WO2016206645A1

WO2016206645A1 - Method and apparatus for loading control data into machine device

Info

Publication number: WO2016206645A1
Application number: PCT/CN2016/087260
Authority: WO
Inventors: 聂华闻
Original assignee: 北京贝虎机器人技术有限公司
Priority date: 2015-06-26
Filing date: 2016-06-27
Publication date: 2016-12-29
Also published as: WO2016206642A1; WO2016206643A1; WO2016206646A1; WO2016206647A1; WO2016206644A1

Abstract

A method and apparatus for loading control data into a machine device are provided. The method comprises the following steps: inputting perceptual data corresponding to a machine device (701), wherein the perceptual data is generated according to at least a portion of the predefined plurality of data elements based on the information perceived by the machine device; providing the perceptual data to a plurality of search modules (702), wherein each of the plurality of search modules is used for searching for control data matched with a perceptual data in different groups of control data, wherein the control data is used for urging the machine device to generate actions, and the actions include an action of the behavior type and an action of the motion type; providing at least one control data candidate derived from the plurality of search modules is provided (703); and urging the machine device to generate an action using control data candidate (704). The method improves the control effect of the robot and other equipments through a plurality of control data groups.

Description

Method and device for loading control data for machine devices

Technical field

The present invention relates to the field of intelligent control technologies, and in particular, to a method and apparatus for loading control data for a machine device.

Background technique

For a machine device such as a robot, there are various control methods in the related art, but the control method in the related art is also not satisfactory, and the control method is complicated.

Summary of the invention

This Summary is provided to introduce a selection of concepts in the <RTIgt; The summary is not intended to identify key features or essential features of the claimed subject matter, and is not intended to be used in any way to limit the claimed subject matter.

Briefly, various aspects of the subject matter described herein relate to methods and apparatus for loading control data for a machine device to facilitate at least control of the machine device. Machine devices may include, but are not limited to, smart devices or devices such as robots.

In one aspect, a method for loading control data for a machine device is provided, comprising: inputting perceptual data corresponding to a machine device, wherein the perceptual data is generated according to at least a portion of a predefined plurality of data elements based on information sensed by the machine device; Providing the perceptual data to a plurality of search modules, wherein each of the plurality of search modules is configured to search for control data matching the perceptual data from different sets of control data, wherein the control data is used to cause the machine device to generate an action And wherein the action generated by the machine device includes an action type and/or a motion type of action; providing a candidate for at least one control data from the plurality of search modules; and controlling the machine device to generate an action using the candidate of the control data.

In one example, the perceptual data described above may be provided to a plurality of search modules separately until one of the plurality of search modules provides control data. Without limitation, a plurality of search modules may be ordered in order of priority, and the perceptual data is separately provided to the plurality of search modules in accordance with the priority.

In one example, the perceptual data can be provided in parallel to a plurality of search modules, in which case candidates for control data from the plurality of search modules can also be selected based on the global algorithm.

In one example, one of the different sets of control data is stored locally on the machine device and/or the local area to which the machine device is connected. Without limitation, another of the different sets of control data may be stored in a remote computer system.

In one example, each of the different sets of control data data corresponds to a different mode of the machine device.

In one example, the different sets of control data include any combination of the control data set of the machine device, or the control data set of the machine device corresponding machine version, or a control data set common to a plurality of machine devices. Without limitation, different sets of control data may be ordered by priority.

In one example, the perceptual data is separately provided to the plurality of search modules until one of the plurality of search modules provides the control data; and the plurality of search modules are ordered according to the priority of the different control data sets The sensing data is separately provided to the plurality of search modules in accordance with the priority.

In one example, the perceptual data may be provided to a plurality of search modules in parallel, and candidates for control data from the plurality of search modules may also be selected based on a global algorithm. Without limitation, the parameters of the global algorithm may include different priorities of the control data set.

In one example, the plurality of search modules can search for control data based on conditional data associated with the control data, wherein the conditional data is generated based on a plurality of data elements corresponding to a predefined plurality of data elements that generate the perceptual data.

In one example, a plurality of data elements in the condition data may be at least partially ordered in order of priority, and the plurality of search modules sort the candidates of the control data that the perceptual data matches in order of priority.

In one example, the plurality of search modules can load candidates for the control data that are aware of the data match based on the sensing data matching at least a portion of the data elements in the condition data.

In one example, where the control data corresponds to a basic behavior or has a plurality of basic behaviors that perform logical constraints, wherein each of the basic behaviors can be directly performed by the machine device. Without limitation, each of the basic behaviors is defined by a behavior name and behavior control parameters.

In one example, multiple search modules correspond to different search algorithms.

In another aspect, a method for loading control data for a machine device is provided, comprising: inputting perceptual data corresponding to the machine device, wherein the perceptual data is based on at least a portion of the predefined plurality of data elements based on the information sensed by the machine device generate;

Providing the perceptual data to a plurality of search algorithms to search for control data for causing the machine device to generate an action from a different range of control data, wherein the different control data ranges include control data sets of the machine device, or machine The device corresponds to any combination of a control data set of the machine device version or a control data set common to the plurality of machine devices, wherein the action generated by the machine device includes an action type and/or a motion type action; determining at least one based on the plurality of search algorithms The candidate of the item control data; and the candidate for controlling the data is used to control the machine device to generate an action.

In one example, the perceptual data is provided separately to a plurality of search algorithms until one of the plurality of search algorithms provides control data.

In one example, multiple search algorithms are ordered by priority, and perceptual data is separately provided to multiple search algorithms in accordance with priority. Non-limiting, the order of priority from high to low is: a control data set of the machine device, a control data set corresponding to the machine device version of the machine device, and a control data set common to a plurality of machine devices.

In one example, the perceptual data may be provided to multiple search algorithms in parallel, and candidates for control data from multiple search algorithms may also be selected based on the global algorithm.

In one example, the control data corresponds to a basic behavior or has a plurality of basic behaviors that perform logical constraints, wherein each of the basic behaviors can be directly executed by the machine device. Without limitation, each of the basic behaviors is defined by a behavior name and behavior control parameters.

In one example, the control data is associated with conditional data, and the plurality of search algorithms can search for control data based on conditional data associated with the control data, wherein the conditional data is based on a plurality of predefined plurality of data elements corresponding to the generated perceptual data Data elements are pre-generated.

In still another aspect, an apparatus for loading control data for a machine device is provided, comprising: an input module, configured to input sensory data corresponding to the machine device, wherein the sensory data is based on a plurality of predefined data elements based on information sensed by the device device At least partially generated; a first providing module, configured to provide the sensing data to the plurality of search modules, wherein each of the plurality of search modules is configured to search for control data matching the sensing data from different control data sets, wherein The control data is used to cause the machine device to generate an action, wherein the action generated by the machine device includes an action type and/or a motion type of action; the second providing module is configured to provide a candidate for the at least one control data from the plurality of search modules; And a control module for controlling the machine device to generate an action using the candidate of the control data.

In one example, the perceptual data can be provided to the plurality of search modules separately until one of the plurality of search modules provides control data.

In one example, multiple search modules may be prioritized and perceptual data is provided to multiple search modules separately according to the priority.

In one example, the perceptual data is provided in parallel to a plurality of search modules, and the second providing module is further configured to select candidates for control data from the plurality of search modules based on the global algorithm.

In one example, one of the different sets of control data is stored locally on the machine device and/or the local area to which the machine device is connected. Without limitation, the other of the different sets of control data is stored in a remote computer system.

In one example, the different sets of control data include any combination of the control data set of the machine device, or the control data set of the machine device corresponding machine version, or the control data set common to the plurality of machine devices, but is not limited thereto.

In one example, different sets of control data are ordered by priority.

In one example, the perceptual data is separately provided to a plurality of search modules until one of the plurality of search modules provides the control data; and the plurality of search modules are ordered according to a prioritization of different sets of control data, the perceptual data They are separately provided to a plurality of search modules in accordance with the priority.

