WO2017177444A1

WO2017177444A1 - Cloud robot system, robot, and robot cloud platform

Info

Publication number: WO2017177444A1
Application number: PCT/CN2016/079438
Authority: WO
Inventors: 王斌; 黄晓庆
Original assignee: 深圳前海达闼云端智能科技有限公司
Priority date: 2016-04-15
Filing date: 2016-04-15
Publication date: 2017-10-19
Also published as: CN107708938B; CN107708938A

Abstract

A cloud robot system, comprising a robot body (101) and a robot cloud platform (102). The robot body (101) comprises an acquisition module (1011), a first judgment module (1012), and a sending module (1013). The robot cloud platform (102) comprises a determining module (1021), a cloud robot module (1022), and an agent module (1023). The acquisition module (1011) is used for acquiring environmental information; the first judgment module (1012) is used for determining whether the complexity of the environmental information exceeds the processing range of the robot body (101); the sending module (1013) is used for sending the environmental information to the robot cloud platform (102) when the complexity of the environmental information exceeds the processing range of the robot body (101); the determining module (1021) is used for determining the cloud robot module (1022) and/or the agent module (1023) to process the environmental information; the determined cloud robot module (1022) and/or agent module (1023) are used for processing the environmental information and sending an instruction to the robot body (101). A cloud robot and an agent can assist in processing when a robot terminal cannot perform processing, so that user experience is improved. Also disclosed are a robot (101) and a robot cloud platform (102).

Description

Cloud robot system, robot and robot cloud platform

Technical field

The invention relates to the field of intelligent robot technology, in particular to a cloud robot system, a robot and a robot cloud platform.

Background technique

The cloud robot is an intelligent robot that puts the cognitive system in the cloud, the body, the drive and the sensor are placed on the robot body, and connects the two through mobile communication; the cloud robot is the development direction of the intelligent humanoid robot.

At present, the cloud robot solution generally adopts a combination of a terminal robot body and a cloud robot module. The terminal robot body is mainly responsible for interaction with the user and the surrounding environment, such as: voice interaction, image acquisition, service provision, etc. The cloud robot module is mainly responsible for complex intelligent processing, such as: speech recognition, speech understanding, image recognition, and the like.

Since the existing cloud robot solution relies on the cloud robot module for intelligent processing, if the cloud robot module has a busy work, a fault, and the like, the quality of service to the user cannot be guaranteed.

The disadvantages of the prior art are:

The existing cloud robot implementation scheme may have a situation in which the quality of service to the user cannot be guaranteed, resulting in a reduced user experience.

Summary of the invention

The embodiment of the present application provides a cloud robot system, a robot, and a robot cloud platform, so as to solve the technical problem that the cloud robot implementation solution in the prior art may not guarantee the quality of service to the user, resulting in a reduced user experience.

In a first aspect, an embodiment of the present application provides a cloud robot system, including a robot And a robot cloud platform, the robot body includes an acquisition module, a first determination module, and a transmission module, and the robot cloud platform includes a determination module, a cloud robot module, and a manual seat module, wherein

The collecting module is configured to collect environmental information;

The first determining module is configured to determine whether the complexity of the environment information exceeds a processing range of the robot body;

The sending module is configured to send the environment information to the robot cloud platform when the complexity of the environment information exceeds a processing range of the robot body;

The determining module is configured to determine the cloud robot module and/or the manual agent module to process the environment information;

The determined cloud robot module and/or manual agent module is configured to process the environment information and send an instruction to the robot body.

The benefits are as follows:

Due to the cloud robot system provided by the embodiment of the present application, the intervention of the third-party artificial seat module is added to the existing artificial intelligence robot implementation solution, and the complexity of the processing environment information can be predicted by the robot body first, if the robot The ontology cannot process the environment information and then processed by the robot cloud platform. After the terminal side robot body sends the collected environment information to the robot cloud platform, the robot cloud platform may first determine that the cloud robot module processes the environment information. The environment information is processed by the manual agent module, or the environment information is jointly processed by the cloud robot module and the manual agent module, and finally determined by the cloud robot module and/or the artificial seat module. The environment information is processed and sent to the robot body on the terminal side, and remotely controlled by the robot cloud platform to ensure the quality of service to the user and improve the user experience.

In a second aspect, an embodiment of the present application provides a robot, including:

An acquisition module for collecting environmental information;

a first determining module, configured to determine whether the complexity of the environment information exceeds a processing range of the robot body;

a sending module, configured to send the environment information to the robot cloud platform when a complexity of the environment information exceeds a processing range of the robot body;

The receiving module is configured to receive an instruction sent by the cloud robot module of the robot cloud platform or the artificial seat module.

The benefits are as follows:

The robot provided by the embodiment of the present application first performs complexity prediction after collecting the environmental information, and if it is determined that the complexity of the environmental information exceeds the processing range of the robot body, the collected environmental information is Sending to the robot cloud platform, receiving a cloud robot module of the robot cloud platform, or an instruction sent by the manual agent module, that is, remotely controlling the robot body by the robot cloud platform when the robot body cannot process the environment information, and the robot The cloud platform includes two control modes, namely, cloud robot and manual seat, which solves the technical problem that the cloud robot module is unable to provide services or provide poor service quality if the cloud robot module is busy or faulty, and ensures better service for the user. , improved user experience.

In a third aspect, the embodiment of the present application provides a robot cloud platform, including: an information receiving module, a determining module, a cloud robot module, and a manual seating module, where

The information receiving module is configured to receive environment information sent by the robot body;

The benefits are as follows:

The robot cloud platform provided by the embodiment of the present application may determine, after receiving the environment information sent by the terminal-side robot body, that the cloud robot module processes the The environmental information is still processed by the manual agent module, or the environment information is jointly processed by the cloud robot module and the manual agent module, and then determined by the cloud robot module and/or artificial The agent module processes the environment information and sends an instruction to the robot body on the terminal side, and remotely controls the terminal side robot body in two ways by introducing a manual seat, thereby ensuring the service quality of the terminal side robot body to the user, and thereby improving user experience.

