sequence of messages received;
predicting, after receiving each valid message, a plurality of sequential future operations to control the aircraft, the plurality of sequential future operations to control the aircraft predicted based at least in part on the valid message received, wherein the plurality of sequential future actions predicted corresponds to a plurality of future messages expected to immediately follow the valid message received in the serial data-stream;
performing, after receiving each valid message, the operation to control the aircraft communicated by the data packet contained in the valid message received;
storing, after receiving each valid message, the plurality of sequential future operations to control the aircraft; and
performing, after receiving each invalid message, a next one of the plurality of sequential future operations to control the aircraft stored after a last valid message was received.