In one example, the perceptual data is provided to the plurality of search modules in parallel, wherein the first providing module is further configured to select candidates of the control data from the plurality of search modules based on the global algorithm, wherein the parameters of the global algorithm include at least different Controls the priority of the data collection.

In one example, the control data is associated with conditional data, the plurality of search modules searching for control data based on conditional data associated with the control data, wherein the conditional data is based on data elements corresponding to a predefined plurality of data elements that generate the perceptual data produce.

In one example, the plurality of data elements in the condition data are at least partially ordered in order of priority, and the plurality of search modules sort the candidates of the control data that the perceptual data matches in order of priority.

In one example, the control data corresponds to a basic behavior or has a plurality of basic behaviors that perform logical constraints, wherein each of the basic behaviors can be directly executed by the machine device. Optionally, each of the basic behaviors is defined by a behavior name and a behavior control parameter.

In still another aspect, an apparatus for loading control data for a machine device is provided, comprising: an input module, configured to input sensing data corresponding to the machine device, wherein the sensing data is based on the information sensed by the machine device according to a predefined plurality of data elements. At least partially generating; providing a module for providing the sensing data to the plurality of search algorithms to search for control data for causing the machine device to generate an action from a different control data range, wherein the different control data ranges include the machine device Controlling a data set, or a control data set of a machine device corresponding machine version, or any combination of control data sets common to a plurality of machine devices, wherein the action comprises an action type and/or a motion type of action; the determining module is configured to The search algorithm determines candidates for at least one piece of control data; and a control module for controlling the machine device to generate an action using the candidate of the control data.

In one example, the perceptual data is provided separately to the plurality of search algorithms until one of the plurality of search algorithms provides a candidate for the control data.

In one example, the plurality of search algorithms are ordered by priority, and the perceptual data is separately provided to the plurality of search algorithms in accordance with the priority. Without limitation, the order of priority from high to low is: a control data set of the machine device, a control data set corresponding to the machine device version of the machine device, and a control data set common to the plurality of machine devices.

In one example, the perceptual data is provided in parallel to a plurality of search algorithms, wherein candidates for control data from the plurality of search algorithms are also selected based on the global algorithm.

In one example, the control data is associated with conditional data, the plurality of search algorithms searching for control data based on the conditional data associated with the control data, the conditional data being based on a plurality of the plurality of predefined data elements corresponding to the generated perceptual data Data elements are generated.

As a non-limiting example, the machine device may perceive information including information of the physical environment in which the machine device is located (such as an item, or an entity such as a human user, or a machine device), and/or information of the machine device itself. The machine device may include one or more sensor devices, which may include sensor modules of software combination and/or hardware components, may include sensor devices physically coupled to the device devices, and/or sensor devices communicatively coupled to the device devices, But it is not limited to this.

As a non-limiting example, the information sensed by the machine device may include at least one or any combination of visual, or tactile, or audible, or gesture, etc., but is not limited thereto, and any information that can be perceived may be.

Restrictively, the machine device may include a robot having human-computer interaction capabilities, a mobile robot, etc., and in some examples, the machine device may include moving parts (such as limbs, wheels, tracked, etc.) to generate mechanical motion. , but not limited to this.

Through the embodiment of the invention, the machine device is controlled to generate actions according to different control data sets, and the effect of controlling the machine device is improved.

DRAWINGS

The drawings described herein are provided to provide a further understanding of the invention, and are not intended to limit the invention. In the drawing:

1 is a schematic diagram of an example of a machine device communication system 100;

2 is a schematic structural view of a machine device 110;

3 is a schematic diagram of a system 300 for loading control data for a machine device 110;

4 is a schematic diagram of loading control data for the machine device 110 at the local and server;

Figure 5 is a schematic diagram of loading control data from different control data ranges;

6 is a schematic structural view of a machine device 110;

7 is a flow chart of a method of loading control data for machine device 110.

detailed description

In order to make the objects, technical solutions and advantages of the present invention more comprehensible, the present invention will be further described in detail with reference to the embodiments and drawings. The illustrative embodiments of the present invention and the description thereof are intended to explain the present invention, but are not intended to limit the invention.

It should be understood that any examples herein are non-limiting. Therefore, the invention is not limited to any specific embodiments, aspects, concepts, structures, functions or examples described herein. On the contrary, any one embodiment, aspect, structure, function, or example described herein is limited.

Features described and/or illustrated with respect to one embodiment or example may be used in the same manner or in a similar manner in one or more other embodiments or examples, and/or combined with features of other embodiments or examples or Instead of the features of other embodiments or examples.

It should be emphasized that the words "including", "based on" or "comprising", when used in the specification, are used to mean the presence of the recited features, elements, steps, components or components, but do not exclude one or more The presence or addition of other features, elements, steps, components, or combinations thereof.

FIG. 1 is a schematic diagram of an example of a machine device communication system 100.

Referring to FIG. 1, communication system 100 includes machine device 110, local user terminal 120, remote user terminal 130, one or more servers 140, static sensor 180, and static sensor 181.

Machine device 110 can communicate with local user terminal 120, and/or static sensor 180 and/or static sensor 181 via private network 150. Machine device 110 can communicate with remote user terminal 130, and/or server 140 via private network 150 and public network 160. Machine device 110 can also communicate with static sensor 180 and/or static sensor 181, etc. via hub 170. It should be understood that the communication link herein is for illustrative purposes only and is not a limitation of the manner of communication. In fact, any suitable communication link is possible. This embodiment does not rely on a particular communication link.

In some examples, machine device 110 may perceive information, which may include information of the physical environment in which machine device 110 is located and/or information of machine device 110 itself, including but not limited to items, or human users. The information of the human user may include at least one or any combination of visual, or audible, or tactile, or gestures, etc., but is not limited thereto.

In some examples, machine device 110 may include one or more sensor devices 111. The sensor device 111 may include a camera (such as a camera, a depth camera, etc.), or a microphone, or an infrared sensor, or a motion sensor, a global positioning system (GPS) module, an accelerometer, a gyroscope, a light sensor, a nearby device signal strength detection module, and the like. . Additionally, some of the sensor devices 111 include software-based sensors that can provide high level or fine granularity of information. In addition to the raw sensor inputs, these sensor devices 111 process the appropriate input data to provide meaningful information, for example, the sensor device 111 can include, for example, a sensor module 112 that can include the words and/or body gesture/movement identification module, Face recognition module and more.

In some examples, sensor device 111 of machine device 110 may also include a proxy for static sensor 180 and/or static sensor 181 as shown in FIG. 1, which may be in communication with static sensor 180 and/or static sensor 181 to The static sensor 180 and/or the static sensor 181 acquire data or send instructions or the like to the static sensor 180 and/or the static sensor 181.

In some examples, local user terminal 120 and remote user terminal 130 can control machine device 110. The local user terminal 120 and the remote user terminal 130 can transmit control data in a message, and the machine device 110 can receive the message and read the control data to generate an action based on the control data. The control data may be edited by the local user terminal 120 and the remote user terminal 130, but is not limited thereto. The local user terminal 120 and the remote user terminal 130 may include an application, such as a mobile application, or a web program, or an application, to control the machine device 110.

In some examples, server 140 can control machine device 110 to generate an action. The server 140 can transmit control data in the communication signal for receipt at the machine device 110, and the machine device 110 can generate an action based on the control data transmitted by the server 140. Without limitation, the machine device 110 may include a robot, or a mobile robot or the like, but is not limited thereto.

In some examples, machine device 110 may transmit information perceived by machine device 110 in a communication signal over private network 150 and public network 160 for receipt at server 140. Server 140 may generate control data for machine device 110 based on information sensed by machine device 110 and transmit control data in the communication signal for receipt at machine device 110. The machine device 110 receives the control data and performs an action based on the control data. The actions to be performed by the machine device 110 include actions of a type of behavior and/or type of motion.

In some examples, machine device 110 may take control data locally on machine device 110 and/or a local network to which machine device 110 is connected, and generate an action based on the control data. The machine device 110 may, based on the sensed information, obtain control data that matches the perceived information locally on the local network to which the machine device 110 and/or the machine device 110 is connected, and generates an action for the perceived information.

