Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 03164768 2022-06-15
SMART-DEVICE-ORIENTATED HUMAN-COMPUTER INTERACTION
APPARATUS AND METHOD THEREOF
BACKGROUND OF THE INVENTION
Technical Field
[0001] The present invention relates to the technical field of human-computer
interaction, and
more particularly to a smart-device-orientated human-computer interaction
apparatus and
a method thereof.
Description of Related Art
[0002] Human-computer interaction (HCI), also known as human-machine
interaction (HMI), is
a science of interactive relationship between systems and users. Herein, the
systems may
be various machines or computer software. For example, a smart robot can have
human-
machine conversation with users using a voice system installed therein. In the
existing
human-computer interaction schemes, a smart robot usually uses the same
channel to
identify or control task capability and behavior capability. The task
capability comprises
listen capability and watch capability, while the behavior capability
comprises speak
capability and action capability, and tends to have logical confusion when
dealing with
task capability and behavior capability at the same time, leading to inferior
smoothness
of interaction and poor human-computer interaction experience at the user
side.
SUMMARY OF THE INVENTION
[0003] The objective of the present invention is to provide a smart-device-
orientated human-
computer interaction apparatus and a method thereof for improving user
experience about
human-computer interaction.
1
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
[0004] To achieve the objective, in one aspect, the present invention provides
a smart-device-
orientated human-computer interaction apparatus, which comprises an
identification
interface, an interaction model unit, an intention-identifying unit, and a
control interface;
[0005] the identification interface serving to initiate and send an
identification request to an
identification channel, wherein the identification request contains listen
capability data
and/or watch capability data;
[0006] the interaction model unit comprising a channel controller, a focus
controller, and a
strategy manager for managing control strategy priority of the identification
request in a
said predetermined identification channel and control strategy priority of a
control request
in a control channel, the focus controller serving to acquire a control logic
for the
identification request according to the control strategy priority of the
identification
request, and the channel controller serving to execute a control action on the
identification
request in the identification channel according to the control logic so as to
obtain task
data;
[0007] the intention-identifying unit serving to identify an intention
instruction of the task data
and generate a control request, so that the focus controller acquires the
control logic of
the control request according to the control strategy priority of the control
request, and
the channel controller further serving to act on the control request in the
control channel
according to the control logic so as to obtain a behavior action; and
[0008] the control interface serving to output the behavior action in a
controlled manner, wherein
the behavior action comprises speak capability data and/or action capability
data.
[0009] Preferably, the channel controller comprises an identification
controller and an action
controller;
[0010] the identification controller serving to, according to an active state
of the identification
request in the current identification channel and an active state of the
control request in
the control channel, use the identification request control logic output by
the focus
controller to execute the control action on the identification request in the
identification
channel so as to obtain the task data; and
2
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
[0011] the action controller serving to, according to an active state of the
identification request
in the current identification channel and an active state of the control
request in the control
channel, use the control request control logic output by the focus controller
to act on the
control request in the control channel so as to obtain the behavior action.
[0012] More preferably, the intention-identifying unit comprises an intention
engine, a data
identification interface, and an intention execution interface;
[0013] the data identification interface serving to receive the task data and
upload the task data
to the intention engine;
[0014] the intention engine serving to identify the task data so as to obtain
the intention
instruction and distribute the intention instruction to the intention
execution interface; and
[0015] the intention execution interface serving to send the intention
instruction to the action
controller to request for communication, so that the action controller
generates a control
request based on the intention instruction and sends it to the control
channel.
[0016] Exemplarily, the identification strategies include one or more of:
stopping executing the
identification request, executing the identification request in the
foreground, executing
the identification request in the background, blocking the identification
request
temporarily, and resuming executing the identification request. The control
strategies
include one or more of: stopping executing the control request, executing the
control
request in the foreground, executing the control request in the background,
blocking the
control request temporarily, and resuming executing the control request.
[0017] Preferably, the identification request includes conversation-type
identification requests
and context-type identification requests, and the conversation-type
identification requests
have the higher priority of the identification strategies than the priority of
the
identification strategies of the context-type identification requests.