DRAWINGS

Specific embodiments of the present invention will be described below with reference to the accompanying drawings, in which:

1 is a schematic structural view 1 of a cloud robot system in Embodiment 1 of the present application;

FIG. 2 is a second schematic structural diagram of a cloud robot system according to Embodiment 1 of the present application;

FIG. 3 is a schematic structural diagram of a cloud robot system according to Embodiment 1 of the present application;

FIG. 4 is a schematic diagram showing changes in processing state in the first embodiment of the present application;

FIG. 5 is a schematic structural view 1 of the robot in the second embodiment of the present application;

6 is a schematic structural view 2 of the robot in the second embodiment of the present application;

FIG. 7 is a schematic structural diagram 1 of a robot cloud platform in Embodiment 3 of the present application;

FIG. 8 is a schematic structural diagram 2 of the robot cloud platform in the third embodiment of the present application.

detailed description

The embodiments of the present invention are further described in detail with reference to the accompanying drawings, in which FIG. An exhaustive example. And in the case of no conflict, the features in the embodiments and the embodiments in the description can be combined with each other.

The inventor noticed during the invention:

At present, the robot body realizes interaction with customers through various interaction methods and sensors, and the perception of the surrounding environment, with a certain degree of intelligence, can handle most of the less complicated The scene (for example: Google's driverless car, most of the time can automatically drive, may account for 99%), basically meet the needs of use.

However, in some and special scenarios (for example, Google's unmanned car, when the signal light is faulty, there is a police or pedestrian gesture to direct traffic, this scene may only account for 1%), the robot body can not handle.

The drawback of the current robotic solution is that in some cases, the robotic body cannot be processed by relying on the local capabilities of the robot.

In view of the above deficiencies, the embodiment of the present invention provides a cloud robot system, a robot, and a robot cloud platform. When the robot body cannot process, the cloud robot and/or a third-party artificial agent are introduced to remotely intervene, through a third-party artificial seat and the cloud. The robots participate in the control decision together to solve the technical problem that the inability of the robot body to provide services to the user in the prior art.

In order to facilitate the implementation of the present application, the cloud robot system, the robot and the robot cloud platform proposed by the present application will be described below in conjunction with specific embodiments.

FIG. 1 is a schematic structural diagram 1 of a cloud robot system according to Embodiment 1 of the present application. As shown in the figure, the cloud robot system may include a robot body 101 and a robot cloud platform 102, and the robot body includes an acquisition module 1011. a judging module 1012 and a sending module 1013, the robot cloud platform includes a determining module 1021, a cloud robot module 1022, and a human agent module 1023, where

The collecting module 1011 is configured to collect environmental information.

The first determining module 1012 is configured to determine whether the complexity of the environment information exceeds a processing range of the robot body;

The sending module 1013 is configured to send the environment information to the robot cloud platform when the complexity of the environment information exceeds a processing range of the robot body;

The determining module 1021 is configured to determine the cloud robot module and/or the artificial agent a module to process the environmental information;

The determined cloud robot module 1022 and/or the artificial agent module 1023 is configured to process the environment information and send an instruction to the robot body 101.

In a specific implementation, the robot body may be a mobile robot structure (such as a housekeeping service robot, a medical robot, a blind robot, etc.), or may be a guide blind helmet, a guide cane, or the like.

The collection module in the robot body may be used to collect environment information, and the environment information may be information such as a sound of a surrounding environment, a voice input by a user, an image or a video of a surrounding environment. The sending module in the robot body can transmit information collected by the collecting module (such as voice, voice, image or video, etc.) to the robot cloud platform via a network in a wireless or wired manner, and is processed by the robot cloud platform.

The robot cloud platform may include a cloud robot module and a manual seat module. In a specific implementation, the cloud robot module and the artificial seat module may both analyze and process environmental information collected by the terminal side robot body, and generate instruction feedback. The ability to give the terminal side robot body. Specifically, it is processed by the cloud robot module, processed by the manual agent module, or processed by the cloud robot module and the manual agent module, and can be determined by the determining module.

Finally, the module determined by the determining module (the cloud robot module and/or the manual agent module) processes the environmental information and sends an instruction to the robot body.

In the embodiment of the present application, a third-party artificial agent module may be introduced in an existing cloud robot implementation solution, where the cloud cloud platform determines whether the environment information is processed by the cloud robot module, or is processed by the manual agent module. The environmental information is processed by the cloud robot module and the manual agent module together, and the quality of service to the user is ensured by remote manipulation of the robot cloud platform.

Due to the cloud robot system provided by the embodiment of the present application, the existing artificial intelligence robot In the implementation scheme, the intervention of the third-party artificial seat module is added, and the complexity of the processing environment information can be predicted by the robot body first. If the robot body cannot process the environment information and then processed by the robot cloud platform, the terminal-side robot body will After the collected environment information is sent to the robot cloud platform, the robot cloud platform may first determine whether the environment information is processed by the cloud robot module, or whether the environment information is processed by the manual agent module, or The cloud robot module and the manual agent module jointly process the environment information, and finally the environment information is processed by the determined cloud robot module and/or the artificial seat module, and the instruction is sent to the robot body on the terminal side, through the robot Remote control of the cloud platform to ensure the quality of service to users and enhance the user experience.

In an implementation, the environment information may include the user's selection information, and the determining module may be specifically configured to: according to the user's selection information, determine to send the environment information to the cloud robot module of the robot cloud platform and/or Manual seating module.

