US 12,265,762 B1
Human machine teaching system for sequential task training
Myana Anderson, Minneapolis, MN (US); Tiffany Hwu, San Diego, CA (US); Steven W. Skorheim, North Hills, CA (US); and David J. Huber, Calabasas, CA (US)
Assigned to HRL LABORATORIES, LLC, Malibu, CA (US)
Filed by HRL Laboratories, LLC, Malibu, CA (US)
Filed on Aug. 24, 2021, as Appl. No. 17/411,053.
Claims priority of provisional application 63/118,263, filed on Nov. 25, 2020.
Int. Cl. G06F 30/27 (2020.01); G05B 13/02 (2006.01); G05B 13/04 (2006.01); G06F 30/12 (2020.01); G06N 5/02 (2023.01); G06F 119/02 (2020.01)
CPC G06F 30/27 (2020.01) [G05B 13/0265 (2013.01); G05B 13/041 (2013.01); G06F 30/12 (2020.01); G06N 5/02 (2013.01); G06F 2119/02 (2020.01)] 14 Claims
OG exemplary drawing
 
1. A system for improving machine operation performance, the system comprising:
an interface having a plurality of interactive controls; and
one or more processors and a non-transitory computer-readable medium having executable instructions encoded thereon such that when executed, the one or more processors perform an operation of:
assigning and displaying, on the interface, a performance score for each skill of a sequential task in a simulation of operation of a machine;
based on the performance scores, identifying and displaying, on the interface, one or more skills to improve with targeted training;
based on the one or more identified skills, recommending a training scenario comprising one or more skills to be performed via at least one of the plurality of interactive controls in a subsequent simulation to improve the performance scores;
following performance of the training scenario in the subsequent simulation, assigning and displaying, on the interface, a new performance score for each skill performed;
adapting the training scenario based on the new performance scores;
using k-means clustering to determine boundaries of phases of the sequential task;
obtaining a time sequence of segments based on the determined boundaries; and
determining a set of metric functions, each metric function associated with a segment position on a segment.