[0018] Preferably, further comprising,
3
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
[0019] after the identification controller has executed the control action on
the identification
request in the identification channel and has generated the task data, the
identification
request is automatically switched to an inactive focus state from a focus
state; and
[0020] after the action controller has executed the control action on the
control request in the
control channel and has generated the behavior action, the behavior action is
automatically switched to an inactive focus state from a focus state.
[0021] As compared to the prior art, the smart-device-orientated human-
computer interaction
apparatus of the present invention provides the following beneficial effects:
[0022] The smart-device-orientated human-computer interaction apparatus of the
present
invention comprises an identification interface, an interaction model unit, an
intention-
identifying unit, and a control interface. The identification interface
receives an
identification request initiated by a user, and transmits the identification
request to the
identification channel, so as to input listen capability data and/or watch
capability data by
means of human-computer interaction. At this time, the focus controller reads
the priority
of the identification strategies of the identification channel from the
strategy manager and
generates a control logic for the identification request, so that the channel
controller can
perform action control on the identification request in the identification
channel according
to the control logic, so as to acquire task data. Then the intention-
identifying unit
identifies the intention instruction of the task data, thereby generating a
control request.
The channel controller inputs the control request to the control channel. At
this time, the
focus controller reads priority of the identification strategies of the
control channel from
the strategy manager and generates a control logic for the control request. At
last, the
channel controller performs action control on the control request in the
control channel
according to the control logic, thereby outputting a behavior action, so that
the control
interface can output the behavior action. Thereby, human-computer interaction
control of
speak capability data and/or action capability data can be achieved.
[0023] It is thus clear that the present embodiment uses dual-channel control
in virtue of the
4
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
identification channel and the control channel to separately control
identification requests
and control requests, so as to reasonably control identification requests
initiated by users
and behavior actions that are output based on identification request, thereby
improving
smoothness of action control of bionic capability and making human-computer
interaction more user friendly. Additionally, due to the interactive design of
the channel
controller, the focus controller, and the strategy manager, the implementation
logic of
identification requests and behavior actions can be flexibly controlled based
on priority
of the control strategies, so as to increase bionic experience of human-
computer
interaction.
[0024] In another aspect, the present invention provides a smart-device-
orientated human-
computer interaction method, which is applied to the smart-device-orientated
human-
computer interaction apparatus described previously. The method comprises:
[0025] initiating and sending an identification request to an identification
channel, wherein the
identification request contains listen capability data and/or watch capability
data;
[0026] acquiring a control logic for the identification request according to
priority of
identification strategies for the identification request, and executing a
control action on
the identification request in the identification channel according to the
control logic;
[0027] identifying an intention instruction of the task data and generating a
control request, then
acquiring a control logic for the control request according to control
strategy priority of
the control request, and acting on the control request in the control channel
according to
the control logic, so as to obtain a behavior action; and
[0028] outputting the behavior action in a controlled manner, wherein the
behavior action
comprises speak capability data and/or action capability data.
[0029] Optionally, the identification strategies include one or more of:
stopping executing the
identification request, executing the identification request in the
foreground, executing
the identification request in the background, blocking the identification
request
temporarily, and resuming executing the identification request, and the
control strategies
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
include one or more of: stopping executing the control request, executing the
control
request in the foreground, executing the control request in the background,
blocking the
control request temporarily, and resuming executing the control request.
[0030] Preferably, the identification request includes conversation-type
identification requests
and context-type identification requests, and the conversation-type
identification requests
have the higher priority of the identification strategies than the priority of
the
identification strategies of the context-type identification requests.
[0031] As compared to the prior art, the disclosed human-computer interaction
method provides
beneficial effects that are similar to those provided by the disclosed human-
computer
interaction apparatus as enumerated above, and thus no repetitions are made
herein.
[0032] In a third aspect, the present invention provides a computer-readable
storage medium,
which stores a computer program that when executed performs the steps of the
human-
computer interaction method as described above.