In a specific implementation, the user may select, according to his own needs, the robot body to be operated by the cloud robot module, the robot body controlled by the manual seat module, or the cloud robot module and the manual seat module. operating. The user's selection information may be that the robot body is controlled by the cloud robot module, the robot body may be controlled by the artificial seat module, and the robot body may be controlled by the cloud robot module and the artificial seat module.

In the embodiment of the present application, the user's selection information may be transmitted to the robot cloud platform together with other sound, voice, image, or video information, and the determining module of the robot cloud platform may determine the environmental information according to the user's selection. The cloud robot module sent to the robot cloud platform is also a manual seat module.

In the embodiment of the present application, the user may actively select who controls the robot. Some users may feel that the computer processing method is more accustomed and convenient than the manual method, and then the artificial agent mode may be selected for interaction, and some other users may feel that the manual method is not as good as the computer processing mode. Faster, Concise, you can choose cloud robot mode interaction and so on. In summary, the embodiment of the present application can allow a user to have an active option to meet the user's usage habits, thereby further improving the user experience.

In an implementation, the collection module may be specifically configured to collect user selection information by using a switch button.

During the specific implementation, the acquisition module of the robot body can collect the selection information of the user by switching the button, and the user can perform button switching and the like according to his own needs.

The switch button may be a one-piece structure, which is determined by pressing/bounced at both ends, or may be a separate structure. For example, a guide helmet may be provided with a button set on the left ear portion of the helmet and another button set on the helmet. The right ear area. In addition, the switching button may be a method in which one end is pressed, the other end is automatically bounced, or a button is pressed or bounced. The specific structure and specific implementation manner of the switching button are not limited in this application, as long as the cloud robot module or the artificial seat module can be determined.

The embodiment of the present application collects the user's selection information by switching buttons, which is simpler and easier to implement than the voice mode, and can quickly determine the user's selection without performing voice recognition and the like, thereby improving the processing speed to a certain extent. Save costs.

FIG. 2 is a schematic structural diagram 2 of the cloud robot system in the first embodiment of the present application. As shown in the figure, the robot cloud platform may further include:

The second determining module 1024 is configured to determine whether the complexity of the environment information exceeds a processing range of the cloud robot module, or whether the cloud robot module is faulty;

The determining module may be specifically configured to: when the complexity of the environment information exceeds a processing range of the cloud robot module, or when the cloud robot module fails, determine the manual agent module to process the environment information; When the complexity of the environment information is within the processing range of the cloud robot module, the cloud robot module is determined to process the environment information.

In the embodiment of the present application, the robot cloud platform may receive the robot body and send the After the environment information, determining whether the cloud robot module can be processed, determining whether to send the environment information to the cloud robot module or the manual agent module according to whether the cloud robot can process, the right to select may be in the Robot cloud platform side.

In a specific implementation, if the complexity of the environment information exceeds the processing range of the cloud robot module, the human agent module may be determined to process the environment information for a better service user.

In the embodiment of the present application, if the complexity of the environment information is within the processing range of the cloud robot module, in order to reduce labor costs, the environment information may be preferentially processed by the cloud robot module.

In the embodiment of the present application, it is considered that even if the complexity of the environment information is within the processing range of the cloud robot module, if the cloud robot module has a system failure, the user cannot be provided with a service. The determining module can directly determine that the environmental information is processed by the manual agent module to ensure the quality of service to the user.

In a specific implementation, if the detecting module detects that the cloud robot module is all normal, the cloud robot module may be determined to process the environment information. If the complexity of the environment information exceeds the processing range of the cloud robot, it may be determined that the manual agent module is processed; when the cloud robot module is operating normally and the complexity of the environment information is in the cloud When the processing range of the robot module is within the range, the cloud robot module may be determined to process the environment information, and the specific implementation may be set as needed.

In a specific implementation, the environment information may include the user's selection information or the user's selection information, and the environment information includes the user's selection information and the user's selection information is determined by the robot cloud platform. When the results are inconsistent, the robot body (which can be understood as the terminal side or the user) can send a prompt message to inform the user that the selected module cannot provide the service or is not optimal.

The embodiment of the present application determines the cloud machine by determining the complexity of the environment information. The human module is also the manual agent module to process the environment information, and can intelligently determine whether the cloud robot module can be processed, so that the module that is most suitable for processing the environment information can be selected to process the environment information.

In an implementation, the second determining module may specifically include:

a calculating unit, configured to calculate a confidence level for performing voice recognition, face recognition, image recognition, and/or obstacle avoidance on the environmental information;

a comparing unit, configured to compare the confidence level with a preset first threshold;

And a determining unit, configured to determine, according to the comparison result, whether the complexity of the environment information exceeds a processing range of the cloud robot module.

In the embodiment of the present application, when determining whether the complexity of the environment information is within the processing range of the cloud robot, the identifier may be determined by using the identifier, and it is assumed that the environment information needs to be recognized by the voice, the face, and the image (including: For operations such as text, gestures, etc., or obstacle avoidance, the preset threshold is 95%. Then, the specific judgment process can be:

1) speech recognition, when the confidence level of the speech recognition is ≥95%, the judgment complexity is within the treatable range, and the cloud robot is determined to process the environmental information;

2) Face recognition, when the confidence level of the face recognition is ≥95%, the judgment complexity is within the treatable range, and the cloud robot is determined to process the environment information;

3) image recognition, when the confidence of the image recognition is ≥95%, the judgment complexity is within the treatable range, and the cloud robot is determined to process the environmental information;

4) Obstacle avoidance, when the confidence level of the obstacle avoidance is ≥95%, the judgment complexity is within the treatable range, and the cloud robot is determined to process the environmental information.

In a specific implementation, the confidence preset thresholds of different operations may be different. The operations of the speech recognition, the face recognition, the image recognition, the obstacle avoidance and the like may all adopt the existing implementation manners, and the calculation of the confidence of these operations may also adopt the existing confidence calculation formula, and the corresponding development and design by those skilled in the art. This can be achieved, and the present application does not describe it here.

