US 11,989,625 B2
Method and system for detecting and avoiding loss of separation between vehicles and updating the same
Megan Hawley, Roseville, MN (US); and Raj Mohan Bharadwaj, Maple Grove, MN (US)
Assigned to HONEYWELL INTERNATIONAL INC., Charlotte, NC (US)
Filed by Honeywell International Inc., Morris Plains, NJ (US)
Filed on Mar. 29, 2019, as Appl. No. 16/370,141.
Prior Publication US 2020/0311602 A1, Oct. 1, 2020
Int. Cl. G06Q 30/00 (2023.01); G06N 20/00 (2019.01); G08G 5/00 (2006.01)
CPC G06N 20/00 (2019.01) [G08G 5/0095 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for updating a vehicle interaction machine learning model, the method comprising:
obtaining trajectory information indicating trajectories of a plurality of vehicles over a set period of time;
analyzing the trajectory information to determine whether a trajectory, among the trajectories within the set period of time, is a new trajectory type;
analyzing the trajectory information to determine whether the trajectory is a member of a new interaction, wherein an interaction is a trajectory, among the trajectories within the set period of time, of a first vehicle of the plurality of vehicles within a threshold distance of a trajectory, among the trajectories within the set period of time, of a second vehicle of the plurality of vehicles, and wherein the interaction includes combinations of vehicles that were not operating at the same time, but that would have been within a threshold distance of each other if the respective vehicles had been operating at the same time;
in response to determining the trajectory is a new trajectory type or the trajectory is a member of a new interaction, updating training data for the vehicle interaction machine learning model with the new trajectory type or the new interaction; and
training the vehicle interaction machine learning model based on the updated training data including the new trajectory type or the new interaction.