In some examples, the control data may correspond to a basic behavior or a plurality of basic behaviors with execution logic. Execution of logical constraints may include logic that is constrained by time logic and/or event logic. Machine device 110 may include/have a plurality of components for performing basic behaviors, each of which performs a basic behavior for performing a corresponding basic behavior. Without limitation, the basic behavior may be defined by a behavior name and a behavior control parameter, and the behavior control parameter may be "Null" (empty). Components that perform basic behavior can perform basic behavior based on behavioral control parameters.

Basic behavior

As a non-limiting example, the basic behavior can be written in the JSON language, but is not limited thereto, and other methods are also possible. Non-limiting basic behaviors may include:

1. Let the machine device 110 speak

The behavior name is: audio_speak;

Behavior control parameters can include: text (content to say), volume (volume of speech), etc. (eg, vocal gender, or vocal age, etc.)

JSON is expressed as follows:

"audio_speak": {"text": "How are you", "volume": "50%"}

Without limitation, "text" may include a conversion character that corresponds to a parameter. For example, the "owner" conversion character can be defined as "@master". As an example, the JSON representation containing the conversion characters is as follows:

"audio_speak": {"text": "Hello, @master", "volume": "50%"}

When "audio_speak" is executed, "@master" can be replaced with the name of "owner".

In addition, in the above example, "volume" is set to a percentage, and the machine device 110 can calculate the specific parameters of the machine device 110 based on the percentage value of "volume". As another example, "volume" may also be represented as a specific parameter of machine device 110.

2. Let the machine device 110 play music

The behavior name is: audio_sound_music;

Behavior control parameters may include: path (path to play music, or file name, etc.), volume (volume of playing music), etc.

JSON is expressed as follows:

"audio_sound_music": {"path": "http//bpeer.com/happy.mp3", "volume": "50%"}

3. Let the machine device 110 play the prompt tone

The behavior name is: audio_sound_info;

Behavior control parameters include: name (the name of the tone to be played), volume (play volume), etc.

JSON is expressed as follows:

"audio_sound_info": {"name": "warning", "volume": "normal"}

4. Let the head movement of the machine unit 110

The behavior name is: motion_head;

Behavior control parameters may include: motor (motor that performs motion), velocity (motor speed), angle (motor motion angle), etc.

JSON is expressed as follows:

"motion_head": {"motor": "1", "velocity": "1", "angle": "45"}

In the above example, "velocity" is represented as a gear position, and the machine device 110 can calculate a specific "velocity" based on the gear position. In fact, "velocity" can also be expressed as a specific parameter of the head movement of the machine device 110.

In addition, in the above example, "angle" is represented as the angle of the motor, and actually, "angle" can be expressed as relative data such as percentage, for example, "angle": "50%", and the machine device 110 can determine according to the angle range The specific parameters, for example, the maximum angle is 180 degrees, then the specific angle is calculated to be 90 degrees, but is not limited thereto.

5. Let the neck of the machine device 110 move

The behavior name is: motion_neck;

JSON is expressed as follows:

"motion_neck": {"motor": "1", "velocity": "2", "angle": "60"}

6. Let the shoulder movement of the machine device 110

The behavior name is: motion_shoulder;

JSON is expressed as follows:

"motion_shoulder": {"motor": "1", "velocity": "3", "angle": "60"}

7. Let the elbow movement of the machine unit 110

The behavior name is: motion_elbow;

JSON is expressed as follows:

"motion_elbow": {"motor": "1", "velocity": "2", "angle": "50"}

8. Let the wrist of the machine device 110 move

The behavior name is: motion_wrist;

JSON is expressed as follows:

"motion_wrist": {"motor": "1", "velocity": "2", "angle": "50"}

9. Let the waist of the machine unit 110 move

The behavior name is: motion_waist;

JSON is expressed as follows:

"motion_waist": {"motor": "1", "velocity": "2", "angle": "50"}

10. Let the eye movement of the machine device 110

The behavior name is: motion_eye;

JSON is expressed as follows:

"motion_eye": {"motor": "1", "velocity": "2", "angle": "50"}

11. Let the machine device 110 display an expression

The behavior name is: display_emotion;

Behavior control parameters can include: content (displayed emoticons), velocity (display speed), etc.

JSON is expressed as follows:

"display_emotion": {"content": "happy", "velocity": "3"}

12, let the machine device 110 take pictures

The behavior name is: program_photo;

The behavior control parameters may include: flash (whether the flash is turned on) or the like (such as known camera control parameters, but are not limited thereto)

JSON is expressed as follows:

"program_photo": {"flash": "1"}

13. Let the machine device 110 control the television

The behavior name is: control_tv;

Behavior control parameters can include: state (eg open, close), etc.

JSON is expressed as follows:

"control_tv": {"state": "open"}

14. Let the machine unit 110 control the LED lights

The behavior name is: control_led;

Behavior control parameters can include: state (eg open, close), color, etc.

JSON is expressed as follows:

"control_led": {"state": "open", "color": "yellow"}

It should be understood that the basic behavior is merely illustrative and is not an exhaustive description of the basic behavior and the classification of the basic behavior. In fact, the basic behavior can use any behavior name, and the behavior control parameter can be any data. For example, for the mobile device 110 having mobility capabilities, basic behaviors related to mobility capabilities may also be defined, but are not limited thereto.

Control data

In some examples, the execution logic constraint can include at least a time logic constraint and/or an event logic constraint to cause at least the machine device 110 to generate one or a set of actions in accordance with time logic, and/or to generate an event corresponding to the event in response to the event. Or a set of actions, but not limited to this.

In some examples, the time logic constraint includes a time-dependent constraint. One or more basic actions may be included to start execution at the same time; or one or more basic actions may be performed after performing a predetermined time to perform one or more basic actions of the next time node; or after one or more basic behaviors are executed, execution is started One or more basic behaviors of the node at the next time; or a plurality of basic behaviors according to the "timeline" are distributed at corresponding time points on the timeline, and the corresponding basic behavior is performed at the time of arrival. It should be understood that the temporal logic constraint may be one or a combination of the above, as well as other time constraints, but is not limited thereto.

Based on the definition of the basic behavior described above, as an example, the control data may be a script containing execution logic of basic behavior and basic behavior, and the corresponding JSON control data may be expressed as follows ("//" is a commentary description of the corresponding item) :

Trigger:

[“0”: “motion_eye”: {“motor”: “1”, “velocity”: “2”, “angle”: “90”}, // eye opens “90” with speed “2” (eg 90 %);

"1": "display_emotion": {"content": "happy", "velocity": "3"}, / / display "happy" expression, and the display speed is "3";

"2": "audio_speak": {"text": "How are you", "volume": "50%"} / / Make the machine say "How are you?" The volume is 50%;

]

Among them, "0", "1" and "2" are time logic constraints of basic behavior, and these behaviors are sequentially executed from "0" to "1" to "2". The control data corresponds to: first open the eyes, then display the "happy" expression, and finally say "How are you?"

In some examples, the event logic constraint includes performing an action in response to the event. The actions of the event logic constraints may include interactive action flows. The control data of the interactive action flow may include a behavior frame and a logical control frame, the behavior frame includes a start behavior frame, and each logical control frame is connected to one or more previous behavior frames and one or more subsequent behavior frames, and the logical control frame A one or more control conditions and subsequent behavior frames corresponding to the one or more control conditions are included, wherein the behavioral frame corresponds to a basic behavior or a plurality of basic behaviors that are logically constrained by time.

As a non-limiting example, based on the definition of the basic behavior described above, the JSON control data of the interactive action flow can be as follows (the content comment description after "//").

