Prompt word generation method, control system, cleaning robot, terminal and medium

Abstract

A prompt word generation method, comprising: acquiring instruction description information; and on the basis of the instruction description information, generating a prompt word adapted to the instruction description information, and determining a behavior policy associated with the prompt word, such that a cleaning robot (10) executes, on the basis of the prompt word, the behavior policy associated with the prompt word.

Description

Method for generating prompt words, control system, cleaning robot, terminal and medium

[0001] Cross-reference to Related Applications

[0002] This application claims priority to Chinese Patent Application No. 2024113542280, filed on September 26, 2024, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0003] The present disclosure relates to the field of automatic control, and in particular, to a method for generating prompt words, a control system, a cleaning robot, a terminal and a medium. BACKGROUND

[0004] Nowadays, cleaning robots with automatic cleaning capabilities are favored by most consumers. For example, a sweeping robot can automatically clean garbage in a user's home environment, improve the cleanliness of the home, and free the user from daily household cleaning. SUMMARY

[0005] The present disclosure provides a method for generating prompt words, a control system, a cleaning robot, a terminal and a medium.

[0006] In a first aspect of the present disclosure, a method for generating prompt words is provided, comprising: obtaining instruction description information; and based on the instruction description information, generating a prompt word adapted to the instruction description information, and determining a behavior strategy associated with the prompt word, so that a cleaning robot executes the behavior strategy based on the prompt word.

[0007] In a second aspect of the present disclosure, a method for generating prompt words is also provided, comprising: obtaining instruction description information input by a user; based on the instruction description information, generating a prompt word adapted to the instruction description information, and determining a behavior strategy associated with the prompt word; and sending the prompt word and the behavior strategy associated with the prompt word to a cleaning robot, so that the cleaning robot executes the behavior strategy based on the prompt word.

[0008] In a third aspect of the present disclosure, a control system of a cleaning robot is provided, comprising: a user terminal and a cleaning robot. The user terminal is configured to obtain instruction description information input by a user, based on the instruction description information, generate a prompt word adapted to the instruction description information, and determine a behavior strategy associated with the prompt word, and send the prompt word and the behavior strategy associated with the prompt word to the cleaning robot; and the cleaning robot is configured to execute the behavior strategy based on the prompt word.

[0009] In a fourth aspect of the present disclosure, a cleaning robot is provided, comprising a processor and a memory storing a computer program executable on the processor, which, when executed by the processor, implements the steps of the prompt word generation method provided in the first aspect above.

[0010] In a fifth aspect of the present disclosure, a user terminal is provided, comprising a processor and a memory storing a computer program executable on the processor, which, when executed by the processor, implements the steps of the prompt word generation method provided in the first aspect or the second aspect above.

[0011] In a sixth aspect of the present disclosure, a computer readable storage medium is provided, comprising computer instructions stored thereon, which, when executed by a processor, implements the steps of the prompt word generation method provided in the first aspect or the second aspect above. BRIEF DESCRIPTION OF DRAWINGS

[0012] FIG. 1 shows a structural schematic diagram of a cleaning robot according to some embodiments of the present disclosure.

[0013] FIG. 2 shows an exemplary scenario schematic diagram suitable for some embodiments of the present disclosure.

[0014] FIG. 3 shows a flowchart of a prompt word generation method according to some embodiments of the present disclosure.

[0015] FIG. 4 shows a schematic diagram of a keyword generation and screening process according to some embodiments of the present disclosure.

[0016] FIG. 5A shows a flowchart of a first prompt word generation mode according to some embodiments of the present disclosure.

[0017] FIG. 5B shows a flowchart of a second prompt word generation mode according to some embodiments of the present disclosure.

[0018] FIG. 6 shows a flowchart of a prompt word generation method according to some other embodiments of the present disclosure.

[0019] FIG. 7 shows a structural schematic diagram of a control system according to some embodiments of the present disclosure.

[0020] BRIEF DESCRIPTION OF DRAWINGS

[0021] Control system 1, cleaning robot 10, user terminal 20, cloud server 30, machine body 110, perception module 120, position determination device 121, front collision structure 122, human-computer interaction module 130. Embodiments of the present disclosure

[0022] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be thoroughly understood, and complete conveyance of the scope of the present disclosure to those skilled in the art is achieved.

