Granted Patent B2

US 12,607,743 · App. 17/632,747 · Granted Apr 21, 2026

Inventor: Kyoji Yokoyama (Kanagawa, JP)

Assignee: SONY SEMICONDUCTOR SOLUTIONS CORPORATION

G01S17/08G01S17/86G06N3/082G06V20/35G06V10/82

Abstract

Claims (72)

1 . A device, comprising:

a light source configured to emit light toward a measurement object;

a light receiving element configured to:

receive the light reflected from the measurement object; and

output an output signal based on the reception of the reflected light; and

a central processing unit (CPU) configured to:

convert the output signal into time information, wherein the time information indicates a time range corresponding to a time from the emission of the light by the light source to a time of the reception of the light by the light receiving element;

acquire first distribution of the time information;

change, based on the acquired first distribution of the time information, values of a plurality of parameters of the light source, wherein

the plurality of parameters includes at least one of a direction of irradiation by the light source, power of the irradiation by the light source, or a pulse shape of the irradiation by the light source;

acquire second distribution of the time information based on the changed values of the plurality of parameters of the light source;

select, based on the acquired second distribution of the time information, a first machine learning model from a plurality of machine learning models, wherein

the selected first machine learning model recognizes specific information corresponding to a distance to the measurement object, and

the specific information is recognized based on the time information;

acquire a sensing result of a subject space in which the measurement object exists; and

switch, based on the sensing result, from the first machine learning model to a second machine learning model of the plurality of machine learning models.

2 . The device according to claim 1 , wherein

the CPU is further configured to change, based on the acquired first distribution of the time information, a value of a parameter associated with the light receiving element, and

the parameter comprises sensitivity of the light receiving element.

3 . The device according to claim 1 , wherein

the sensing result includes first information corresponding to a type of the measurement object.

4 . The device according to claim 3 , wherein

the sensing result further includes second information corresponding to a scene in the subject space.

5 . The device according to claim 4 , wherein

the second information includes at least one of information of a weather in the subject space, information of a time associated with the subject space, or information of a place of the subject space, and

the CPU is further configured to switch from the first machine learning model to the second machine learning model based on the second information.

6 . The device according to claim 5 , wherein the CPU is further configured to determine a distance measuring point in the subject space based on the sensing result.

7 . A measuring device, comprising:

a light source configured to emit light toward a measurement object;

a light receiving element configured to:

receive the light reflected from the measurement object; and

output an output signal based on the reception of the reflected light; and

a central processing unit (CPU) configured to:

convert the output signal into time information, wherein the time information indicates a time range corresponding to a time from the emission of the light by the light source to a time of the reception of the light by the light receiving element;

acquire first distribution of the time information;

change, based on the acquired first distribution of the time information, values of a plurality of parameters associated with the light source, wherein

the plurality of parameters includes at least one of a direction of irradiation by the light source, power of the irradiation by the light source, or a pulse shape of the irradiation by the light source;

acquire second distribution of the time information based on the changed values of the plurality of parameters of the light source;

select, based on the acquired second distribution of the time information, a first machine learning model from a plurality of machine learning models;

recognize, based on the selected first machine learning model, specific information corresponding to a distance to the measurement object, wherein the recognition of the specific information corresponding to the distance to the measurement object is based on the time information;

acquire a sensing result of a subject space in which the measurement object exists; and

switch, based on the sensing result, from the selected first machine learning model to a second machine learning model of the plurality of machine learning models.

8 . A distance measuring system, comprising:

a light source configured to emit light toward a measurement object, wherein the measurement object exists in a subject space;

a light receiving element configured to:

receive the light reflected from the measurement object; and

output an output signal based on the reception of the reflected light;

a plurality of sensors configured to sense the subject space; and

a central processing unit (CPU) configured to:

convert the output signal into time information, wherein the time information indicates a time range corresponding to a time from the emission of the light by the light source to a time of the reception of the light by the light receiving element;

acquire first distribution of the time information;

change, based on the acquired first distribution of the time information, values of a plurality of parameters of the light source, wherein

the plurality of parameters includes at least one of a direction of irradiation by the light source, power of the irradiation by the light source, or a pulse shape of the irradiation by the light source;

acquire second distribution of the time information based on the changed values of the plurality of parameters of the light source;

select, based on the acquired second distribution of the time information, a first machine learning model from a plurality of machine learning models; and

recognize, based on the selected first machine learning model, information corresponding to a distance to the measurement object, wherein the recognition of information corresponding to the distance to the measurement object is based on the time information;

acquire, from at least one sensor of the plurality of sensors, a sensing result of the subject space in which the measurement object exists; and

switch, based on the sensing result, from the first machine learning model to a second machine learning model of the plurality of machine learning models.

9 . A method, comprising:

emitting, by a light source, light toward a measurement object;

receiving, by a light receiving element, the light reflected from the measurement object;

outputting, by the light receiving element, an output signal based on the reception of the reflected light;

converting the output signal into time information, wherein the time information indicates a time range corresponding to a time from the emission of the light by the light source to a time of the reception of the light by the light receiving element;

acquiring first distribution of the time information;

changing, based on the acquired first distribution of the time information, values of a plurality of parameters associated with the light source, wherein

the plurality of parameters includes at least one of a direction of irradiation by the light source, power of the irradiation by the light source, or a pulse shape of the irradiation by the light source;

acquiring second distribution of the time information based on the changed values of the plurality of parameters of the light source;

selecting, based on the acquired second distribution of the time information, a first machine learning model from a plurality of machine learning models, wherein

the selected first machine learning model recognizes specific information corresponding to a distance to the measurement object, and

the specific information is recognized based on the time information;

acquiring a sensing result of a subject space in which the measurement object exists; and

switching, based on the sensing result, from the first machine learning model to a second machine learning model of the plurality of machine learning models.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (22)