WO2017000785A1

WO2017000785A1 - System and method for training robot

Info

Publication number: WO2017000785A1
Application number: PCT/CN2016/085910
Authority: WO
Inventors: 蔡明峻
Original assignee: 芋头科技（杭州）有限公司
Priority date: 2015-06-30
Filing date: 2016-06-15
Publication date: 2017-01-05
Also published as: TW201700237A; CN106313113A; TWI594857B; HK1231439A1; CN106313113B

Abstract

A system for training a robot, comprising: a collecting unit (3) used for collecting conditional data; a fetching unit (2) used for fetching an operation instruction transmitted by a user; a storage unit (5) used for storing the operation instruction; an operating unit (1) used for executing an operation corresponding to an edit instruction; a processing unit (4) connected respectively to the collecting unit (3), the fetching unit (2), the storage unit (5), and the operating unit (1) and used for storing the operation instruction in the storage unit (5) and controlling the operating unit (1) and/or the collecting unit (3) to execute an operation corresponding to the operation instruction; and a display unit (6) connected to the processing unit (4) and comprising a preset template. This system is structurally simple and facilitates user observation. Also provided is a method for training the robot.

Description

System and method for training robot

Technical field

The present invention relates to the field of robots, and more particularly to a system and method for training a robot through an image interface.

Background technique

At present, the training method for robots is limited to modifying the program of the robot by writing code. For users who do not understand the programming language, the training method is difficult to write, and requires the user to have professional programming development skills and excellent code. Management literacy, have a deep understanding of the bottom of the system, can find problems. Due to the complexity of the process of writing code, it is prone to training logic errors, so the current training methods of robots can not meet the needs of users who do not understand programming languages.

Summary of the invention

In view of the above problems existing in the training methods of existing robots, there is now provided a system and method for implementing user behavior training that can support the user without a programming development foundation.

The specific technical solutions are as follows:

A system for training robots, including:

An acquisition unit for collecting condition data;

An obtaining unit, configured to acquire an operation instruction sent by the user, each of the operation instructions corresponding to a corresponding operation, where the operation instruction includes an acquisition instruction and/or an editing instruction;

a storage unit for storing the operation instruction;

An operation unit, configured to execute the operation corresponding to the editing instruction;

a processing unit, which is respectively connected to the collection unit, the acquisition unit, the storage unit and the operation unit, for storing the operation instruction in the storage unit, and controlling the operation unit and/or The collecting unit performs the operation corresponding to the operation instruction;

A display unit is connected to the processing unit, and the display unit includes a preset template, the preset template includes an area for displaying the condition data and an area for displaying an execution image of the operation.

Preferably, the condition data comprises: a light value, and/or a humidity value, and/or a temperature value, and/or pyroelectric data, and/or a PM2.5 value, and/or a carbon dioxide content, and/or time. , and / or user attributes.

Preferably, the user attributes include a manager and/or a visitor.

Preferably, the obtaining unit comprises:

The voice recognition module is configured to receive a voice operation instruction sent by the user, and convert the voice operation instruction into a text operation instruction.

Preferably, the processing unit comprises:

a receiving module, configured to receive the condition data and the operation instruction;

An identification module is connected to the receiving module, configured to identify whether the operation instruction is the collection instruction, and output a recognition result;

If the operation instruction is the collection instruction, controlling an operation state of the collection unit according to the collection instruction;

An extraction module, connected to the identification module, when the operation instruction is not the acquisition instruction, the operation instruction is the editing instruction, and the extraction module is configured to extract the operation key in the editing instruction Word and reference condition data, each set of the operation keywords corresponds to one of the operations, and outputs the operation.

Preferably, the processing unit further includes:

a matching module, which is respectively connected to the receiving module and the extraction module, for using the condition Matching the data with the reference condition data to obtain a matching result;

When the results match, the operation unit performs the operation corresponding to the editing instruction.

A method of training a robot for controlling the operation of the robot includes the following steps:

S1. Collecting condition data;

S2. Obtain an operation instruction sent by the user, where each operation instruction corresponds to a corresponding operation, where the operation instruction includes an acquisition instruction and/or an edit instruction;

S3. storing the operation instruction;

S4. Control the robot to perform the operation corresponding to the operation instruction, and display the condition data and the execution image of the operation in a preset template.