{

Page_id: 1000; / / interactive action flow unique identifier;

Item:0{

Trigger:[

0:["audio_speak":{"text": "How are you", "volume": "50%"}];//Basic behavior "0", the basic behavior number is the chronological order of basic behavior execution; basic The behavior “0” is speaking, the content of the speech is set to “How are you?”, and the volume is set to “50%”;

1:["audio_sound_music":{"path":"http//bpeer.com/happy.mp3", "volume": "50%"}];//Basic behavior "1", here is playing music, Its parameters are set to the path "path" and the volume "volume";

2: ["motion_head": {"motor": "1", "velocity": "1", "angle": "45"}, "motion_neck": {"motor": "1", "velocity": "2", "angle": "60"}]; / / basic "2", contains two basic behaviors performed simultaneously, head movement and neck movement, head movement is set to be by head motor "1 "Execution, the speed of movement is "1" and the angle is "45" degrees, the neck movement is similar;

Behavior frame in //item0

Flow_map:[

Ifs{

Ear: "text": ["good", "I am very good", "you are good"]; / / control conditions for the conversation heard;

Goto:item 2;

}//When the content heard is "good", or "I am fine", or "you are good", enter "item2" to execute the corresponding behavior frame;

Ifs{

Ear: "text": ["good sad", "I am not good", "too sad"];

Goto:item 1;}

a logical control frame in /item0, which is connected to the trigger of item0 as its previous behavior frame, and is linked to the behavior frame in "item1" and "item2" based on the control condition as its subsequent behavior frame;

}; / / interactive action item "0", its behavior frame as the starting behavior frame;

Item:1{

Trigger:[

0: ["motion_eye": {"motor": "1", "velocity": "2", "angle": "50"}];

1:["audio_speak":{"text": "Don't be sad, how about telling a joke to you?", "volume": "50%"}];];

Flow_map:[

Ifs{

Ear: "text": ["good", "okay", "may"];

Goto:item 3;};

Ifs{

Ear: "text": ["No need", "Don't"];

Goto:item 4;}};

Item:2{

Trigger:[

0: ["display_emotion": {"content": "happy", "velocity": "3"}]; / / The basic action of the behavior frame "0", the basic behavior is to display the expression, the expression content is set to " Happy", and the speed at which the expression is displayed is set to "3"

];

End;

};// interactive action item "1";

Item:3{

Trigger:[

0: ["audio_speak": {"text": "joke", "volume": "50%"}];]

End;]};

Item:4{

Trigger:[

0:["motion_shoulder": {"motor": "1", "velocity": "3", "angle": "60"}];]

End;]};

}

FIG. 2 is a schematic structural view of the machine unit 110.

Referring to FIG. 2, the machine device 110 includes a sensor device 112, a communication unit 113, and a computing unit 114. Computing unit 114 can include one or more processors and storage devices and one or more meters stored in the storage device Computer program or instruction set. The storage device of computing unit 114 includes local storage devices and/or storage devices on the network, and storage devices on the network provide access to content stored by one or more processors.

The sensor device 112 of the machine device 110 can sense information external to the machine device 110, such as an item in the environment in which the machine device 110 is located, or an entity such as a human user, or a robot, or a computing device, or information of a physical environment such as temperature, or Humidity, or gas concentration, etc. The sensor device of machine device 110 may also sense information of machine device 110 itself, such as the amount of power of machine device 110, or the temperature of machine device 110, or the length of time of machine device 110, the posture state of machine device 110, and the like.

In some examples, sensor device 112 can include any sensor that can provide information, such as a microphone, or a camera, or a camera, or an infrared sensor, or a light sensor, or a GPS position sensor, or a gyroscope, motion detection sensor, or acceleration. One or any combination of sensors, tactile sensors, but not limited thereto, it should be understood that sensor device 112 may also include other sensors that provide information.

In some examples, sensor device 112 can include a software-based sensor that can provide high level or fine granularity of information. In addition to the raw sensor inputs, these sensor devices 112 process the appropriate input data to provide meaningful information, for example, the sensor device 112 can include the words and/or body gesture/moving identification module, facial recognition module, event detection module. and many more.

In some examples, sensor device 112 can include a voice recognition device that can recognize the audio to a corresponding text using known speech recognition techniques. The speech recognition device can be a local speech recognition module of the machine device 110, and the process of speech recognition is performed locally at the machine device 110. Or the voice recognition device may be a proxy with the remote voice recognition server for transmitting at least part of the information of the audio data to the remote voice recognition server, so that the remote voice recognition server recognizes the audio data to obtain text corresponding to the audio, and receives The text returned by the remote speech recognition server. It should be understood that the speech recognition apparatus is not limited to the above form, and virtually any speech recognition method can be employed.

In some examples, sensor device 112 can include an image recognition device that can employ at least some of the information captured by the camera of machine device 110 using known image recognition algorithms. The image recognition device may recognize the device device 110 locally, or may cooperate with the remote image recognition server to complete image recognition, but is not limited thereto. This example does not limit the image recognition device.

In some examples, sensor device 112 may also include an agent of sensors distributed in a physical environment, such as a static sensor 180 and/or a proxy for static sensor 181 as shown in FIG. 1, which may be associated with static sensor 180 and/or The static sensor 181 communicates to acquire data from the static sensor 180 and/or the static sensor 181, or to send instructions or the like to the static sensor 180 and/or the static sensor 181.

In some examples, computing unit 114 can include data fusion component 119 that can incorporate multi-sensor input data input by sensor device 112 or the like to obtain sensory data. For example, data fusion component 119 can generate perceptual data containing one or more data elements based on multi-sensor input data in accordance with a plurality of predefined data elements. The information perceived by the machine device 110 can thus be at least partially formatted into data for a plurality of data elements. Without limitation, data fusion component 119 and, when at least a portion of sensor device 112 senses a change in information, generates perceptual data containing one or more data elements based on the multi-sensor input data, but is not limited thereto.

As a non-limiting example, a plurality of data elements may be pre-set, it being understood that the setting of the exemplary data elements described below is not a division of data elements, or a number of data elements, or a definition of a data element, in fact any The division of data elements can all be considered. Examples of data elements are shown in Table 1.

Table 1 Data element example table

With the definition of the data elements in Table 1, an exemplary perceptual data is given below. It should be understood that the following perceptual data is not the number of elements of perceptual data, or the definition of perceptual data elements, or the format or perception of perceptual data. The definition of the way the data is expressed. The JSON-aware data of an example case is expressed as follows, but is not limited thereto, and other methods are also possible.

In the exemplary perceptual data, "vision_human_position" records that the human user is behind ("back") relative to the machine device 110, and "back" can also be represented by other characters, which can distinguish different positions, and should The position of understanding can also be expressed by "angle value", such as "vision_human_position": "45°" and the like. "sensing_touch" records the touch of the human user on the machine device 110. The position of the touch is a hand ("hand"), and the "hand" can also be represented by other characters, which can distinguish different positions, it should be understood The touch position can be multiple, and the value of "sensing_touch" can be an array that records multiple locations. "audio_speak_txt" records what the human user said "very happy to see you", and the content can also be audio data. "audio_speak_language" records the language "chinese" spoken by human users. "vision_human_posture" records the human user's gesture "posture1", and "posture1" can also be represented by other characters, which can distinguish different postures. "system_date" records the date "2016/3/16" of the generation of the perceptual data, and "system_time" records the time "13-00-00" of the perceptual data generation. "system_power" records the "80%" of the power of the machine unit 110, it being understood that the amount of power can also be identified in other ways.

As a non-limiting example, the condition data may be expressed as follows, it being understood that the condition data of the following examples is not a data element included in the condition data, or a format of the condition data, or a condition setting of the condition data, or other aspects of the condition data. The limitation is merely exemplified. In fact, various types of conditions can be set based on a plurality of predefined data elements to obtain condition data of each type, and the condition data is expressed by any expression. A JSON condition data is shown below, but is not limited thereto, and other methods are also possible.

