Wakefulness determination method

Patent metadata

Field	Value
Publication number	US-10470697-B2
Application number	US-201615770622-A
Country	US
Kind code	B2
Filing date	Oct 28, 2016
Priority date	Oct 30, 2015
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

The present disclosure provides a wakefulness determination method for accurately determining wakefulness. The wakefulness determination method uses a respiration sensor that obtains respiratory data about respiration of a seated occupant, a calculation unit that calculates the respiratory data obtained from the respiration sensor, and a controller including a determination unit that determines a state of the seated occupant. The wakefulness determination method includes: obtaining, by the respiration sensor, respiratory data of the seated occupant; calculating, by the calculation unit, a degree of change in respiration from the obtained respiratory data; and determining, by the determination unit, wakefulness of the seated occupant by using a Bayesian filter where a probability of occurrence of drowsiness in the seated occupant for the degree of change in respiration is taken as a likelihood and the likelihood is multiplied by a prior probability of occurrence of drowsiness.

First claim

The invention claimed is:

1. A wakefulness determination method using a respiration sensor that obtains respiratory data about respiration of a person, a calculation unit that calculates the respiratory data obtained from the respiration sensor, and a controller including a determination unit that determines a state of the person, the wakefulness determination method comprising:
obtaining, by the respiration sensor, respiratory data of the person;
calculating, by the calculation unit, a degree of change in respiration, which is a degree of change in the obtained respiratory data per unit time; and
determining, by the determination unit, wakefulness of the person by using a Bayesian filter where a probability of occurrence of drowsiness in the person for the degree of change in respiration is taken as a likelihood and the likelihood is multiplied by a prior probability of occurrence of drowsiness.

2. The wakefulness determination method according to claim 1 , wherein the method uses an obtaining unit that obtains a wakefulness signal, which is associated with the wakefulness of the person, in accordance with operation by the person, and a storage unit that stores the wakefulness signal sent from the obtaining unit and various conditions at the time of the operation;
wherein calculating the degree of change in respiration comprises setting the likelihood or the prior probability on the basis of data stored in the storage unit.

3. The wakefulness determination method according to claim 1 , wherein calculating the degree of change in respiration comprises calculating a pattern associated with a number of times that the degree of change in respiration has decreased in a predetermined time period, and
the likelihood includes a probability of occurrence of drowsiness in the person for the pattern.

4. The wakefulness determination method according to claim 1 , wherein calculating the degree of change in respiration comprises calculating a respiration interval (RI), which is an interval for one respiration of the person, from the respiratory data obtained from the respiration sensor, and
the likelihood includes a probability of occurrence of drowsiness in the person for the RI.

5. The wakefulness determination method according to claim 1 , wherein the controller includes a waveform generation unit that generates respiration interval (RI) waveform data for a respiratory waveform from the respiratory data obtained from the respiration sensor, the RI waveform data indicating transitions in a predetermined time period of a respiration interval (RI) that is an interval for one respiration of the person,
in a case where a Respiration root Mean Square Successive Difference (RrMSSD) is defined by

RrMSSD
=

{

RI
⁡

(
t
)

-

RI
⁡

(

t
-
1

)

}

2

+
⋯
+

{

RI
⁡

(

t
+
n
-
1

)

-

RI
⁡

(

t
+
n
-
2

)

}

2

n

where the t-th RI in the RI waveform data generated by the waveform generation unit is determined as RI (t) and the n-th RI after the t-th RI in the generated RI waveform data is defined as RI (t+n), calculating the degree of change in respiration comprises calculating the RrMSSD from the RI waveform data, and
the likelihood includes a probability of occurrence of drowsiness in the person for the RrMSSD.

6. The wakefulness determination method according to claim 1 , wherein the likelihood includes five or six pieces of data.

Assignees

• Ts Tech Co Ltd

Inventors

• Sugiyama Shinji

Classifications

• B60W2540/221

  Physiology, e.g. weight, heartbeat, health or special needs · CPC title

• B60N2002/981

  Warning systems, e.g. the seat or seat parts vibrates to warn the passenger when facing a danger · CPC title

• A61B5/725

  using specific filters therefor, e.g. Kalman or adaptive filters (specific diagnostics methods using using bioelectric or biomagnetic signals A61B5/316) · CPC title

• A61B5/113

  occurring during breathing · CPC title

• B60W2050/0029

  Mathematical model of the driver · CPC title