[0033] As compared to the prior art, the disclosed computer-readable storage
medium provides
beneficial effects that are similar to those provided by the disclosed human-
computer
interaction metho as enumerated above, and thus no repetitions are made
herein.
BRIEF DESCRIPTION OF THE DRAWINGS
[0034] The accompanying drawings are provided herein for better understanding
of the present
invention and form a part of this disclosure. The illustrative embodiments and
their
descriptions are for explaining the present invention and by no means form any
improper
limitation to the present invention, wherein:
[0035] FIG. 1 is a flowchart of a smart-device-orientated human-computer
interaction method of
an embodiment of the present invention.
6
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
DETAILED DESCRIPTION OF THE INVENTION
[0036] To make the foregoing objectives, features, and advantages of the
present invention
clearer and more understandable, the following description will be directed to
some
embodiments as depicted in the accompanying drawings to detail the technical
schemes
disclosed in these embodiments. It is, however, to be understood that the
embodiments
referred herein are only a part of all possible embodiments and thus not
exhaustive. Based
on the embodiments of the present invention, all the other embodiments can be
conceived
without creative labor by people of ordinary skill in the art, and all these
and other
embodiments shall be encompassed in the scope of the present invention.
[0037] Embodiment 1
[0038] Referring to FIG. 1, the present embodiment provides a smart-device-
orientated human-
computer interaction apparatus, which comprises an identification interface,
an
interaction model unit, an intention-identifying unit, and a control
interface. The
identification interface serves to initiate and send an identification request
to an
identification channel. The identification request includes listen capability
data and/or
watch capability data. The interaction model unit comprises a channel
controller, a focus
controller, and a strategy manager for managing control strategy priority of
the
identification request in a said predetermined identification channel and
control strategy
priority of a control request in a control channel. The focus controller is
for acquiring a
control logic for the identification request according to priority of
identification strategies
for the identification request. The channel controller is for performing a
control action on
the identification request in the identification channel according to the
control logic, so
as to obtain task data. The intention-identifying unit is for identifying an
intention
instruction of the task data and generating a control request. The focus
controller acquires
the control logic of the control request according to the priority of the
control strategies
of the control request. The channel controller is further for according to the
control logic,
acting on the control request in the control channel so as to obtain behavior
actions. The
7
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
control interface is for outputting the behavior action in a controlled
manner. The
behavior action comprises speak capability data and/or action capability data.
[0039] The smart-device-orientated human-computer interaction apparatus of the
present
invention comprises an identification interface, an interaction model unit, an
intention-
identifying unit, and a control interface. The identification interface
receives an
identification request initiated by a user, and transmits the identification
request to the
identification channel, so as to input listen capability data and/or watch
capability data by
means of human-computer interaction. At this time, the focus controller reads
the priority
of the identification strategies of the identification channel from the
strategy manager and
generates a control logic for the identification request, so that the channel
controller can
perform action control on the identification request in the identification
channel according
to the control logic, so as to acquire task data. Then the intention-
identifying unit
identifies the intention instruction of the task data, thereby generating a
control request.
The channel controller inputs the control request to the control channel. At
this time, the
focus controller reads priority of the identification strategies of the
control channel from
the strategy manager and generates a control logic for the control request. At
last, the
channel controller performs action control on the control request in the
control channel
according to the control logic, thereby outputting a behavior action, so that
the control
interface can output the behavior action. Thereby, human-computer interaction
control of
speak capability data and/or action capability data can be achieved.
[0040] It is thus clear that the present embodiment uses dual-channel control
in virtue of the
identification channel and the control channel to separately control
identification requests
and control requests, so as to reasonably control identification requests
initiated by users
and behavior actions that are output based on identification request, thereby
improving
smoothness of action control of bionic capability and making human-computer
interaction more user friendly. Additionally, due to the interactive design of
the channel
controller, the focus controller, and the strategy manager, the implementation
logic of
8
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
identification requests and behavior actions can be flexibly controlled based
on priority
of the control strategies, so as to increase bionic experience of human-
computer
interaction.