In the condition data of the above example, the condition data includes the following conditions: 1) "vision_human_position": "back"", indicating that the human user is behind the machine device 110; 2) "sensing_touch": "hand"", indicating The human user touches the hand of the machine device 110; 3) "audio_speak_txt": "I am very happy to see you", indicating that the human user said "very happy to see you" to the machine device 110; and 4) "audio_speak_language ": "chinese"" means that the language spoken by human users is Chinese. The condition data indicates a scene when the human user is behind the machine device 110 and touches the hand of the machine device 110 and says "I am very happy to see you" in Chinese.

The condition data is associated with the control data, and the corresponding control data can be obtained at least by the condition data. The control data corresponding to the condition data may be a reaction to the scenario. For example, as a non-limiting example, the control data corresponding to the condition data may be:

"trigger": {

“0”: “motion_neck”: {“motor”: “1”, “velocity”: “2”, “angle”: “180”}, // neck motor “1” rotates at speed “2” “180” "degree

"1": "motion_eye": {"motor": "1", "velocity": "2", "angle": "90"}, / / eyes open "90" with speed "2" (for example 90% );

"2": "display_emotion": {"content": "happy", "velocity": "3"}, / / display "happy" expression, and the display speed is "3";

"3": "audio_speak": {"text": "hi, nice to meet you", "volume": "50%"}// Make the machine say "hi, nice to see you" volume is 50 %;

}

In the exemplary control data, there are several basic behaviors of “motion_neck”, “motion_eye”, “display_emotion”, and “audio_speak”, and the execution order of the basic behavior is defined in the control data. In this example, the head is turned first, then Open your eyes, then show the expression "happy", then say "hi, nice to meet you."

The computing unit 114 can include a scheduling component 117 that can schedule a plurality of components 118 for performing basic behavior based on the control data to cause the machine device 110 to generate an action. Each component 118 for performing basic behavior is used to perform a corresponding basic behavior. The scheduling component 117 can schedule a plurality of components 118 for performing basic behavior based on execution logic constraints of the basic behavior corresponding to the control data, and the scheduling component 117 can determine the basic behavior for performing the behavior name according to the behavior name in the control data. The component 118 is executed to perform the basic behavior, and the behavior control parameter is passed to the component 118 for performing the basic behavior such that the component 118 for performing the basic behavior performs the basic behavior in accordance with the behavior control parameter. The execution logic of the basic behavior may include time logic constraints and/or event logic constraints.

In some examples, the scheduling component 117 is further configured to convert control parameters in the base behavior to behavior control parameters of the component 118 for performing the base behavior based on the parameter conversion strategy. For example, the scheduling component 117 can convert the volume "80%" to a particular volume value "32" of the component 118 for performing the basic behavior, or convert the volume "comfort" to a specific volume value of the component 118 for performing the basic behavior. "32", but not limited to this. In fact, the behavior control parameters of the basic behavior in the control data can be converted into behavior control parameters of the component 118 for performing the basic behavior by a preset mapping relationship or algorithm or the like.

Taking the control data of the above example as an example, as a non-limiting example, the scheduling component 117 can determine that the control data is a plurality of basic behaviors of temporal logic constraints, and the scheduling component 117 determines the basic behavior as "neck motion" according to "motion_neck" ", and then dispatch the component 118 for "neck movement", which will be "motor", Behavioral control parameters such as "velocity", "angle" are passed to component 118 for "neck motion", and component 118 of "neck motion" performs actions based on behavioral control parameters such as "motor", "velocity", "angle", and the like. . When the component 118 of the "neck motion" performs the action completion, the execution completion status can be returned to the dispatch component 117. The scheduling component 117 determines that the component 118 for "eye movement" is scheduled to perform an action based on "motion_eye", and after the execution of the component 118 for "eye movement" is completed, the scheduling component 117 determines the schedule for "expression display" based on "display_emotion" The component 118 displays an expression, and the scheduling component 117 determines to schedule the component 118 for "talking" to speak based on "audio_speak."

The control data may correspond to an interactive action flow (including a plurality of basic behaviors constrained by event logic), and the scheduling component 117 may schedule a plurality of components 118 for performing basic behavior to cause the machine device 110 to generate an action based on the interactive action flow.

As a non-limiting example, taking the above interactive action flow as an example, "page_id" is a unique identifier of the interactive action flow, "item" is an interactive action item, and "item" has a unique identifier in the interactive action flow. "item" may include a behavior frame "trigger" and a logical control frame "flow_map", "trigger" may include one or more basic behaviors that are logically constrained by time, and "flow_map" may include one or more control conditions "ifs" and The subsequent behavior frame corresponding (marked with "goto" in this example) is referred to in this example by the unique identifier of the interactive action item "item". The end tag "end" may also be included in "item".

Upon entering the interactive action flow, the scheduling component 117 can determine the starting behavior frame based on the unique identification of the item and execute the starting behavior frame, in this example, the starting behavior frame is in "item0", executing in "item0" "trigger", the "trigger" includes four behaviors that are logically constrained by time, and the time logic is represented by numbers "0", "1", "2", and the like. The number "0" corresponds to the basic behavior, the content to be said is set to "How are you?", the volume of the speech is set to "50%", and the component that causes the machine device 100 to speak can be scheduled according to the set parameters. How are you". After the execution of the number "0" is completed, the number "1" is executed, and the component 118 for "speaking" returns the execution completion result after execution and starts execution of the number "1". The number "1" corresponds to the basic behavior of playing music, the music to be played is "http//bpeer.com/happy.mp3", the volume of the play is "50%", and the scheduling component 117 can call the machine device 110 to play music. The component plays "http//bpeer.com/happy.mp3". After the number "1" is executed, the number "2" is executed. The number "2" includes two basic behaviors performed simultaneously, namely head movement and neck movement, wherein the head movement is performed by the head motor "1", the movement speed is set to "1", and the movement angle is "45". "degree; neck movement is performed by the neck motor "1", the movement speed is set to "2", and the movement angle is "60" degrees.

In some examples, component 118 performing head motion and performing neck motion may determine a motor to be controlled based on the motor identification of the performance behavior, and machine device 110 may maintain a mapping table that will perform the behavior in the interactive motion stream. The motor identification is mapped to a motor corresponding to machine device 110. In addition, a mapping table can be maintained to map the motion speed to the motion speed performed by the motor. The motion speed in the interactive motion stream can be a speed gear. For example, “1” means slow speed, “2” means normal speed, “3”. Expressed quickly. The angle of motion can also be a relative value that machine device 100 can convert to a final value of execution. Of course, the motor is not limited to the actuator, and other controlled objects can be controlled in a similar manner.

After the initial behavior frame is executed, enter the logical control frame "flow_map" connected to it. In "item0", "flow_map" corresponds to two subsequent behavior frames based on the control condition. If you hear "good sad" or "I" Not good, or "too sad", enter "item 1" to perform the behavior frame in "item 1" and enter the logical control frame in "item 1"; if you hear "good", or "I am fine", or "you are good", enter "item2" to perform the behavior frame in "item 2" and enter the logical control frame in "item 2" to judge.

Referring to the above code, the behavior frame "trigger" of "item 1" includes two basic behaviors, firstly the "0" eye movement, which is performed by the eye motor "1", and the movement speed is "2". "50" degrees; then speak for "1", the content of the speech is "Don't be sad, how about telling a joke to you?", and the volume of the speech is "50%". Although the text of the spoken content is given in the example, in some examples, the audio data of the spoken content may also be directly given. After the basic behavior "1" is completed, it enters the logical control frame "flow_map" connected thereto, and the "flow_map" is connected to two subsequent behavior frames based on the control condition, which are behavior frames in "item3" and "item4", respectively. If the content heard in the control input is "good" or "okay" or "may", enter "item3" to execute the behavior frame; if the content heard in the control input is "not used" or " Don't go to "item4" to execute the behavior frame in it.

In the above example, "0: ["display_emotion": {"content": "happy", "velocity": "3"}]" defines the behavior of the machine device 110 displaying an expression, and the displayed expression is "happy" The machine device 110 can maintain a correspondence between an expression animation and an expression identifier, acquire an animation corresponding to the expression based on “happy”, and display the animation of the expression, but is not limited thereto.