The embodiment of the present application can use the confidence concept in statistics as a reference to determine whether the complexity of the environmental information is within the machine processable range.

In an implementation, the determining module may be further configured to determine, after determining the manual agent module to process the environment information, if a confidence level of an algorithm for processing the environment information by the cloud robot module exceeds a preset second threshold, The cloud robot module and the manual agent module process the environment information, the cloud robot module is configured to send an instruction to the robot body; if the cloud robot module processes the environment information, the confidence of the algorithm exceeds a pre The second threshold is set and continues for a preset time, and the cloud robot module is determined to process the environment information and send an instruction to the robot body.

In a specific implementation, after determining that the manual agent module processes the environment information, if the cloud robot module returns to normal, and the confidence of the algorithm for processing the environment information by the cloud robot module increases and exceeds a preset second threshold, Then, it can be determined that the cloud robot module and the manual agent module jointly intervene and process the environment information, and the cloud robot module sends an instruction to the robot body.

If the confidence level of the algorithm for processing the environment information by the cloud robot module exceeds a preset second threshold and continues for a certain time (assuming the preset time is 1 minute), it may be determined that the cloud robot module processes the The environmental information is sent to the robot body.

Through the above process, after the cloud robot module can process the environment information, the manual agent module can slowly exit and no longer participate to relieve manual pressure.

In an implementation, the determining module may be further configured to: after determining that the cloud robot module processes the environment information and sends an instruction to the robot body, if the algorithm of the robot body processing the environment information has a confidence level exceeding a preset number The third threshold determines that the robot body processes the environmental information.

In a specific implementation, when the cloud robot module remotely controls the robot body, if the confidence of the algorithm for processing the environment information by the robot body is found to be gradually increased, When the reliability is increased to exceed the preset third threshold, it may be determined that the environment information is processed by the robot body itself, without remote control of the robot cloud platform, reducing the network transmission pressure and improving the response speed to a certain extent. The user experience is better.

In implementation, the robot body and the robot cloud platform can perform data transmission through a dedicated communication network.

For traditional Internet services, the network transmits messages as pipes. The quality and speed of transmission mostly depend on the network quality and transmission speed of the telecom operators. In the existing cloud robot implementation scheme, the message transmission cannot guarantee the real-time performance of message transmission when the network is busy and the transmission quality is poor, and it is easy to be attacked and invaded by the hacker during the transmission process, and the accuracy and real-time of the message are satisfied. Aspects seriously affect the service to users.

3 is a schematic structural diagram of a cloud robot system according to Embodiment 1 of the present application. As shown in the figure, the cloud robot system may include a terminal side and a cloud platform. The terminal side includes a robot body, and the robot cloud platform includes a cloud robot. Module and manual agent module, where:

The robot body on the terminal side and the cloud robot module or the artificial seat module of the robot cloud platform can interact through a dedicated communication network.

The embodiment of the present application supports real-time interaction through a high-speed dedicated transmission network, and the terminal-side robot body can interact with the robot cloud platform via a secure high-speed communication network in a wireless or wired manner, thereby performing security protection and accelerated transmission for the transmitted message. It ensures the real-time and accuracy of the message and improves the quality of service to the user.

FIG. 4 is a schematic diagram showing changes in processing state in the first embodiment of the present application. As shown in the figure, the state change between the robot body, the cloud robot module, and the artificial seat module may be as follows:

First, start the robot body;

After starting the robot body, it can be decided by the robot body;

When the robot body cannot be processed, it can be converted into a cloud robot intervention, and the cloud robot is responsible for decision making;

If the user requires manual intervention, it can also be directly transferred to the cloud robot and the artificial agent to intervene, and the human agent is responsible for the decision;

If the robot ontology algorithm increases the confidence level and reaches the preset threshold, it can be transferred back to the robot ontology decision;

If the cloud robot can't handle it, it can be transferred to the cloud robot and the artificial seat to intervene, and the decision is made by the artificial seat;

The algorithm of the cloud robot increases the confidence level and reaches a preset threshold, and can be transferred back to the robot ontology decision;

When the confidence of the robot ontology algorithm is increased and the preset threshold is reached, or the user requests to be processed by the robot body, the manual agent decision can be transferred to the robot ontology decision.

After understanding the environment to be used, the terminal side and the network side can be implemented as follows. In the description process, the implementation of the terminal robot and the network side robot cloud platform will be explained separately, but this does not mean that the two must be implemented together. In fact, when the terminal robot is implemented separately from the network robot cloud platform, They solve the problems on the terminal side and the network side respectively, but when they are combined, they will get better technical effects.

Embodiment 2

Based on the same inventive concept, the embodiment of the present application also proposes a robot, which will be described below.

FIG. 5 is a schematic structural diagram 1 of the robot in the second embodiment of the present application. As shown in the figure, the robot may include:

The collecting module 501 is configured to collect environment information.

The first determining module 502 is configured to determine whether the complexity of the environment information exceeds a processing range of the robot body;

The sending module 503 is configured to send the environment information to the robot cloud platform when the complexity of the environment information exceeds a processing range of the robot body;

The receiving module 504 is configured to receive an instruction sent by a cloud robot module of the robot cloud platform or a manual agent module.

The robot in the embodiment of the present application may be a navigation robot, a guide robot, or a housekeeping service robot. The shape/structure of the robot may be a movable independent structure (eg, a clean disk shape, a "human" shape, etc.) It can also be a structure such as a guide cane or a blind guide helmet.

The environment information collected by the collection module may be information such as sound, image, video, etc. of the surrounding environment, or voice information sent by the user. The sending module sends the collected environmental information to the robot cloud platform, and the robot cloud platform performs the next processing.