[0041] Still referring to FIG. 1, the channel controller of the described
embodiment has an
identification controller and an action controller. The identification
controller is for,
according to an active state of the identification request in the current
identification
channel and an active state of the control request in the control channel,
using the
identification request control logic output by the focus controller to execute
the control
action on the identification request in the identification channel so as to
obtain the task
data; and an action controller for according to an active state of the
identification request
in the current identification channel and an active state of the control
request in the control
channel, using the control request control logic output by the focus
controller to act on
the control request in the control channel so as to obtain the behavior
action.
[0042] The intention-identifying unit in the embodiment comprises an intention
engine, a data
identification interface, and an intention execution interface. The data
identification
interface receives the task data and uploads the data to the intention engine.
The intention
engine identifies the task data, so as to obtain an intention instruction and
distribute the
intention instruction to the intention execution interface. The intention
execution
interface sends the intention instruction to the action controller to request
for
communication, so that the action controller generates the control request
based on the
intention instruction and sends the control request to the control channel.
Therein, the
intention engine may be any known intention identification engine, and the
details are not
described herein.
[0043] Therein, the identification strategies include one or more of: stopping
executing the
identification request, executing the identification request in the
foreground, executing
the identification request in the background, blocking the identification
request
9
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
temporarily, and resuming executing the identification request, and the
control strategies
include one or more of: stopping executing the control request, executing the
control
request in the foreground, executing the control request in the background,
blocking the
control request temporarily, and resuming executing the control request.
[0044] In general, the identification request includes conversation-type
identification requests
and context-type identification requests, and the conversation-type
identification requests
have the higher priority of the identification strategies than the priority of
the
identification strategies of the context-type identification requests.
[0045] In order to realize the technical scheme of the present embodiment, an
example is
described below with reference to FIG. 1.
[0046] The first step is to establish the identification interface (Input
Interface) and the control
interface (Output Interface). The identification interface is a general-
purpose data
collecting interface, which is used to connect external bionic capability
having input
capability, such as the listen capability data and/or the watch capability
data. When a
smart device equipped with the human-computer interaction apparatus starts to
work, an
identification request from an external bionic capability can be acquired by
calling the
identification interface. Then for the identification request, the intention
instruction can
be identified using the intention engine. Similarly, the control interface is
a general-
purpose output control interface. The control interface connects external
bionic capability
that has output capability. When the smart device equipped with the human-
computer
interaction apparatus starts to work, a control request corresponding to the
identified
intention instruction can be output to control actions of the external bionic
capability.
[0047] The second step is to establish the interaction model unit, which
comprises a channel
controller, a focus controller, and a strategy manager. Therein, the channel
controller at
least comprises an identification controller (Channel 1) and an action
controller (Channel
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
2). When the identification request provided by the identification interface
passes through
the identification channel, the focus controller execute strategical control
over
identification requests in the identification channel according to the
predetermined
priority of the control strategies, so as to obtain task data correspondingly.
Then the task
data are transmitted to the intention-identifying unit to identify an
intention instruction,
thereby generating a control request. The control module then transmits the
control
request to the control channel. Afterward, the focus controller executes
strategical control
over the control requests in the control channel according to the
predetermined priority
of the control strategies, so as to obtain behavior actions correspondingly.
At last,
behavior actions of the external bionic capability can be indirectly
controlled through the
control interface.
[0048] The strategy manager is a caching tool that defines channel strategies
and priority levels
in the channel controller. For example, when the identification request in the
identification controller is in the focus state (Activity), the action
controller may,
according to the strategies, preferentially execute identification request
actions in the
identification channel. On the contrary, when the control request in the
action controller
is in the focus state (Activity), the identification controller may
alternatively
preferentially execute control request actions in the control channel
according to the
strategies.
[0049] The focus controller is for, according to the strategy manager,
generating a specific
control logic for the channel controllers in the identification channel and
the control
channel. Exemplarily, the control logic is as below.