[0023] It should be noted that the term "and / or" appearing in the present document merely describes an association relationship of associated objects, and means that three relationships can exist. For example, A and / or B can represent three cases of A existing alone, A and B existing together, and B existing alone. The terms "front", "back", "upper", "lower", "left", "right", and the like are merely used to represent relative positional relationships, and when the absolute positions of the described objects are changed, the relative positional relationships can also be changed accordingly.

[0024] The cleaning robot is an intelligent cleaning device with self-moving function, which is used to clean target objects in the environment. For example, the cleaning robot can be a sweeping robot, a mopping robot, a sweeping and mopping integrated robot, a floor polishing robot, or a weeding robot, etc. The embodiments in the present disclosure mainly take the sweeping robot as an example for illustration.

[0025] Taking the cleaning robot as a sweeping robot, Fig. 1 shows a structural schematic diagram of the cleaning robot according to some embodiments of the present disclosure. It should be noted that the structure and shape of the cleaning robot shown in Fig. 1 are only schematic and are not limiting. As shown in Fig. 1, the cleaning robot can include a robot main body 110, a perception module 120, a controller, a driving module, a cleaning assembly, a power module, and a human-computer interaction module 130. As shown in Fig. 1, the robot main body 110 includes a front portion 111 and a rear portion 112, and has an approximately circular shape (circular in front and back), or other shapes, including but not limited to an approximately letter D shape with a circular front and a rectangular or square shape with a circular front and back.

[0026] For example, the perception module 120 includes a position determination device 121 located on the robot main body 110, a collision sensor arranged on the front collision structure 122 of the front portion 111 of the robot main body 110, a wall sensor located on the side of the robot, a cliff sensor arranged at the lower part of the robot main body 110, and a magnetometer, an accelerometer, a gyroscope, an odometer, and other sensing devices arranged inside the robot main body 110, which are used to provide various position information and motion state information of the robot to the controller.

[0027] As shown in FIG. 1, the front portion 111 of the machine body 110 can carry a front bumper structure 122, which, via a sensor system, such as a collision sensor or a proximity sensor (e.g., an infrared sensor) arranged thereon, detects one or more events in the travel path of the cleaning robot 10 when the drive module propels the cleaning robot 10 to travel on the ground during a cleaning process, and the cleaning robot 10 can control the drive module to respond to the events, such as obstacles, walls, etc., detected by the front bumper structure 122, such as performing an obstacle avoidance operation away from the obstacles, etc.

[0028] For example, the controller can be arranged on a circuit board within the machine body 110, including a non-transitory memory (e.g., a hard disk, a flash memory, and a random access memory, etc.) and a processor (e.g., a central processing unit and an application processor, etc.).

[0029] For example, the drive module can maneuver the machine body 110 to travel across the ground based on a drive command having distance and angle information. For example, the cleaning assembly can include a dry cleaning assembly and / or a wet cleaning assembly, and the specific structure can refer to the related art. When the cleaning robot 10 is in a working mode, i.e., performing a cleaning task, the cleaning assembly can clean a target surface, such as the ground.

[0030] For example, the power module can include a rechargeable battery, such as a nickel-hydrogen battery and a lithium battery. The rechargeable battery can be charged by connecting a charging electrode arranged on the side or below the machine body 110 to an electrode on the cleaning base station 20.

[0031] For example, the human-machine interaction module 130 can include a key on the host panel for a user to select a function. The human-machine interaction module 130 can also include one or more of a display screen, an indicator light, and a speaker to show the user the current mode of the machine or the function selection item. The human-machine interaction module 130 can also include a microphone to receive voice instructions from the user to implement a voice control function. In some embodiments, the user can also interact with the cleaning robot 10 through a client arranged on a user terminal in communication connection with the cleaning robot 10.

[0032] The structures of the cleaning robot 10 listed above are only exemplary, and the cleaning robot 10 can also include other structures, which are not limited by the present disclosure.

[0033] According to some embodiments of the present disclosure, a prompt word generation method for generating an instruction prompt word of a cleaning robot 10 is provided, which facilitates the user to issue an instruction to the cleaning robot 10 and is conducive to enriching the functions of the cleaning robot 10.