The receiving module may receive an instruction sent by the cloud robot module of the robot cloud platform, or receive an instruction sent by the artificial seat module of the robot cloud platform, which compensates for the prior art that only the cloud robot module can be received. Directives, so that when the cloud robot module is busy or faulty, it can not receive instructions, and thus can not provide users with better service defects.

The determining, by the first determining module, whether the complexity of the environment information exceeds the processing range of the robot body may be: calculating a confidence level for performing voice recognition, face recognition, image recognition, obstacle avoidance, etc. on the environment information. And comparing the confidence level with a preset threshold, and determining, according to the comparison result, whether the complexity of the environment information exceeds a processing range of the robot body.

The robot provided by the embodiment of the present application can be remotely controlled by the cloud robot module of the robot cloud platform or remotely controlled by the artificial seat module of the robot cloud platform, thereby ensuring the quality of service to the user.

The robot provided by the embodiment of the present application first performs complexity prediction after collecting the environmental information, and if it is determined that the complexity of the environmental information exceeds the processing range of the robot body, the collected environmental information is Sending to the robot cloud platform, receiving the cloud robot module of the robot cloud platform, or the instruction sent by the manual seat module, that is, the robot body is remotely controlled by the robot cloud platform when the robot body cannot process the environment information, and The robot cloud platform includes two control modes of the cloud robot and the artificial seat, and solves the technical problem that the cloud robot module is unable to provide services or provide poor service quality if the cloud robot module is busy or faulty in the prior art, and ensures that the user is provided with a better technical problem. Good service improves the user experience.

In an implementation, the collecting module may be specifically configured to collect environment information including selection information of a user, where the selecting information is a cloud computing robot module and/or a human agent module to process the environment information.

In a specific implementation, the collected environment information may include selection information of the user, and the selection information of the user may be used to select the cloud robot module to process the environment information, or select a manual agent module to process the environment information, or select a cloud robot module and The manual agent module processes the environmental information together.

In the embodiment of the present application, the user may select to determine which module of the cloud platform processes the environment information, so that the user selects a service mode that conforms to his own usage habit, thereby further improving the user experience.

FIG. 6 is a schematic structural diagram 2 of the robot in the second embodiment of the present application. As shown in the figure, the robot may further include:

The processing module 505 is configured to process the environment information when a confidence level of an algorithm for processing the environment information by the robot body exceeds a preset third threshold.

In a specific implementation, if the robot body cannot process the environment information, the cloud platform may determine, by the cloud platform, the environment information is processed by the cloud robot module and the artificial agent module, and when the robot body processes the algorithm of the environment information, When the confidence level is increased and exceeds the preset third threshold, the environment information can be processed by the robot body without remote control of the robot cloud platform.

In implementation, the robot may further include:

A toggle button for collecting user selection information.

In the embodiment of the present application, a switching button can be set on the robot, and the user selects by using a button. You can choose the way you want to receive the service. In the specific implementation, the user can also select by voice. The switch button mode is simpler and easier to implement than the voice mode.

In an implementation, the sending module may be configured to send the environment information to a robot cloud platform via a dedicated communication network; the receiving module may be specifically configured to receive, by using the dedicated communication network, the determined cloud platform of the robot An instruction sent by a Cloud Robot module or a human agent module.

In a specific implementation, the sending module of the robot may send the environment information to a robot cloud platform through a dedicated communication network, and the receiving module of the robot may receive the cloud robot module of the robot cloud platform through the dedicated communication network or The instruction sent by the manual agent module. That is, the robot can perform data interaction with the robot cloud platform through the dedicated communication network.

Since the dedicated communication network can be used for the communication link of the robot and the robot cloud platform, the information transmission speed and security are higher than the existing mobile communication network, and the information security of the user can be ensured, and the information transmission speed is improved. The corresponding user waiting time is greatly reduced, thereby improving the quality of service.

Embodiment 3

Based on the same inventive concept, the embodiment of the present application also proposes a robot cloud platform, which will be described below.

FIG. 7 is a schematic structural diagram 1 of the robot cloud platform in the third embodiment of the present application. As shown in the figure, the robot cloud platform may include: an information receiving module 701, a determining module 702, a cloud robot module 703, and a manual seating module 704. ,among them,

The determined cloud robot module and/or manual agent module for processing the environment Information and send instructions to the robot body.

In a specific implementation, the cloud robot module and the artificial seat module may each have the capability of analyzing and processing environmental information collected by the terminal side robot body, and generating an instruction feedback to the terminal side robot body. Specifically, it is processed by the cloud robot module, processed by the manual agent module, or processed by the cloud robot module and the manual agent module, and can be determined by the determining module. Finally, the module determined by the determining module (the cloud robot module and/or the manual agent module) processes the environmental information and sends an instruction to the robot body.

In the embodiment of the present application, a third-party artificial seat module may be introduced in an existing cloud robot implementation solution, and the robot cloud platform determines whether the cloud robot module remotely controls the robot body, or is the manual seat module Remotely manipulating the robot body, using the cloud robot module and/or the artificial seat module to remotely control to ensure the quality of service to the user.

The robot cloud platform provided by the embodiment of the present application may determine whether the environment information is processed by the cloud robot module or is processed by the manual agent module after receiving the environment information sent by the terminal side robot body. The environment information, or the environment information is jointly processed by the cloud robot module and the manual agent module, and then the environment information is processed by the determined cloud robot module and/or the artificial seat module and the command is sent to The robot body on the terminal side remotely controls the terminal side robot body in two ways by introducing a manual seat, thereby ensuring the service quality of the terminal side robot body to the user, thereby improving the user experience.

In an implementation, the environment information may include the user's selection information, and the determining module may be specifically configured to determine, according to the user's selection information, the cloud robot module and/or the manual agent module to process the environment information. .