Component 118 for performing basic behaviors can include software components and/or hardware components. The component 118 for performing the basic behavior can be scheduled by the scheduling component 117 and can provide one or more behavior control parameters, such as a call interface via a software interface.

As a non-limiting example, the component 118 for "speaking" may include a speech synthesis portion and an audio playback portion, the speech synthesis portion may synthesize audio corresponding to the text, and the speech synthesis portion may include local speech synthesis, or may be For the interface with the speech synthesis server, the text is transmitted to the speech synthesis server and the audio returned by the speech synthesis server is received. The audio playback portion includes an audio circuit and a speaker, and the like, and an audio processing program for converting the data signal into an electrical signal, transmitting the electrical signal as a speaker, the speaker generating the sound based on the electrical signal, and the audio processing program for the audio data. Perform decoding and so on. The component 118 for "talking" may also not include a speech synthesis portion, such as audio data may be included in the control data.

As a non-limiting example, the component 118 for "expression display" may include an expression acquisition portion and a display portion for acquiring a corresponding expression image (for example, an action composed of a multi-frame image) according to the expression control parameter, the display portion A program for causing the display to display content allows the display to display an emoticon image.

As a non-limiting example, component 118 for "eye movement" can include an eye executor and a command generation portion for generating a control command to the actuator based on the behavior control parameter, the actuator can include a motor , or a relay or the like that causes the actuator to produce motion.

It should be understood that the component 118 for performing the basic behavior is not limited to the above-described form, and the above examples are not limitations on the component 118 for performing basic behavior, and may actually include any type of component 118 for performing basic behavior. I will not repeat them here.

FIG. 3 is a schematic diagram of a system 300 for loading control data for a machine device 110.

Referring to FIG. 3, a system 300 for loading control data for a machine device 110 can include a search management component 310, a plurality of search modules 320 (shown as 320 ₁ - 320 _n in FIG. 3).

The search management component 310 is configured to provide the sensory data corresponding to the machine device 110 to the plurality of search modules 320. Each of the plurality of search modules 320 is configured to search for control data that matches the perceptual data from a different set of control data. The search management component 310 provides candidates for at least one control data from the plurality of search modules 320. The machine device 110 controls the machine device 110 to generate an action using the candidate of the control data.

As a non-limiting example, one of the plurality of search modules 320 is configured to search for control data matching the perceptual data from a set of control data of the local network to which the machine device 110 is local and/or connected to the machine device 110, multiple searches The other of the modules 320 is configured to search for control data matching the perceptual data from the control data set of the remote computer system, but is not limited thereto.

As a non-limiting example, one of the plurality of search modules 320 can be configured to search for control data that matches the perceptual data from the control data set of the machine device 110, and the other of the plurality of search modules 320 can be configured to Searching for control matching the sensory data from the control data set corresponding to the machine device version of the machine device 110 Data, yet another one of the plurality of search modules 320 is configured to search for control data that matches the perceptual data from a common set of control data, but is not limited thereto.

Without limitation, the control data set of machine device 110 may include user-configured control data for machine device 110, and the control data set for the machine device version of machine device 110 may include control data provided by the manufacturer of the machine device, multiple medium machine devices The general control data set may include control data suitable for a variety of machine devices, but is not limited thereto.

As a non-limiting example, the plurality of search modules 320 are configured to search for control data that matches the perceptual data from a set of control data data corresponding to the different modes. Without limitation, the different modes may correspond to different sets of control data, which may be divided in various ways, such as different scenes, different emotions, different personalities, different professional attributes, and the like, but are not limited thereto.

In some examples, the search management component 310 is configured to provide the perceptual data corresponding to the machine device 110 to the plurality of search modules 320, respectively, until one of the plurality of search modules 320 provides control data candidates. The search management component 310 can provide control data to the machine device 110, which can use the control control data of the search management component 310 to control the machine device 110 to generate actions.

In some examples, the search management component 310 is configured to provide the perceptual data to the plurality of search modules 320 in parallel, the search management component 310 selecting controls from the candidates of the control data provided by the plurality of search modules 320 based on a global algorithm. The data is used to obtain control data that controls the machine device 110 to generate an action. Without limitation, the global algorithm may select control data provided to the machine device 110 from among a plurality of candidates for control data data.

In some examples, search management component 310 is configured to maintain the priority of different control data sets or multiple search modules 320. Without limitation, the search management component 310 can be configured to provide the perceptual data to the plurality of search modules 320, respectively, according to the priority, until one of the plurality of search modules 320 provides a candidate for control data that matches the perceptual data item. Without limitation, the global algorithm of the search management component 310 can select control data provided to the machine device 110 from among a plurality of candidates for the control data data, depending at least on the priority of the different control data sets or the plurality of search modules 320.

In some examples, the multiple search modules 320 may use different algorithms or the same algorithm, which is not limited in this embodiment.

As a non-limiting example, the plurality of search modules 320 can be configured to match the perceptual data to the condition data to obtain control data that matches the perceptual data. In some examples, the sensing data and the condition data may include a plurality of data elements, and the plurality of data elements may be prioritized, the priority may be used to determine the degree of matching of the sensing data with the condition data, but is not limited thereto.

An example of loading control data for the machine device 110 will now be described.

4 is a schematic diagram of loading control data for machine device 110 locally and at a server.

Referring to FIG. 4, the machine device 110 includes a search management module 121 and a first search module 122, wherein the first search module 122 is configured to search and search for local control data sets on the local network connected to or connected to the machine device 110. The candidate of the control data matched by the sensor data of the machine device 110.

Server 140 includes a second search module 142. The search management module 121 can transmit the sensory data of the machine device 110 in the communication signal for receipt at the server 140. The server 140 can receive the perceptual data in the communication signal, and the second search module 142 can search the server control data set 143 for candidates for the control data that match the perceptual data. Server 140 may send a candidate for control data in the communication signal to be received at search management module 121 at machine device 110. The search management module 121 at the machine device 110 can receive candidates for control data provided by the second search module 142 at the server 140.

As a non-limiting example, the search management module 121 at the machine device 110 can provide the perceptual data to the first search module 122 and the second search module 142 at the server 140, respectively, until the first search module 122 or the second search One of the modules 142 provides control data candidates. In some examples, the search management module 121 can be configured to maintain a prioritization in which the perceptual data is provided to the first search module 122 and the second search module 142, respectively, but is not limited thereto.

As another non-limiting example, the search management module 121 at the machine device 110 can provide the perceptual data to the first search module 122 and the second search module 142 at the server 140 in parallel, the search management module 121 based on a global algorithm Control data for controlling the machine device 110 to generate an action is selected among the control data candidates provided by the first search module 122 and the second search module 142. In some examples, the global algorithm may be selected for control based on the priority of the first search module 122 and the second search module 142, and/or the degree of matching of the control data, and/or the time at which the data candidates are provided, and the like. The machine device 110 generates control data for the action. It should be noted that the example is merely illustrative, and the embodiment of the present invention is not limited thereto.

Figure 5 is a schematic diagram of loading control data from different control data ranges.

Referring to FIG. 5, in the example, three control data sets are included, as shown in FIG. 5, respectively, a first control data set 511, a second control data set 521, and a third control data set 531. The first search module 510 is configured to search the first control data set 511, and the second search module 520 is configured to search the second control data set 521, the third search The cable module 530 is used to search for the third control data set 531. It should be understood that in this example, more or better control data sets and/or search modules may be included, and FIG. 5 is not a limitation on the number.

In one example, the first control data set 511 can be a control data set of the machine device 110, and the control data set of the machine device 110 can be control data specific to the machine device 110, but is not limited thereto; the second control data set 521 The control data set of the machine device version of the machine device 110 may be different. The different device device versions may be set with different or the same control data set, which is not limited in this embodiment; the third control data set 531 may be a general control data set. It can be used by a variety of machine versions.