[0034] FIG. 2 shows a schematic diagram of an example scenario applicable to some embodiments of the present disclosure. As shown in FIG. 2, the cleaning robot 10 can establish a communication connection with a user terminal 20 to implement data interaction between the cleaning robot 10 and the user terminal 20, and the user terminal 20 can establish a communication connection with a cloud server 30 to implement data interaction between the user terminal 20 and the cloud server 30. The user terminal 20 can be installed with a client (i.e., an APP) for managing and controlling the cleaning robot 10. The user can issue control instructions to the cleaning robot 10 through the client and understand the status and task progress of the cleaning robot 10. The cloud server 30 can provide the client with services related to the cleaning robot 10, such as software update services, storage services, and computing services, etc. In some embodiments, the cleaning robot 10 can also establish a communication connection with the cloud server 30 to implement data interaction with the cloud server 30.

[0035] FIG. 3 shows a flowchart of a prompt word generation method according to some embodiments of the present disclosure, which can be executed by the user terminal 20 or by the cleaning robot 10. As shown in FIG. 3, the prompt word generation method can at least include:

[0036] Step S101, obtaining instruction description information; and

[0037] Step S102, generating a prompt word adapted to the instruction description information based on the instruction description information, and determining a behavior strategy associated with the prompt word, so that the cleaning robot executes the behavior strategy based on the prompt word.

[0038] The instruction description information is description information of an instruction that needs to be executed by the cleaning robot. For example, the description information can at least include a relevant description of an object to which the instruction is directed. For another example, the description information can also include a relevant description of a behavior strategy corresponding to the instruction. Automatically generating an adapted prompt word based on the instruction description information and determining a behavior strategy associated with the prompt word can simplify the operation process of the user, facilitate the user to issue instructions to the cleaning robot, and be conducive to enriching the functions of the cleaning robot.

[0039] As some implementations, in step S101, the instruction description information input by the user can be obtained. There can be various implementations for the user to input the instruction description information. For example, the user can interact with the client installed on the user terminal to input the instruction description information. In some embodiments, the user terminal is in communication connection with the cleaning robot. Exemplarily, the two can be connected through wireless communication modes such as Bluetooth or WIFI. Exemplarily, the user terminal can be a mobile terminal such as a mobile phone, a tablet computer, or a wearable device (such as a smart watch or a bracelet). For another example, the user can also input the instruction description information through interaction with the cleaning robot.

[0040] The input form of the instruction description information can be various, and can be set according to the actual product needs. For example, the instruction description information can include one or more of the following information: text, voice, picture, and video. It should be noted that the instruction description information can include a single one of text, voice, picture, and video or a combination of multiple information. For example, the instruction description information can include text and voice, or text and picture or video, or voice and picture or video, or text, voice, and picture or video. In some embodiments, the picture or video is input as the instruction description information, without the user thinking about how to describe the instruction, which is beneficial to further reduce the input difficulty of the user, and the generation of the prompt word does not need to depend on the user description, which can avoid being affected by the personal preference of the user.

[0041] For example, in the case of inputting the instruction description information by text or voice, the instruction description information can include the related description of the object to which the instruction is directed and the related description of the behavior strategy corresponding to the instruction. For example, the instruction description information can be "sweep the area around the square black object at the wall", the related description of the object to which the instruction is directed is "the square black object at the wall", and the related description of the behavior strategy corresponding to the instruction is "sweeping". In the case of inputting the instruction description information by picture or video, the instruction description information is a picture or video containing the object to which the instruction is directed.

[0042] As some embodiments, the process of obtaining the instruction description information input by the user can include: showing the input interface of the instruction description information to the user; and obtaining the instruction description information input by the user in the input interface. For example, the input interface can provide one or more of the following input channels: a voice input channel, a text input channel, and an image input channel. The voice input channel is used to input the instruction description information by voice, the text input channel is used to input the instruction description information by text, and the image input channel is used to input the picture or video as the instruction description information.

[0043] For example, the user can perform an operation for triggering the input of the instruction description information by interacting with the user terminal, so that the user terminal shows the input interface of the instruction description information to the user. The operation for triggering the input of the instruction description information can be various, for example, it can be a key operation, a voice wake-up operation, a gesture operation, or a touch operation (such as a click operation, a sliding operation, or a tapping operation), which can be set according to the actual product needs.

[0044] For example, the input interface is provided with a text input box, and the user can input a description of the instruction expected to be executed by the cleaning robot from the text input box. For another example, the input interface is provided with a voice input button, and the user can click the voice input button and then input a description of the instruction expected to be executed by the cleaning robot through voice input. For yet another example, the input interface is provided with an image input channel, and the user can input a real-time captured image or video as the description of the instruction or select a locally stored image or video as the description of the instruction. In actual implementation, the input interface can provide multiple types of input channels, and the user can select the required input mode on the input interface according to actual needs.