Preferably, the step S4 includes:

S41. Receiving the condition data and the operation instruction;

S42. Identify whether the operation instruction is the collection instruction, and output a recognition result;

If the operation instruction is the collection instruction, stop acquiring the condition data or start collecting the condition data according to the collection instruction;

If the operation instruction is not the acquisition instruction, the operation instruction is the editing instruction, and the operation keyword and the reference condition data in the editing instruction are extracted, and each group of the operation keyword corresponds to one The operation and outputting the operation.

Preferably, the step S4 further includes:

S43. Match the condition data with the reference condition data to obtain a matching result;

When the matching results match, controlling the robot to perform the operation corresponding to the editing instruction;

When the matching results do not match, the process returns to step S1.

The beneficial effects of the above technical solutions:

In the technical solution, in the system for training the robot, the operation instruction sent by the user is acquired by the acquisition unit, the operation instruction is stored in the storage unit by the processing unit, and the control operation unit and/or the acquisition unit is executed to execute the operation instruction. Operation, display condition data through the display unit And the execution image of the operation, the user can train the robot in an intuitive manner, and the training process is simple, and the user who does not have the basis of programming development can be supported to participate in the behavior training of the robot. In the method of training the robot, the robot can be controlled to perform an operation corresponding to the operation instruction, and display the condition data and the execution image of the operation in a preset template, the operation process is simple, and the user can intuitively understand the training process.

DRAWINGS

1 is a block diagram of an embodiment of a system for training a robot according to the present invention;

2 is a flow chart of a method of an embodiment of a method for training a robot according to the present invention;

3 is a schematic diagram of an embodiment of the display unit in a normal state according to the present invention;

4 is a schematic diagram of an embodiment of the display unit in an edit state according to the present invention.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

It should be noted that the embodiments in the present invention and the features in the embodiments may be combined with each other without conflict.

The invention is further illustrated by the following figures and specific examples, but is not to be construed as limiting.

As shown in Figure 1, a system for training a robot includes:

An acquisition unit 3, configured to collect condition data;

An obtaining unit 2 is configured to acquire an operation instruction sent by the user, where each operation instruction corresponds to a corresponding operation, and the operation instruction includes an acquisition instruction and/or an editing instruction;

a storage unit 5 for storing an operation instruction;

An operation unit 1 configured to perform an operation corresponding to the editing instruction;

A processing unit 4 is connected to the acquisition unit 3, the acquisition unit 2, the storage unit 5 and the operation unit 1, respectively, for storing the operation instruction in the storage unit 5, and controlling the operation unit 1 and/or the acquisition unit 3 to execute the operation instruction Corresponding operation;

A display unit 6 is connected to the processing unit 4. The display unit 6 includes a preset template including an area for displaying condition data and an area for displaying an execution image of the operation.

Further, the acquiring instruction includes stopping the collecting instruction and starting the collecting instruction; the collecting instruction is for controlling the working state of the collecting unit 3.

As shown in FIG. 3 and FIG. 4, in the system of the embodiment, it is embedded in the robot system in actual application, and the robot can be trained when the robot system is in the integrated development environment. The acquisition unit 3 collects the condition data at a predetermined time interval, and the display unit 6 displays the collected condition data in real time in the display area of the condition data, and the user can input the stop acquisition instruction, specifically: a method of starting the edit mode by triggering (such as clicking “coagulation” The "button" causes the robot to enter a solid state. In this state, all the acquisition condition data of the robot is not updated, and is solidified at the triggering moment, and the user performs editing in the operation execution image of the display area of the display unit 6 in the solidified state, and the operation is performed. The image can be associated with the corresponding operation action corresponding to the editing instruction. After the user completes the editing, the user can input the start acquisition instruction. Specifically, the editing action can be uploaded to the robot by triggering the end of the editing mode (clicking the "release" button). (Operation unit 1) is up and running, and the data in the robot continues to be updated in real time.

In this embodiment, the operation instruction sent by the user is acquired by the acquisition unit 2, the operation instruction is stored in the storage unit 5 by the processing unit 4, and the operation unit 1 and/or the acquisition unit 3 are controlled to perform an operation corresponding to the operation instruction. The display unit 6 displays the condition data and the execution image of the operation, so that the user can train the robot in an intuitive manner, and the training process is simple, and can support no programming development. The basic users participate in the behavioral training of the robot.