In this example, the search management module 500 can provide the perceptual data to the first search module 510, the second search module 520, and the third search module 530, respectively, until one of them provides a candidate for the control data. Without limitation, the search management module 500 can maintain a priority, and the search management module 500 can provide the perceived data to the first search module 510, the second search module 520, and the third search module 530, respectively, according to the priority. The first search module 510, the second search module 520, and the third search module 530 may correspond to different search algorithms, and may also correspond to the same search algorithm, which is not limited in this example. As an example, the order of priority from high to low is: a control data set of the machine device 110, a control data set corresponding to the machine device version of the machine device 110, and a general control data set.

In this example, the search management module 500 can provide the perceptual data to the first search module 510, the second search module 520, and the third search module 530 in parallel. The search management module 500 may select control data for controlling the machine device 110 to generate an action from the candidates of the control data provided by the first search module 510, the second search module 520, and the third search module 530 based on a global algorithm.

In another example, the first control data set 511, the second control data set 521, and the third control data set 531 may correspond to different modes, and the different modes correspond to different actions of the machine device. As an example, the first control data set 511 may correspond to a solitude mode, the second control data set 521 may correspond to an interaction mode, and the third control data set 531 may correspond to a sleep mode, but is not limited thereto. In the control data set corresponding to the different modes, the data elements in the corresponding condition data may be different, and the basic behavior in the control data may be different, but this embodiment does not limit this.

FIG. 6 is a schematic structural view of the machine unit 110.

Referring to Figure 6, machine device 110 includes a sensory device that senses an entity or environment, which can represent sources 202 _{1 -} 202 _m of data corresponding to possible input modules, the data of which is comprised of multi-sensor input data 206.

Although the sources 202 _{1 -} 202 ₄ may show specific examples of various sensor modules, these examples are not exhaustive of the potential configuration of the sensor modules, and other sensor modules 202 _m may include any number of sensor modules including sensing entity activity states, such as Infrared body sensors and more. Other input data that can be utilized include electronic ink from the device, touch, voice, body position/body language, facial expressions, brainwave computer input, keyboard, manipulation of physical interfaces (eg, gloves or tactile interfaces), and the like. Emotional sensing, such as facial expressions and facial color changes, temperature, grip pressure and/or other possible indications of emotions, can also be used as a viable input data.

In some examples, motion detection sensor module 202 ₁ provides various environmental and/or entity data, including physical movements and/or gestures, and the like. The speech recognition module 202 ₂ can parse the audio data into words and/or sentences according to the syntax of the speech. Alternatively, the voiceprint of the sound source, the direction of the sound source, and the like can be extracted from the audio data. Event detection device 202 ₃ can detect events of the environment and/or entity, providing event data for entities and/or environments. The facial recognition module 202 ₄ can detect and identify (human) faces in image data (eg, an image or a set of images) and/or video data (eg, video frames). The facial recognition module 202 ₄ can include hardware components and/or software components.

In some examples, data fusion component 204 can integrate multi-sensor input data 206 of sources 202 _{1 -} 202 _{m to} form one or more data elements based on at least a portion of the multi-sensor input data in accordance with a predefined plurality of data elements Perceptual data 212. Data fusion component 204 can update perceptual data 206 based on multi-sensor input data. The data elements can be referred to the data elements shown in Table 1, but are not limited thereto. The data fusion component 204 can form the perceptual data comprising one or more data elements based on at least a portion of the plurality of sensor input data in accordance with the predefined plurality of data elements as the multi-sensor input data 206 of the sources 202 _{1 -} 202 _m changes. 212. The sources 202 _{1 -} 202 _m may update the multi-sensor input data when the perceived information changes, causing the data fusion component 204 to generate the perceptual data, but is not limited thereto.

The search component 222 can search the control data 216 in the control data set 214 based on the perceptual data 212 to obtain control data 220 for controlling the machine device 110 to generate an action. Control data 216 in control data set 214 may include control data edited by a user of machine device 110, and/or control data executed by machine device 110, as well as other more control data, and the like.

The scheduling component 218 can schedule the component 210 for performing the basic behavior based on the control data 220 to cause the machine device 110 to generate an action. Referring to Figure 6, illustrates components for performing the basic behavior 210 ₁ -210 _m, the fundamental behavior for each component performs a corresponding executable basic behavior of the 210 ₁ -210 _m.

Transmission component 208 can transmit perceptual data 212 in the communication signal to be received at transmission module 141 at server 140, and/or at local user terminal 120 or remote user terminal 130. Transmission component 208 can receive control data 220 transmitted by server 140 and/or receive control data 220 transmitted by local user terminal 120 or remote user terminal 130. The scheduling component 218 can schedule the components 210 _{1 -} 210 _m for performing the basic behavior based on the control data 220 received by the transmission component 208 to cause the machine device 110 to generate an action.

In some examples, the search component 222 can cause the transmission component 208 to transmit the sensing data 212 to the server 140 to obtain the control data 220 corresponding to the sensing data 212 from the server 140 when the control data 220 corresponding to the sensing data 212 is not found. But it is not limited to this.

7 is a flow chart of a method of loading control data for machine device 110.

Referring to FIG. 7, the method includes steps 701 to 704.

In step 701, the sensory data of the machine device 110 is input.

Step 702, providing the sensing data to the plurality of search modules. Wherein each of the plurality of search modules is for searching for control data matching the perceptual data from a different set of control data, wherein the control data is for causing the machine device to generate an action, wherein the action comprises a behavior type and/or a motion type Actions.

Step 703, providing candidates for at least one control data from the plurality of search modules.

Step 704, using the candidate of the control data to control the machine device 110 to generate an action.

In one example, the perceptual data may be provided to a plurality of search modules in parallel, and candidates for control data from the plurality of search modules may also be selected based on a global algorithm. Without limitation, the parameters of the global algorithm include at least the priority of different sets of control data.

For other parts of the above method, refer to the description of the above example, and details are not described herein again.

The embodiment of the invention achieves the following technical effects: controlling the machine device to generate an action according to different control data sets, and improving the effect of controlling the machine device.

Obviously, those skilled in the art should understand that the above modules or steps of the embodiments of the present invention can be implemented by a general computing device, which can be concentrated on a single computing device or distributed in multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from The steps shown or described are performed sequentially, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated into a single integrated circuit module. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.

The above description is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various changes and modifications may be made to the embodiments of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Claims

A method for loading control data for a machine device, comprising:

Entering perceptual data corresponding to the machine device, wherein the perceptual data is generated according to information sensed by the machine device according to at least part of a predefined plurality of data elements;

Providing the perceptual data to a plurality of search modules, wherein each of the plurality of search modules is configured to search for control data matching the perceptual data from different sets of control data, wherein the control data is for Generating an action by the machine device, wherein the action comprises an action type and/or a motion type of action;

Providing candidates for at least one control data from the plurality of search modules;

The machine device is used to generate an action using the candidate of the control data.
The method of claim 1 wherein said perceptual data is provided to said plurality of search modules separately until one of said plurality of search modules provides said control data.
The method of claim 2, wherein the plurality of search modules or the different sets of control data are prioritized, the perceptual data being separately provided to the plurality according to the priority Search modules.
The method of claim 1 wherein said perceptual data is provided to said plurality of search modules in parallel, wherein said method further comprises selecting control data from said plurality of search modules based on a global algorithm Candidate.
A method according to claim 1 or 2, wherein one of said different sets of control data is stored locally in said machine device and/or in a local area network to which said machine device is connected.
The method of claim 5 wherein the other of the different sets of control data is stored in a remote computer system.
The method of claim 1 wherein each of said different sets of control data data corresponds to a different mode of the machine device.
The method according to claim 1 or 5 or 6, wherein said different control data set comprises a control data set of said machine device, or a control data set of said machine device corresponding machine device version, or more Any combination of control data sets common to machine devices.
The method of claim 8 wherein said different sets of control data are ordered by priority.
The method of claim 9, wherein the perceptual data is separately provided to the plurality of search modules until one of the plurality of search modules provides the control data;