[0050] When only the identification channel or the control channel has a
controller in the active
state, the controller is set into the focus state and set in the foreground.
[0051] When the identification controller or the action controller has to be
set into the active
11
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
state, and both the current channel and other channel have controllers in the
active state,
the focus controller first checks the priority level strategies of all these
controllers, and
runs the corresponding controllers according to their priority levels.
[0052] For facilitating understanding, the foregoing embodiment of the present
invention is
further explained with reference to some application scenarios.
[0053] For a smart robot equipped with the human-computer interaction
apparatus of the
embodiment, the listen capability may be realized using a microphone array
that collects
voice data; the watch capability may be realized using a camera that collects
image
information; the speak capability may be realized using a speaker that sounds;
and the
action capability may be realized using movable mechanisms of the robot.
Therein, the
listen capability and the watch capability are connected to the identification
interface, and
the speak capability and the action capability are connected to the control
interface.
[0054] When a user talks to the smart robot, the identification channel is
first requested to
perform interactive control. When the priority of the control strategies of
the
identification request acquired by the current identification channel through
the strategy
manager is of the highest priority level, the focus controller sets the
identification request
in the foreground, and sets requests in other channel, such as a control
request in the
control channel, into the background. Exemplarily, the control behavior in the
foreground
may be in the following situations:
[0055] The smart robot is listening, and is preferentially processing
identification request that it
is listening to;
[0056] The smart robot is watching, is preferentially processing the
identification request that it
is watching; and
[0057] The smart robot is watching and listening, and is preferentially
processing the
identification request that it is watching and listening to.
12
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
[0058] The control behavior in the background may be in the following
situations:
[0059] The smart robot is speaking and acting, and speaking and acting are to
be stopped or
paused or blocked;
[0060] The smart robot is only speaking, and speaking is to be stopped or
paused or blocked; and
[0061] The smart robot is only acting, and acting is to be stopped or paused
or blocked.
[0062] When the priority of the control strategies of the identification
request acquired by the
current identification channel through the strategy manager is not of the
highest priority
level, the focus controller sets the identification request into the
background, and sets
control requests in other channel requests, such as the control channel, in
the foreground.
Exemplarily, the control behavior in the background may be in the following
situations:
[0063] The smart robot is listening, and the identification request it is
listening to is stopped or
paused or blocked;
[0064] The smart robot is watching, and the identification request it is
watching is stopped or
paused or blocked; and
[0065] The smart robot is watching and listening, and the identification
request it is watching
and listening to is stopped or paused or blocked.
[0066] At this time, the control behavior in the foreground may be in the
following situations:
[0067] The smart robot is speaking and acting, and speaking and acting are
resumed;
[0068] The smart robot is only speaking, and speaking is resumed; and
[0069] The smart robot is only acting, and acting is resumed.
[0070] In addition, after the foregoing process, the data identification
interface uploads the
identification request to the intention engine for intention identification.
After
identification, the intention instruction is distributed to the intention
execution interface.
The interface requests for interaction through the action controller. An
example is
described below.
13
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
[0071] When the priority of the control strategies of the control request
acquired by the current
control channel through the strategy manager is of the highest priority level,
the focus
controller sets the control request in the foreground, and sets requests in
other channels,
such as an identification request in the identification channel in the
background.
Exemplarily, the control behavior in the foreground may be in the following
situations:
[0072] The smart robot is speaking or acting. When the control channel
receives the present
control request, it preferentially executes the present control request, and
downgrades the
priority of the control request for which the smart robot is speaking and
acting to a lower
priority level. After the present control request has been executed, execution
of the earlier
downgraded control request is resumed.
[0073] The smart robot has earlier speaking or acting stopped, paused, or
blocked. When the
control channel receives the present control request, it preferentially
executes the present
control request, and the control request of the earlier speaking or acting
remains stopped,
paused, or blocked.