As shown in Figures 3 and 4, in a preferred embodiment, the condition data includes: light values, and/or humidity values, and/or temperature values, and/or pyroelectric data, and/or PM2.5 values. , and / or carbon dioxide (CO ₂ ) content, and / or time, and / or user attributes.

Further, the collecting unit 3 may include: a light sensor for collecting light values; a humidity sensor for collecting humidity values; a pyroelectric infrared sensor for collecting whether there is a user near the robot; and a PM2.5 collecting module for Collecting PM2.5 values; carbon dioxide sensors for collecting CO ₂ content; and an authentication module for identifying user attributes.

In a preferred embodiment, the user attributes include a manager and/or a visitor.

In this embodiment, the identity verification module can be used to identify that the user is a manager or a visitor. If the administrator can directly train the robot, if the visitor can acquire the relationship with the administrator, the corresponding behavior is performed on the robot according to the corresponding authority. operating.

In a preferred embodiment, the obtaining unit 2 may include:

The voice recognition module 42 is configured to receive a voice operation instruction sent by the user, and convert the voice operation instruction into a text operation instruction.

In this embodiment, a natural language processing module of artificial intelligence may be further included for converting the text operation instruction into digital data for the robot to recognize.

In a preferred embodiment, the processing unit 4 can include:

a receiving module 41, configured to receive condition data and an operation instruction;

An identification module 42 is connected to the receiving module 41 for identifying whether the operation instruction is an acquisition instruction, and outputting the recognition result;

If the operation instruction is an acquisition instruction, controlling the working state of the collection unit 3 according to the acquisition instruction;

An extraction module 44 is connected to the identification module 42. When the operation instruction is not an acquisition instruction, the operation instruction is an editing instruction, and the extraction module 44 is configured to extract an operation keyword and reference condition data in the editing instruction, and each group of operation keywords corresponds to An operation and output operation.

In this embodiment, the operation may be to turn on the phone function to make a call to a pre-stored contact in the robot, or to activate the prompt tone function, or to control the LED light on the robot to be turned on/off, or to enable the video player function, or to turn on the music. The player function, or open the search engine function to search for keywords that the user needs to search, or turn on the TTS (Text To Speech) voice module function to convert the text into a voice function.

In a preferred embodiment, the processing unit 4 further includes:

a matching module 43 is respectively connected to the receiving module 41 and the extracting module 44 for matching the condition data with the reference condition data to obtain a matching result;

When the results match, the operation unit 1 performs an operation corresponding to the editing instruction.

In the embodiment, the operation keyword may be pre-stored in the storage unit 5, and when the matching result is matched and the operation keyword in the received editing instruction is the same as the pre-stored operation keyword, the control robot performs the corresponding operation.

As shown in FIG. 2, a method for training a robot for controlling the operation of the robot includes the following steps:

S1. Collecting condition data;

S2. Obtain an operation instruction sent by the user, where each operation instruction corresponds to a corresponding operation, and the operation instruction includes an acquisition instruction and/or an editing instruction;

S3. storing operation instructions;

S4. The control robot performs an operation corresponding to the operation instruction, and displays the condition data and the execution image of the operation in a preset template.

Further, the acquiring instruction includes stopping the acquiring instruction and starting the collecting instruction; and the collecting instruction is used to control whether the robot performs condition data collecting.

In this embodiment, the robot can be controlled to perform an operation corresponding to the operation instruction, and display the condition data and the execution image of the operation in a preset template, the operation process is simple, and the user can intuitively understand the training process.

In a preferred embodiment, step S4 may include:

S41. Receiving condition data and an operation instruction;

S42. Identifying whether the operation instruction is an acquisition instruction, and outputting the recognition result;

If the operation instruction is an acquisition instruction, stop acquiring the condition data or start collecting the condition data according to the acquisition instruction;

If the operation instruction is not an acquisition instruction, the operation instruction is an edit instruction, and the operation keyword and the reference condition data in the edit instruction are extracted, and each group of operation keywords corresponds to an operation, and the operation is output.

In a preferred embodiment, step S4 further includes:

When the matching results match, the control robot performs an operation corresponding to the editing instruction;

When the matching results do not match, the process returns to step S1.