The plurality of search modules are ordered according to priorities of the different sets of control data, and the perceptual data is separately provided to the plurality of search modules according to the priority.
The method of claim 8 wherein said perceptual data is provided to said plurality of search modules in parallel, wherein said method further comprises selecting control data from said plurality of search modules based on a global algorithm Candidate.
The method according to any one of claims 1 to 11, wherein the plurality of search modules search for control data based on condition data associated with control data, wherein the condition data is based on a location with which the sensory data is generated A plurality of data elements corresponding to a plurality of predefined data elements are generated.
The method according to claim 12, wherein a plurality of data elements in said condition data are at least partially ordered in order of priority, said plurality of search modules searching for said sensing data matching control according to said priority ranking The candidate for the data.
The method according to claim 12 or 13, wherein the plurality of search modules load candidate of control data matching the perceptual data based on the sensing data matching at least part of the data elements in the condition data item.
The method according to any one of claims 1 to 14, wherein the control data corresponds to a basic behavior or a plurality of basic behaviors having execution logic constraints, wherein each of the basic behaviors It can be directly executed by the machine device.
The method of claim 15 wherein each of said basic behaviors is defined by a behavior name and a behavior control parameter.
The method of any of claims 1 to 16, wherein the plurality of search modules correspond to different search algorithms.
A method for loading control data for a machine device, comprising:

Entering perceptual data corresponding to the machine device, wherein the perceptual data is generated according to at least part of a predefined plurality of data elements based on information sensed by the machine device;

Providing the perceptual data to a plurality of search algorithms to search for control data for causing a machine device to generate an action from a different control data range, wherein the different control data ranges include control data sets of the machine device, Or any combination of the control device data set of the machine device version or the control data set common to the plurality of machine devices, wherein the action comprises an action type and/or a motion type action;

Determining at least one candidate for control data based on the plurality of search algorithms;

The machine device is used to generate an action using the candidate of the control data.
The method of claim 18 wherein said perceptual data is separately provided to said plurality of search algorithms until one of said plurality of search algorithms provides said control data.
The method according to claim 19, wherein said plurality of search algorithms or said different control data ranges are ordered in order of priority, said sensing data being separately provided to said plurality according to said priority Search algorithms.
The method according to claim 20, wherein the order of priority from high to low is: a control data set of said machine device, a control data set of said machine device corresponding machine version, and a plurality of machines A set of control data common to the device.
The method of claim 18, wherein the perceptual data is provided to the plurality of search algorithms in parallel, wherein the method further comprises selecting control data from the plurality of search algorithms based on a global algorithm Candidate.
The method according to any one of claims 18 to 22, wherein the control data corresponds to a basic behavior or a plurality of basic behaviors having execution logic constraints, wherein each of the basic behaviors It can be directly executed by the machine device.
The method of claim 23 wherein each of said basic behaviors is defined by a behavior name and a behavior control parameter.
The method according to any one of claims 18 to 23, wherein the plurality of search algorithms search for control data based on condition data associated with control data, wherein the condition data is based on a location with which the sensory data is generated A plurality of data elements corresponding to a plurality of predefined data elements are generated.
An apparatus for loading control data for a machine device, comprising:

a module for inputting perceptual data corresponding to a machine device, wherein the perceptual data is generated based on at least a portion of a predefined plurality of data elements based on information sensed by the machine device;

a module for providing the perceptual data to a plurality of search modules, wherein each of the plurality of search modules is configured to search for control data matching the perceptual data from different sets of control data, wherein Control data is used to cause an action by the machine device, wherein the action comprises an action type and/or a motion type of action;

a module for providing candidates for at least one control data from the plurality of search modules;

A module for controlling the machine device to generate an action using a candidate for the control data.
The apparatus of claim 26, wherein the perceptual data is separately provided to the plurality of search modules until one of the plurality of search modules provides the control data.
The apparatus according to claim 27, wherein said plurality of search modules or said different control data sets are ordered in order of priority, said sensing data being separately provided to said plurality according to said priority Search modules.
The apparatus of claim 26, wherein the perceptual data is provided to the plurality of search modules in parallel, wherein the apparatus further comprises selecting control data from the plurality of search modules based on a global algorithm Candidate.
Apparatus according to claim 26 or 27, wherein one of said different sets of control data is stored locally in said machine means and/or in a local area network to which said machine means is connected.
30. Apparatus according to claim 30 wherein the other of said different sets of control data is stored in a remote computer system.
Apparatus according to claim 26 or 27 or 29 wherein each of said different sets of control data corresponds to a different mode of the machine means.
The apparatus of claim 26, wherein said different set of control data comprises a control data set of said machine device, or a control data set of said machine device corresponding machine version, or a plurality of machine means Any combination of control data sets.
The apparatus of claim 33 wherein said different sets of control data are ordered by priority.
The apparatus of claim 34, wherein the perceptual data is separately provided to the plurality of search modules until one of the plurality of search modules provides the control data;

The plurality of search modules are ordered according to priorities of the different sets of control data, and the perceptual data is separately provided to the plurality of search modules according to the priority.
The apparatus of claim 33, wherein the perceptual data is provided to the plurality of search modules in parallel, wherein the apparatus further comprises selecting control data from the plurality of search modules based on a global algorithm Candidate.
The apparatus according to any one of claims 26 to 36, wherein the plurality of search modules search for control data based on condition data associated with control data, wherein the condition data is based on a location with which the sensory data is generated A plurality of data elements corresponding to a plurality of predefined data elements are generated.
The apparatus according to claim 37, wherein a plurality of data elements in said condition data are at least partially ordered in order of priority, said plurality of search modules searching for said sensing data matching control according to said priority ranking The candidate for the data.
The apparatus according to claim 37, wherein said plurality of search modules load candidates of said control data matching control data based on said perceptual data matching at least a portion of said data elements.
The apparatus according to any one of claims 26 to 39, wherein the control data corresponds to a basic behavior or has a plurality of basic behaviors of performing logical constraints, wherein each of the basic behaviors is Directly executed by the machine device.
The apparatus of claim 40 wherein each of said basic behaviors is defined by a behavior name and a behavior control parameter.
The apparatus of any one of claims 26 to 41, wherein the plurality of search modules correspond to different search algorithms.
An apparatus for loading control data for a machine device, comprising:

a module for inputting perceptual data corresponding to a machine device, wherein the perceptual data is generated according to at least a portion of a predefined plurality of data elements based on information sensed by the machine device;

Means for providing the perceptual data to a plurality of search algorithms to search for control data for causing a machine device to generate an action from a different control data range, wherein the different control data ranges include the machine device Controlling the data set, or any combination of the control data set of the machine device corresponding machine version, or the control data set common to the plurality of machine devices, wherein the action comprises an action type and/or a motion type of action;

a module for determining candidates for at least one piece of control data based on the plurality of search algorithms;

A module for controlling the machine device to generate an action using a candidate for the control data.
The apparatus of claim 43, wherein the perceptual data is separately provided to the plurality of search algorithms until one of the plurality of search algorithms provides the control data.
The apparatus according to claim 44, wherein said plurality of search algorithms or said different control data ranges are ordered in order of priority, said sensing data being separately provided to said plurality according to said priority Search algorithms.
The apparatus according to claim 45, wherein said order of priority from high to low is: a control data set of said machine device, a control data set of said machine device corresponding machine version, and a plurality of machines A set of control data common to the device.
The apparatus of claim 43, wherein the perceptual data is provided to the plurality of search algorithms in parallel, wherein the apparatus further comprises selecting control data from the plurality of search algorithms based on a global algorithm Candidate.
The apparatus according to any one of claims 43 to 47, wherein the control data corresponds to a basic behavior or a plurality of basic behaviors having execution logic constraints, wherein each of the basic behaviors It can be directly executed by the machine device.
The apparatus of claim 48 wherein each of said basic behaviors is defined by a behavior name and a behavior control parameter.
The apparatus according to any one of claims 43 to 48, wherein the plurality of search algorithms search for control data based on condition data associated with control data, wherein the condition data is based on a location with which the sensory data is generated A plurality of data elements corresponding to a plurality of predefined data elements are generated.