[0074] At this time, the control behavior in the background may be in the
following situations:
[0075] The smart robot is listening, and the identification request it is
listening to is stopped or
paused or blocked;
[0076] The smart robot is watching, and the identification request it is
watching is stopped or
paused or blocked; and
[0077] The smart robot is watching and listening, and the identification
request it is listening to
and watching is stopped or paused or blocked.
[0078] When the priority of the control strategies of the control request
acquired by the current
control channel through the strategy manager is not of the highest priority
level, the focus
controller sets the control request into the background, and sets requests in
other channels,
14
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
such as an identification request in the identification channel, into the
foreground.
Exemplarily, the control behavior in the background may be in the following
situations:
[0079] After the control channel receives the present control for request
speaking or acting, it
preferentially executes the previous control request. After the previous
control request
has been executed, it executes the present control request for speaking or
acting.
[0080] At this time, the control behavior in the foreground may be in the
following situations:
[0081] The smart robot is listening or watching, and the current is paused,
stopped, or blocked,
and then resumed; and
[0082] The smart robot is listening or watching, and no intervention is made
before the current
execution is done.
[0083] Embodiment 2
[0084] The present embodiment provides a smart-device-orientated human-
computer interaction
method, which comprises:
[0085] initiating and sending an identification request to an identification
channel, wherein the
identification request contains listen capability data and/or watch capability
data;
[0086] acquiring a control logic for the identification request according to
priority of
identification strategies for the identification request, and executing a
control action on
the identification request in the identification channel according to the
control logic;
[0087] identifying an intention instruction of the task data and generating a
control request, then
acquiring a control logic for the control request according to control
strategy priority of
the control request, and acting on the control request in the control channel
according to
the control logic, so as to obtain a behavior action; and
[0088] outputting the behavior action in a controlled manner, wherein the
behavior action
comprises speak capability data and/or action capability data.
[0089] Preferably, the identification strategies include one or more of:
stopping executing the
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
identification request, executing the identification request in the
foreground, executing
the identification request in the background, blocking the identification
request
temporarily, and resuming executing the identification request, and the
control strategies
include one or more of: stopping executing the control request, executing the
control
request in the foreground, executing the control request in the background,
blocking the
control request temporarily, and resuming executing the control request.
[0090] Exemplarily, the identification request includes conversation-type
identification requests
and context-type identification requests, and the conversation-type
identification requests
have the higher priority of the identification strategies than the priority of
the
identification strategies of the context-type identification requests.
[0091] As compared to the prior art, the disclosed human-computer interaction
method provides
beneficial effects that are similar to those provided by the disclosed human-
computer
interaction apparatus as enumerated above, and thus no repetitions are made
herein.
[0092] Embodiment 3
[0093] The present embodiment provides a computer-readable storage medium. The
computer-
readable storage medium stores therein a computer program. When executed by a
processor, the computer program performs the steps of the smart- device-
orientated
human-computer interaction method as described previously.
[0094] As compared to the prior art, the disclosed computer-readable storage
medium provides
beneficial effects that are similar to those provided by the disclosed human-
computer
interaction metho as enumerated above, and thus no repetitions are made
herein.
[0095] As will be appreciated by people of ordinary skill in the art,
implementation of all or a
part of the steps of the method of the present invention as described
previously may be
realized by having a program instruct related hardware components. The program
may
16
Date Recue/Date Received 2022-06-15
CA 03164768 2022-06-15
be stored in a computer-readable storage medium, and the program is about
performing
the individual steps of the methods described in the foregoing embodiments.
The storage
medium may be a ROM/RAM, a hard drive, an optical disk, a memory card or the
like.
[0096] The present invention has been described with reference to the
preferred embodiments
and it is understood that the embodiments are not intended to limit the scope
of the present
invention. Moreover, as the contents disclosed herein should be readily
understood and
can be implemented by a person skilled in the art, all equivalent changes or
modifications
which do not depart from the concept of the present invention should be
encompassed by
the appended claims. Hence, the scope of the present invention shall only be
defined by
the appended claims.
17
Date Recue/Date Received 2022-06-15