As shown in FIG. 3 and FIG. 4, for example, when the training robot needs to prompt the user to turn on the air conditioner when the temperature is lower than 16 ° C, the robot system is in an integrated development environment, and the real-time condition state of the robot is displayed on the condition data of the preset template. Area; if the temperature is 20 °C at this time, click the “Coagulation” button (stop the acquisition command) to stop the real-time condition status, and display 20°C in the “Temperature” column (the condition data display area of the preset template). The temperature range can be manually set to 10 ° C to 16 ° C; activate the TTS voice module in the operation operation display area of the preset template, enter the text "temperature below 16 degrees Celsius, please turn on the air conditioner" (editing instructions); click "release "Button (start acquisition command), the automatically generated code is uploaded to the robot and updated to run. Thus completing a robotic behavior training Practice, when the temperature collected by the robot is lower than 16 °C, it will automatically remind the user to turn on the air conditioner.

The above is only a preferred embodiment of the present invention, and is not intended to limit the scope of the embodiments and the scope of the present invention, and those skilled in the art should be able to Alternatives and obvious variations are intended to be included within the scope of the invention.

Claims

A system for training a robot, comprising:

An acquisition unit for collecting condition data;

An obtaining unit, configured to acquire an operation instruction sent by the user, each of the operation instructions corresponding to a corresponding operation, where the operation instruction includes an acquisition instruction and/or an editing instruction;

a storage unit for storing the operation instruction;

An operation unit, configured to execute the operation corresponding to the editing instruction;

a processing unit, which is respectively connected to the collection unit, the acquisition unit, the storage unit and the operation unit, for storing the operation instruction in the storage unit, and controlling the operation unit and/or The collecting unit performs the operation corresponding to the operation instruction;

A display unit is connected to the processing unit, and the display unit includes a preset template, the preset template includes an area for displaying the condition data and an area for displaying an execution image of the operation.
A system for training a robot according to claim 1, wherein said condition data comprises: a light value, and/or a humidity value, and/or a temperature value, and/or pyroelectric data, and/or PM2. .5 value, and / or carbon dioxide content, and / or time, and / or user attributes.
A system for training a robot as claimed in claim 2, wherein said user attributes comprise a manager and/or a visitor.
The system for training a robot according to claim 1, wherein the obtaining unit comprises:

The voice recognition module is configured to receive a voice operation instruction sent by the user, and convert the voice operation instruction into a text operation instruction.
The system for training a robot according to claim 1, wherein the processing unit comprises:

a receiving module, configured to receive the condition data and the operation instruction;

An identification module, connected to the receiving module, to identify whether the operation instruction is the Set instructions and output recognition results;

If the operation instruction is the collection instruction, controlling an operation state of the collection unit according to the collection instruction;

An extraction module, connected to the identification module, when the operation instruction is not the acquisition instruction, the operation instruction is the editing instruction, and the extraction module is configured to extract the operation key in the editing instruction Word and reference condition data, each set of the operation keywords corresponds to one of the operations, and outputs the operation.
The system for training a robot according to claim 5, wherein the processing unit further comprises:

a matching module, configured to respectively connect the receiving module and the extracting module, to match the condition data with the reference condition data to obtain a matching result;

When the results match, the operation unit performs the operation corresponding to the editing instruction.
A method of training a robot, characterized in that, for controlling the operation of the robot, the following steps are included:

S1. Collecting condition data;

S2. Obtain an operation instruction sent by the user, where each operation instruction corresponds to a corresponding operation, where the operation instruction includes an acquisition instruction and/or an edit instruction;

S3. storing the operation instruction;

S4. Control the robot to perform the operation corresponding to the operation instruction, and display the condition data and the execution image of the operation in a preset template.
The method of training a robot according to claim 7, wherein the step S4 comprises:

S41. Receiving the condition data and the operation instruction;

S42. Identify whether the operation instruction is the collection instruction, and output a recognition result;

If the operation instruction is the collection instruction, stop acquiring the condition data or start collecting the condition data according to the collection instruction;

If the operation instruction is not the acquisition instruction, the operation instruction is the editing instruction, and the operation keyword and the reference condition data in the editing instruction are extracted, and each group of the operation keyword corresponds to one The operation and outputting the operation.
The method of training a robot according to claim 8, wherein the step S4 further comprises:

S43. Match the condition data with the reference condition data to obtain a matching result;

When the matching results match, controlling the robot to perform the operation corresponding to the editing instruction;

When the matching results do not match, the process returns to step S1.