[0045] As another implementation, the process of obtaining the instruction description information input by the user can include: collecting voice information input by the user through a voice module in the cleaning robot, and taking the voice information as the instruction description information. In this way, the input of the instruction description information can be conveniently and quickly realized.

[0046] For example, the user can wake up the cleaning robot by interacting with the cleaning robot. Then, the user can issue an instruction to the cleaning robot through voice, and the input voice information is the instruction description information. Exemplarily, the wake-up word can be "Hi, ××", and "××" represents the name of the cleaning robot, which is determined according to actual application scenarios.

[0047] In some application scenarios, through the prompt word generation method of the embodiments of the present disclosure, the user can directly input the instruction description information to control the cleaning robot to execute some targeted tasks, such as cleaning a specified area, avoiding or ignoring a specified obstacle in the environment, performing deep cleaning on a specified area next time, arranging a specified area, or moving a specified object from position A to position B.

[0048] In addition to the input of the instruction description information by the user, as another implementation, the cleaning robot can also autonomously generate the instruction description information when encountering a preset event. For example, the process of obtaining the instruction description information in step S101 can include: taking a picture or video collected in the process of the cleaning robot encountering a preset event as the instruction description information. For example, the preset event can include a trapping event or other events that can be prevented from occurring again by modifying the behavior strategy.

[0049] Taking a stuck event as an example, it is assumed that the cleaning robot is stuck at an obstacle and cannot free itself. In order to avoid the same situation from happening again next time, the cleaning robot can obtain pictures or videos collected in the preset time period before being stuck as instruction description information. The preset time period can be set according to actual needs, for example, it can be set to 5 seconds or 10 seconds, etc. The pictures or videos contain the obstacle that stuck the cleaning robot, so that the prompt words related to the obstacle can be obtained from the pictures or videos, and then the prompt words are associated with the obstacle avoidance strategy, so that the cleaning robot can be controlled to execute the obstacle avoidance strategy when the obstacle is recognized next time, thereby avoiding being stuck by the obstacle again next time.

[0050] After obtaining the instruction description information, step 102 can be performed to generate prompt words adapted to the instruction description information based on the instruction description information.

[0051] As some embodiments, when the instruction description information is text or voice information, the instruction description information can be subjected to word segmentation processing, and the obtained segmented words are matched with the words in the preset word library respectively. The matched words in the preset word library are combined to obtain the prompt words adapted to the instruction description information. It should be noted that if the instruction description information is voice information, the voice information needs to be converted into text first.

[0052] As other embodiments, the prompt words adapted to the instruction description information can be obtained based on the instruction description information and a pre-trained processing model. For example, the processing model is a pre-trained machine learning model. The processing model can be set locally (such as a user terminal or a cleaning robot), or it can also be set in a cloud server to reduce the requirement for local computing power.

[0053] It should be noted that in the case that the processing model is set in the cloud server, the user terminal or the cleaning robot needs to send the obtained instruction description information to the cloud server, so that the cloud server obtains the prompt words adapted to the instruction description information based on the received instruction description information and the preset processing model; and then, the prompt words returned by the cloud server are received.

[0054] As some embodiments, the processing model can be a prompt word generation model, the input is the instruction description information, and the output is the prompt words corresponding to the instruction description information.

[0055] In view of the fact that the prompt word output by the processing model can not be a satisfactory prompt word for the user, in order to improve the accuracy of the prompt word generation, as some embodiments, the prompt word output by the processing model can be displayed to the user for confirmation. For example, the above process of obtaining the prompt word adapted to the instruction description information based on the instruction description information and the pre-trained processing model can include: processing the instruction description information by the processing model, generating a candidate prompt word and displaying it to the user; in response to the operation performed by the user on the candidate prompt word, obtaining the prompt word adapted to the instruction description information. If the user is satisfied with the candidate prompt word, the user can perform a confirmation operation on the candidate prompt word, such as clicking the submit or confirm button next to the candidate prompt word, to determine the displayed candidate prompt word as the prompt word adapted to the instruction description information. If the user is not satisfied, the user can perform an editing operation on the candidate prompt word in the display interface to modify the prompt word.

