Granted Patent B2

US 12,645,758 · App. 17/897,409 · Granted Jun 2, 2026

Inventors: Makihiko Ishitani (Kawasaki, JP); Shinji Takakura (Yokohama, JP); Yoshiyuki Ishihara (Kawasaki, JP)

Assignee: KABUSHIKI KAISHA TOSHIBA

G06F18/217G06F17/15

Abstract

Claims (22)

1 . A learning control device, comprising:

processing circuitry including one or more hardware processors and configured to:

update a modified control input in accordance with a tracking error, the modified control input being used during a learning trial;

calculate, in accordance with a state of a control target at a start of a learning control, a lag between a time when the state of the control target satisfies a learning start condition and a learning control start time when the learning control actually starts;

correct the modified control input having been updated, the modified control input being corrected by using the lag to have a value obtained by offsetting the lag; and

control the control target based on the modified control input.

2 . The learning control device according to claim 1 , wherein the processing circuitry is further configured to:

extract, using a low-pass filter, a low frequency component included in the updated modified control input;

extract, using a high-pass filter, a high frequency component included in the updated modified control input;

perform linear interpolation on an output of the low-pass filter by using the lag; and

output a result of adding up an output of the high-pass filter and an output of the linear interpolation as a corrected modified control input.

3 . The learning control device according to claim 1 , the processing circuitry is further configured to output, to the control target, an input control signal obtained by adding up a feedback signal to cause the state of the control target to track a target value and the modified control input having been corrected by the processing circuitry.

4 . A learning control method, comprising:

updating a modified control input in accordance with a tracking error, the modified control input being used during a learning trial;

calculating, in accordance with a state of a control target at a start of a learning control, a lag between a time when the state of the control target satisfies a learning start condition and a learning control start time when the learning control actually starts;

correcting the modified control input having been updated, the modified control input being corrected by using the lag to have a value obtained by offsetting the lag; and

controlling the control target based on the modified control input.

5 . A computer program product comprising a non-transitory computer-readable recording medium on which a program executable by a computer is recorded, the program instructing the computer to:

update a modified control input in accordance with a tracking error, the modified control input being used during a learning trial;

calculate, in accordance with a state of a control target at a start of the learning control, a lag between a time when the state of the control target satisfies a learning start condition and a learning control start time when the learning control actually starts;

correct the modified control input having been updated, the modified control input being corrected by using the lag to have a value obtained by offsetting the lag; and

control the control target based on the modified control input.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (19)