In a specific implementation, the user can select, according to his own needs, the robot body to be manipulated by the cloud robot module, and the robot body is controlled by the artificial seat module. The robot body is jointly controlled by the cloud robot module and the artificial seat module. The user's selection information may be that the robot body is controlled by the cloud robot module, or the robot body or the like may be manipulated by the artificial seat module.

In the embodiment of the present application, the robot cloud platform may receive the user's selection information and other sound, voice, image, or video information, and the determining module of the robot cloud platform may determine to send the environmental information according to the user's selection information. To the cloud robot module and/or the artificial seat module of the robot cloud platform.

The embodiment of the present application can receive the user's selection information, and the user actively selects who controls the robot. By allowing the user to have the active selection right, the user's usage habit can be satisfied, and the user experience is further improved.

FIG. 8 is a schematic structural diagram 2 of the robot cloud platform in the third embodiment of the present application. As shown in the figure, the robot cloud platform may further include:

The second determining module 705 is configured to determine whether the complexity of the environment information exceeds a processing range of the cloud robot module, or whether the cloud robot module is faulty;

In the embodiment of the present application, after receiving the environment information sent by the robot body, the robot cloud platform may determine whether the cloud robot module can be processed, and determine whether to send the environment information according to whether the cloud robot can process The cloud robot module is also the manual seat module, and the selected right may be on the robot cloud platform side.

In a specific implementation, the environment information may include the user's selection information or the user's selection information, and the environment information includes the user's selection information and the user's selection. When the selected information is inconsistent with the result determined by the robot cloud platform, the robot body (which may be understood as the terminal side or the user) may send the prompt information.

In a specific implementation, if the complexity of the environment information exceeds the processing range of the cloud robot module, the human agent module may be determined to process the environment information for a better service user. If the complexity of the environmental information is within the processing range of the cloud robot module, in order to reduce labor costs, the environment information may be preferentially processed by the cloud robot module.

In the embodiment of the present application, it is also considered that even if the complexity of the environment information is within the processing range of the cloud robot module, if the cloud robot module has a system failure, the user cannot be provided with a service. The determining module can directly determine that the environmental information is processed by the manual agent module to ensure the quality of service to the user.

The embodiment of the present application determines whether the cloud information is processed by the cloud robot module or the manual agent module by determining the complexity of the environment information, and can intelligently determine whether the cloud robot module can be processed, so that the most A module adapted to process the environmental information to process the environmental information.

In an implementation, the second determining module may specifically include:

In an implementation, the determining module may be further configured to determine, after determining the manual agent module to process the environment information, if a confidence level of an algorithm for processing the environment information by the cloud robot module exceeds a preset second threshold, The cloud robot module and the manual agent module process the environment information, and the cloud robot module is configured to send an instruction to the robot book If the confidence level of the algorithm for processing the environment information by the cloud robot module exceeds a preset second threshold and continues for a preset time, determining that the cloud robot module processes the environment information and sends an instruction to the robot body.

In a specific implementation, the environment information may be processed by a human agent module due to user designation, cloud robot module failure, insufficient processing capability of the cloud robot module, and the like; during processing, if the cloud robot module is found to process the The confidence of the algorithm of the environment information is increased and exceeds a preset second threshold, then it may be determined that the environment information is jointly processed by the cloud robot module and the manual agent module, and the cloud robot module sends an instruction to the Robot body

If the confidence level of the algorithm for processing the environment information by the cloud robot module exceeds a preset second threshold and continues for a certain period of time, the cloud robot module may be considered to be stable and can process the environment information, so The environment information is processed by the cloud robot module and an instruction is sent to the robot body, and the manual agent module is no longer involved.

Through the above process, with the enhancement of the processing capability of the cloud robot module, the artificial seat module can gradually withdraw and slowly no longer intervene, and gradually obtain the stable and reliable processing capability for the user, and gradually get rid of the manual intervention and relieve the labor. pressure.

In a specific implementation, in the process of processing the environment information by the cloud robot module, if it is found that the confidence of the algorithm for processing the environment information by the robot body is gradually increased and exceeds a preset third threshold, the The robot body processes the environmental information without remote control of the robot cloud platform.

In a specific implementation, the algorithm for processing the environmental information of the robot body may also be used. After the degree exceeds the preset third threshold and continues for a certain time, the environment information is processed by the robot body.

In an implementation, the information receiving module may be specifically configured to receive environment information sent by the robot body through a dedicated communication network; the determined cloud robot module or the artificial agent module may be specifically configured to process the environment information and pass through a dedicated communication network. Sending an instruction to the robot body.

For convenience of description, the various parts of the above described devices are described in terms of functions divided into various modules or units. Of course, the functions of each module or unit may be implemented in the same software or hardware in the implementation of the present application.

Embodiment 4

The technical solution provided by the present application can be applied to a smart guide blind scene. The following is a description of a guide robot as an example.

Assuming that the structure of the guide robot is a helmet structure, the blind person can wear the guide helmet to the head.

The front of the guide helmet may have a camera for collecting images, videos and the like of the surrounding environment; the right ear portion of the guide helmet is provided with a switching button, and the switching button may include two ends (one end is pressed The representative selects the cloud robot module, and the other end presses the representative to select the artificial seat module), one end is pressed, and the other end is automatically bounced. By default, the guide helmet can select the cloud robot module by default.

When the blind person wears the guide blind helmet to walk on the road, the guide blind helmet can collect the front image or video information, and the current position information of the blind person in real time, and the guide helmet can be first First determine if you can handle the collected information and guide the blind:

If the blind guide helmet can process the collected information and guide the blind person to travel, they can handle and guide the blind person to travel;

If the collected information has exceeded the range that the guide helmet can handle, the guide helmet cannot process the information or give guidance, then the guide helmet can display the image or video information, location information of the blind person The environmental information is transmitted to the robot cloud platform via a dedicated communication network.