[0056] As another embodiment, the processing model can be a keyword generation model, the input is the instruction description information, and the output is the keyword extracted from the instruction description information. In some embodiments, the instruction prompt word is obtained based on the keyword. In order to improve the accuracy of the obtained keyword and thus improve the adaptability of the prompt word, the keyword output by the model can be displayed to the user for the user to select a suitable keyword. At this time, the above process of obtaining the prompt word adapted to the instruction description information based on the instruction description information and the pre-trained processing model can include: processing the instruction description information by the processing model, generating a candidate keyword and displaying it to the user; obtaining the prompt word based on the keyword selected by the user from the candidate keywords. For example, the user terminal can display the candidate keywords output by the model, and the user can select a suitable keyword from the candidate keywords.

[0057] In view of the fact that the candidate keywords generated by the processing model at one time can not be complete or satisfactory to the user, as some embodiments, the above process of obtaining the prompt word based on the keyword selected by the user from the candidate keywords can include: in response to the keyword selection operation of the user, retaining the keyword selected by the user from the candidate keywords; if the user triggers the keyword regeneration instruction, processing the instruction description information by the processing model again to generate new candidate keywords, and displaying the new candidate keywords and the retained keywords to the user for the user to select again; if the user triggers the keyword confirmation instruction after selecting the keywords, obtaining the prompt word based on the selected keywords. In this way, the user can independently select the keywords that meet the instruction description information, limit the content of the generated prompt word, and help to obtain a more accurate prompt word.

[0058] For example, FIG. 4 shows a schematic diagram of a keyword generation and screening process according to some embodiments of the present disclosure. As shown in FIG. 4, the instruction description information is processed by the processing model for the first time, and five candidate keywords are generated, which are keyword 1, keyword 2, keyword 3, keyword 4, and keyword 5. If the user is not satisfied with all the candidate keywords or the keywords are still not complete, the user can trigger the keyword regeneration instruction to let the processing model generate new candidate keywords again. For example, the user selects keyword 2 from the above five candidate keywords when performing the first keyword screening operation, and the newly generated candidate keywords are keyword 6, keyword 7, keyword 8, and keyword 9. Then, the user can be shown keyword 6, keyword 7, keyword 8, keyword 9, and keyword 2 screened out last time (i.e., the previously selected keyword is retained) for the user to continue screening. Assuming that the user selects keyword 7 in this keyword screening, but the keywords are still not complete, the processing model can be triggered again to generate new candidate keywords for the user to screen under the condition of retaining keyword 2 and keyword 7, until the user screens the satisfactory keywords. If the user gets the satisfactory keywords after the Nth screening, the keyword confirmation instruction can be triggered to end the keyword generation and screening, and N is greater than or equal to 1.

[0059] As some implementations, the process of obtaining the prompt word based on the keywords screened by the user can include: concatenating the keywords screened by the user into a coherent description sentence; and obtaining the prompt word based on the description sentence. For example, the keywords screened by the user can be concatenated in sequence to obtain a coherent description sentence.

[0060] As some implementations, the process of obtaining the prompt word based on the keywords screened by the user can include: concatenating the keywords screened by the user into a coherent description sentence; and obtaining the prompt word based on the description sentence. For example, the keywords screened by the user can be concatenated in sequence to obtain a coherent description sentence.

[0061] In order to further improve the accuracy of the prompt word, the user can be prompted to confirm the obtained prompt word. As some embodiments, the process of obtaining the prompt word based on the description sentence can include: displaying the description sentence to the user as a candidate prompt word; determining the candidate prompt word as the prompt word in response to the prompt word confirmation instruction triggered by the user; or, obtaining a modified candidate prompt word in response to the modification operation performed by the user on the candidate prompt word; and determining the modified candidate prompt word as the prompt word in response to the prompt word confirmation instruction triggered by the user.

[0062] That is, if the user is satisfied with the displayed candidate prompt word, the user can directly trigger the prompt word confirmation instruction to determine the candidate prompt word as the prompt word and submit the prompt word; if the user is not satisfied with the displayed candidate prompt word, the user can modify the candidate prompt word to determine the modified candidate prompt word as the prompt word and submit the prompt word.

[0063] In the above prompt word generation process, the user can participate in the selection of the keywords and the modification of the prompt word, which is equivalent to generating the prompt word semi-automatically through interaction with the user, and is conducive to ensuring the accuracy of the prompt word.