The robot cloud platform may first determine whether the blind person has selected a manual seat intervention. If the blind person does not select a manual seat intervention, the robot cloud platform may determine whether the cloud robot module is operating normally:

If the cloud robot failure is detected, the collected information is sent to the manual agent module for processing;

If the cloud robot is running normally, and further determining whether the cloud robot can process the collected information, if the cloud robot is busy, or the collected information has a high complexity, the The collected information is sent to the manual agent module for processing; otherwise, it can be processed by the cloud robot module.

The cloud robot module can perform image recognition, plan a next instruction for the blind according to the current position of the blind person and the preset target position, and the instruction can be a voice instruction such as “continue to advance” and “turn left”.

Because the cloud robot module in the embodiment of the present application sends the voice command to the guide helmet through a dedicated communication network, the transmission speed is fast, and the prior art transmission through the mobile communication network can be avoided. Affect the situation of guiding the blind.

If the collected information is highly complex (for example, complicated scene pictures), or if the cloud robot is busy, resulting in low processing efficiency or even failure, the robot cloud platform can immediately switch to manual seat processing to ensure blindness. User's guidance.

In addition, during walking, blind people can also manually dial the place according to their own wishes. The switching button 2 is switched to the manual agent intervention, and the robot cloud platform can also immediately switch to the manual seat mode after receiving the switching instruction to meet the needs of the blind user.

The manual agent module can receive a guide instruction input by a third-party server (for example, a customer service staff) through an input device such as a keyboard or a microphone, and send the guide instruction to the guide helmet to realize blindness User's guidance.

The guide blind helmet may further include a microphone/microphone device, so that the blind user can directly issue a voice to issue an instruction. For example, the blind person can directly say "I want to go to the Wangjing subway station, how to go?".

After receiving the voice information, the cloud robot module performs voice recognition, semantic understanding, image recognition, or obstacle avoidance processing, and may generate a voice command in conjunction with the current position of the blind person, and send the voice command to the guide helmet. Introducing the blind helmet to play the voice command to guide the blind person;

When the cloud robot module is unable to process, the human service may be provided, and after receiving the voice information, the service personnel may quickly respond to the blind person's current location. In a specific implementation, the service personnel can input a response instruction through a keyboard/touch screen, or can directly input a voice and input a response instruction by voice.

In the process of the human service, if the processing capability of the cloud robot module is changed, and the service can be provided to the user, the human agent and the cloud robot can be jointly involved; the processing capability of the cloud robot module can provide the user with the processing capability. When the service is maintained for 2 minutes, the artificial seat can no longer be involved, and only the cloud robot intervenes. In this process, the artificial seat slowly withdraws.

In the process of the cloud robot service, if the processing capability of the terminal-side robot body changes, when the service can be provided to the user, it can be handled by the robot body itself, or can be maintained while the robot body capability can provide services for the user. After a minute, it is processed by the robot itself.

Embodiment 5

The technical solution provided by the present application can also be applied to the field of domestic service, and the following is an example of a domestic service robot.

Assuming that the shape of the housekeeping service robot is similar to the shape of a person, the housekeeping service robot can move.

When the user is at home, the user can be instructed by voice to the domestic service robot. For example, the user speaks "Please help with the mop".

After receiving the voice information sent by the user, the housekeeping service robot can determine whether the voice information can be recognized by itself. Since the voice information is relatively simple and is a regular service content, the housekeeping service robot can use its own processing. The chip identifies the voice information and controls the corresponding execution unit to perform the mopping operation after the identification.

When the user wants to cook, the user can be voiced to the home service robot. For example, the user speaks "Please use rice cooker to steam rice."

Although the voice information of the user is relatively simple, and the home service robot can recognize the voice, performing the operation requires finding a rice cooker, rice, and putting the rice into the rice cooker, adding water, etc., and the housekeeping service robot cannot give a response. At this time, the voice information and the currently collected image may be sent to the cloud robot module, and the cloud robot module performs voice recognition, semantic understanding, and the like on the voice information, and then issues an action command to the office. In the housekeeping service robot, the action instruction may include a direction command, a walking instruction, and a control instruction indicating that the housekeeping service robot puts rice.

The instruction may be a binary instruction or a synthesized voice instruction.

The housekeeping service robot may perform a corresponding operation according to an instruction sent by the cloud robot module.

Since the housekeeping service robot and the cloud robot module transmit data through a dedicated communication network in the embodiment of the present application, the speed of the transmission instruction is fast, and the user can quickly provide the mopping service, so that the user feels better.

It is assumed that the robot cloud platform can be intervened by the manual agent module after judging and determining that the voice command of the user is too complicated and the cloud robot module cannot be processed.

In a specific implementation, it is assumed that the robot cloud platform determines that the cloud robot module can process the voice instruction of the user, but the user thinks that the service efficiency of the housekeeping service robot is poor, and the user may also request to switch to Manual mode. Specifically, the switching may be indicated by a button or a voice, and the selection information of the user may be sent to the robot cloud platform, and the robot cloud platform may perform corresponding switching according to the user's selection information.

The manual agent module may select an idle service personnel to send the user's voice command to the service personnel. The service personnel may find the placement position of the rice and the rice cooker in the user's kitchen according to the image or video information collected by the domestic service robot, and instruct the robot to perform corresponding operation steps.

It is assumed that the cloud robot has the processing capability of simply putting rice into a rice cooker, etc., after the artificial service is used to find the placement position of the rice and the rice cooker, the cloud robot can make a decision and send an instruction, etc., during which the artificial Monitoring the decision and instruction of the cloud robot is correct. After a certain period of time, the manual service can be slowly separated and no longer involved, and only processed by the cloud robot and remotely controlled.

It is assumed that the housekeeping service robot has the processing capability of pressing a power button or the like, and after the cloud robot instructs the rice and water to be placed in the rice cooker, the housekeeping robot can control the corresponding execution unit to press the power button. The cloud robot is no longer involved.