[0064] It should be noted that the above two prompt word generation methods: the first method of directly generating the prompt word through the prompt word generation model; and the second method of first generating the keywords for the user to select and then generating the prompt word based on the keywords selected by the user, can be used alone or in combination.

[0065] For example, in the scenario where the above two prompt word generation methods are used in combination, the processing model A, the processing model B and the sentence model can be pre-configured, the processing model A is the prompt word generation model, and the processing model B is the keyword generation model.

[0066] FIG. 5A shows a flowchart of the first prompt word generation method according to some embodiments of the present disclosure, and FIG. 5B shows a flowchart of the second prompt word generation method according to some embodiments of the present disclosure. As shown in FIG. 5A, after obtaining the instruction description information, the instruction description information can be input to the processing model A to generate and display the prompt word to the user. The user can choose whether to modify the prompt word to obtain a prompt word adapted to the instruction description information. If the user is not satisfied with the prompt word, the fine modification process shown in FIG. 5B can be triggered.

[0067] As shown in FIG. 5B, the instruction description information is input into the processing model B to generate candidate keywords. If the user is not satisfied with the keyword generation result or the keywords are not comprehensive enough, the processing model B can be used to generate new candidate keywords again until the user filters the satisfactory result. After the keywords are filtered, the keywords filtered by the user are input into the sentence model to infer coherent description sentences, which are used as candidate prompt words to be displayed to the user. The user can choose whether to modify the prompt words, and finally obtain the prompt words adapted to the instruction description information.

[0068] In this way, on the one hand, the prompt words can be quickly generated, and on the other hand, in the case that the user is not satisfied with the quickly generated prompt words, the fine modification of the prompt words can be realized through the interaction with the user, thereby ensuring the accuracy of the generated prompt words.

[0069] In some embodiments of the present disclosure, the prompt words can be prompt words of object categories involved in the instruction description information. For example, in the case that the instruction description information is text or voice information, the prompt words can be prompt words of object categories extracted from the text or voice information. For another example, in the case that the instruction description information is a picture or a video, the prompt words can be prompt words of object categories contained in the picture or the video.

[0070] Based on this, as some implementations, in the above step of processing the instruction description information by the processing model to generate candidate keywords and display them to the user, the processing model can be used to extract keywords belonging to object category description from the instruction description information as candidate keywords and display them to the user. The keywords belonging to object category description may, for example, include keywords related to the shape, position, color, state, and other attributes of the object.

[0071] Still taking the instruction description information "sweep the area around the square black object on the wall" input by text or voice as an example, the keywords filtered by the user can include "wall", "square", "black", and "object", and the obtained prompt words can be "square black object on the wall".

[0072] In step S102, in addition to generating the prompt words adapted to the instruction description information, the behavior strategy associated with the prompt words also needs to be determined, so that the cleaning robot can execute the corresponding behavior strategy based on the prompt words.

[0073] There are various implementations of determining the behavior strategy associated with the prompt words, and several exemplary implementations are listed below for description. It should be noted that in some application scenarios, the way of inputting the instruction description information can be different, and the way of determining the behavior strategy associated with the prompt words can also be different.

[0074] The first, after generating the prompt word, actively ask the user, by the user to choose. For example, the user can be prompted to select the behavior strategy of the prompt word; the user from the selected behavior strategy, determine the behavior strategy associated with the prompt word.

[0075] The optional behavior strategy is configured in advance according to the behavior strategy that the cleaning robot can perform. For example, the optional behavior strategy can include at least two of the following strategies: obstacle avoidance, zone cleaning, and deep cleaning.

[0076] For example, obstacle avoidance can include both execution and ignore (i.e., obstacle avoidance and no obstacle avoidance) strategies. If the user selects "execute", the cleaning robot can be controlled to perform obstacle avoidance actions when it identifies an object that matches the prompt word during subsequent execution of the task.

[0077] In some embodiments, if the user selects "ignore", the cleaning robot ignores the obstacle avoidance strategy associated with the object corresponding to the prompt word. For example, the cleaning robot previously performed obstacle avoidance when it identified object A. After generating the prompt word corresponding to object A based on the instruction description information input by the user, if the user selects the "ignore" obstacle avoidance strategy for the prompt word, the cleaning robot will not perform the previously associated obstacle avoidance strategy when it subsequently identifies object A.

[0078] For example, during daily use of the cleaning robot, the user discovers that the cleaning robot misses cleaning the area near the skirting line. The reason can be that the cleaning robot misidentifies the skirting line as a power cord, and the cleaning robot performs the obstacle avoidance strategy associated with the power cord category to avoid the area near the power cord to prevent entanglement, thereby causing the missed cleaning of the area near the skirting line. In this case, the user can input a picture of the skirting line as instruction description information, and after generating the prompt word corresponding to the skirting line, select "ignore" obstacle avoidance. When an object is identified that matches the prompt word corresponding to the skirting line and belongs to the power cord category, the obstacle avoidance strategy that should have been performed when identified as a power cord category is ignored. In some embodiments, when an object is identified that only belongs to the power cord category, the obstacle avoidance strategy associated with the power cord is still performed.

[0079] For example, if the user selects the zone cleaning strategy to be associated with the prompt word and starts the cleaning task, the cleaning robot can be controlled to search for an object in the environment that matches the prompt word, and then clean a predetermined area around the object, thereby achieving cleaning of the user-specified area.

[0080] For example, the deep cleaning can include one or a combination of the following execution manners: increasing suction, increasing water flow, reciprocating cleaning, and reducing cleaning speed. The execution manner of the deep cleaning can be configured according to the execution manner selected by the user from the above-mentioned multiple optional execution manners, or can be pre-configured before leaving the factory, and determined according to the actual product needs. If the user selects the deep cleaning strategy to be associated with the prompt word and starts the cleaning task, the cleaning robot can be controlled to, in the process of executing the cleaning task, when the object corresponding to the prompt word is recognized, perform deep cleaning on the preset area range around the object according to the configured execution manner.

[0081] In some embodiments, the cleaning robot is provided with a mechanical arm, and the optional behavior strategy can further include: tidying and / or moving position. For example, the optional behavior strategy can include at least two of the following strategies: obstacle avoidance, area selection cleaning, deep cleaning, tidying, and moving position.

[0082] For example, if the user selects the tidying strategy to be associated with the prompt word and starts the tidying task, the cleaning robot can be controlled to, when the object (such as clothes dropped on the carpet) corresponding to the prompt word is recognized, tidy the object through the mechanical arm.

[0083] For example, if the user inputs the instruction description information to control the cleaning robot to move the object A to a specified position (such as beside the sofa), the generated prompt word can include the object A and the specified position. After the user selects the moving position strategy to be associated with the prompt word and starts the moving position task, the cleaning robot can be controlled to recognize the object A according to the prompt word, and after recognizing the object A, the object A is lifted by the mechanical arm and moved to the specified position.

[0084] For example, the cleaning robot can recognize the object in the environment through a pre-set recognition model. For example, the cleaning robot pre-stores a basic category. After the cleaning robot obtains the prompt word, if there is a category in the basic category that matches the object category contained in the prompt word, the recognition result of the recognition model can be matched with the object category contained in the prompt word to determine whether the cleaning robot recognizes the object corresponding to the prompt word. If there is no category in the basic category that matches the object category contained in the prompt word, the object category contained in the prompt word can be input into the recognition model to identify the object that matches the object category from the picture collected by the cleaning robot through the recognition model.

[0085] For example, after generating the prompt word, a strategy selection interface can be displayed on the user terminal to prompt the user to select the behavior strategy associated with the prompt word, and the optional behavior strategies can be displayed in the interface. For example, the optional behavior strategies can be displayed in the form of a list, and the list is expanded for user selection when the user clicks the selection button; or the optional behavior strategies can be displayed in the form of options.

[0086] For example, in the case that the user actively inputs the instruction description information through a picture or a video, the relevant information in the instruction description information that does not involve the behavior strategy, i.e., the behavior strategy without the user's tendency, can be displayed to the user in the preset arrangement order.

[0087] For example, in the case that the instruction description information is a voice or a text, if there is a target behavior strategy in the selectable behavior strategies that matches the instruction description information, the target behavior strategy can be arranged in front of the rest of the selectable behavior strategies when the selectable behavior strategies are displayed to the user, so as to facilitate the user's selection. It can be understood that when the user inputs the instruction description information through a voice or a text, the user usually has an action that needs to be controlled to execute the cleaning robot, and the behavior strategy that reflects the user's tendency can be matched.

[0088] Still taking the above instruction description information "clean the area around the square black object on the wall" as an example, the instruction description information contains the relevant description "clean" of the behavior strategy corresponding to the instruction, and does not contain descriptions such as "intensify" and "focus" that tend to "deep cleaning". Then, the selected area cleaning can be matched from the above selectable behavior strategies, and the selected area cleaning can be placed in the front when the selectable behavior strategies are displayed to the user.