In the embodiment of the present application, when the home service robot cannot be processed locally, the robot cloud platform is introduced, and the smart cloud robot and the artificial seat are used to assist the solution, and according to the local processing capability of the cloud robot and the housekeeping service robot, slowly From manual conversion to cloud robots, then to local processing robots for local processing.

Those skilled in the art will appreciate that embodiments of the present application can be provided as a method, system, or computer program product. Therefore, the present application can adopt an entirely hardware embodiment, complete software implementation. For example, or in combination with an embodiment of software and hardware aspects. Moreover, the application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

While the preferred embodiment of the present application has been described, it will be apparent that those skilled in the art can make further changes and modifications to the embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and the modifications and

Claims

A cloud robot system includes a robot body and a robot cloud platform, the robot body includes an acquisition module, a first determination module, and a transmission module, and the robot cloud platform includes a determination module, a cloud robot module, and a manual seat module. ,among them,

The collecting module is configured to collect environmental information;

The first determining module is configured to determine whether the complexity of the environment information exceeds a processing range of the robot body;

The sending module is configured to send the environment information to the robot cloud platform when the complexity of the environment information exceeds a processing range of the robot body;

The determining module is configured to determine the cloud robot module and/or the manual agent module to process the environment information;

The determined cloud robot module and/or manual agent module is configured to process the environment information and send an instruction to the robot body.
The system according to claim 1, wherein the environment information includes selection information of the user, and the determining module is specifically configured to determine to send the environmental information to the robot cloud according to the selection information of the user. The cloud robot module and/or the artificial seat module of the platform.
The system of claim 1, wherein the robot cloud platform further comprises: a second determining module, configured to determine whether the complexity of the environment information exceeds a processing range of the cloud robot module, or Whether the cloud robot module is faulty;

The determining module is specifically configured to: when the complexity of the environment information exceeds a processing range of the cloud robot module, or when the cloud robot module fails, determine the manual agent module to process the environment information; When the complexity of the environment information is within the processing range of the cloud robot module, the cloud robot module is determined to process the environment information.
The system of claim 3, wherein the second determining module comprises:

a calculating unit, configured to calculate a confidence level for performing voice recognition, face recognition, image recognition, and/or obstacle avoidance on the environmental information;

a comparing unit, configured to compare the confidence level with a preset first threshold;

And a determining unit, configured to determine, according to the comparison result, whether the complexity of the environment information exceeds a processing range of the cloud robot module.
The system according to claim 3, wherein said determining module is further configured to: after determining said artificial agent module to process said environmental information, if said cloud robot module processes an algorithm of said environmental information Determining that the cloud robot module and the manual agent module process the environment information, the cloud robot module is configured to send an instruction to the robot body; and if the cloud robot module processes the The confidence of the algorithm of the environmental information exceeds a preset second threshold for a preset time, and the cloud robot module is determined to process the environmental information and send an instruction to the robot body.
The system of claim 5, wherein the determining module is further configured to: after determining that the cloud robot module processes the environmental information and sends an instruction to the robot body, if the robot body processes the environmental information The confidence of the algorithm exceeds a preset third threshold, and the robot body is determined to process the environmental information.
A robot characterized by comprising:

An acquisition module for collecting environmental information;

a first determining module, configured to determine whether the complexity of the environment information exceeds a processing range of the robot body;

a sending module, configured to send the environment information to the robot cloud platform when a complexity of the environment information exceeds a processing range of the robot body;

The receiving module is configured to receive an instruction sent by the cloud robot module of the robot cloud platform or the artificial seat module.
The robot according to claim 7, wherein said acquisition module is specifically for The environment information including the user's selection information is collected, and the selection information is a cloud host robot module and/or a human agent module to process the environment information.
The robot of claim 7 further comprising:

And a processing module, configured to process the environment information when a confidence level of an algorithm for processing the environment information by the robot body exceeds a preset third threshold.
A robot cloud platform, comprising: an information receiving module, a determining module, a cloud robot module and a manual seat module, wherein

The information receiving module is configured to receive environment information sent by the robot body;

The determining module is configured to determine the cloud robot module and/or the manual agent module to process the environment information;

The determined cloud robot module and/or manual agent module is configured to process the environment information and send an instruction to the robot body.
The robot cloud platform of claim 10, further comprising:

a second determining module, configured to determine whether the complexity of the environment information exceeds a processing range of the cloud robot module, or whether the cloud robot module is faulty;

The determining module is specifically configured to: when the complexity of the environment information exceeds a processing range of the cloud robot module, or when the cloud robot module fails, determine the manual agent module to process the environment information; When the complexity of the environment information is within the processing range of the cloud robot module, the cloud robot module is determined to process the environment information.
The robot cloud platform of claim 11, wherein the second determining module comprises:

a calculating unit, configured to calculate a confidence level for performing voice recognition, face recognition, image recognition, and/or obstacle avoidance on the environmental information;

a comparing unit, configured to compare the confidence level with a preset first threshold;

a determining unit, configured to determine, according to the comparison result, whether the complexity of the environmental information exceeds The processing range of the cloud robot module.
The robot cloud platform according to claim 11, wherein the determining module is further configured to: if the manual agent module determines the environment information, if the cloud robot module processes the environment information Determining a confidence level that exceeds a preset second threshold, determining that the cloud robot module and the manual agent module process the environment information, the cloud robot module is configured to send an instruction to the robot body; if the cloud robot module processes The confidence of the algorithm of the environment information exceeds a preset second threshold for a preset time, and the cloud robot module is determined to process the environment information and send an instruction to the robot body.
The robot cloud platform according to claim 13, wherein the determining module is further configured to: after determining that the cloud robot module processes the environment information and sends an instruction to the robot body, if the robot body processes the The confidence of the algorithm of the environmental information exceeds a preset third threshold, and the robot body is determined to process the environmental information.