[0089] Secondly, in the case that the instruction description information is a voice or a text, if there is a target behavior strategy in the preset selectable behavior strategies that matches the instruction description information, the target behavior strategy is determined as the behavior strategy associated with the prompt word. In this way, the user's operation can be reduced, and the execution speed of the instruction can be improved.

[0090] If there is no target behavior strategy in the preset selectable behavior strategies that matches the instruction description information, the above first method is executed, and the user is actively asked to select the behavior strategy associated with the prompt word by the user himself / herself.

[0091] Thirdly, in the case that the instruction description information is a voice or a text, if there is a target behavior strategy in the preset selectable behavior strategies that matches the instruction description information, the user is confirmed whether to determine the target behavior strategy as the behavior strategy associated with the prompt word. If the user performs a confirmation operation, the target behavior strategy is determined as the behavior strategy associated with the prompt word. If the user performs a denial operation, the user is displayed the selectable behavior strategies, and the behavior strategy selected by the user from the selectable behavior strategies is determined as the behavior strategy associated with the prompt word.

[0092] For example, in the case of matching to the target behavior strategy, a dialog box can be popped up on the user terminal to ask the user whether to determine the target behavior strategy as the behavior strategy associated with the prompt word. For example, the content displayed on the dialog box can be: whether to determine "selective area cleaning" as the behavior strategy associated with "wall-side square black object"? The user can select "yes" or "no". If the user selects "yes", "selective area cleaning" is determined as the behavior strategy associated with "wall-side square black object" so as to use the prompt word to identify the object meeting the requirement when controlling the cleaning robot to perform the selective area cleaning task, thereby determining the cleaning area. If the user selects "no" or there is no target behavior strategy matching the instruction description information in the preset selectable behavior strategy, the process jumps to the strategy selection interface, the selectable behavior strategies are displayed in the interface, and the user is prompted to select the behavior strategy associated with the prompt word.

[0093] In the case of the user terminal as the execution subject of the above prompt word generation method, after generating the prompt word and determining the behavior strategy associated with the prompt word, the above prompt word generation method can further include: sending the prompt word and the behavior strategy associated with the prompt word to the cleaning robot, so that the cleaning robot performs the behavior strategy associated with the prompt word when identifying the object meeting the prompt word. For example, the behavior strategy associated with the prompt word is an obstacle avoidance strategy, and the cleaning robot is controlled to perform the obstacle avoidance action after identifying the object meeting the prompt word. For another example, the behavior strategy associated with the prompt word is a selective area cleaning strategy, and the cleaning robot can be controlled to clean the preset area range around the object after identifying the object meeting the prompt word.

[0094] In the case of the cleaning robot as the execution subject of the above prompt word generation method, after generating the prompt word and determining the behavior strategy associated with the prompt word, the cleaning robot can perform the behavior strategy associated with the prompt word when identifying the object meeting the prompt word.

[0095] In order to facilitate understanding, an exemplary application scenario of the prompt word generation method of some embodiments of the present disclosure is described below.

[0096] When the user wants the cleaning robot to avoid the apple dropped on the ground, the user can input the instruction description information through the instruction description information input interface of the user terminal, select to input "please avoid the red apple dropped on the ground next time" through voice or text, or select to input a picture or a video containing the red apple dropped on the ground, or directly voice wake up the cleaning robot and say "please avoid the red apple dropped on the ground next time" to the cleaning robot. Taking the user terminal input as an example, the user terminal will generate the prompt word such as "red apple on the ground" of the instruction based on the above instruction description information input by the user, and determine the "obstacle avoidance strategy" as the behavior strategy associated therewith. The generated prompt word "red apple on the ground" and the associated behavior strategy "obstacle avoidance strategy" are sent to the cleaning robot. In this way, the cleaning robot will execute the obstacle avoidance action when it recognizes "red apple on the ground" in the subsequent task execution process.

[0097] FIG. 6 shows a flowchart of a prompt word generation method according to some embodiments of the present disclosure, which is executed by the user terminal 20 described above. As shown in FIG. 6, the prompt word generation method can at least include:

[0098] Step S201, obtaining the instruction description information input by the user;

[0099] Step S202, generating a prompt word adapted to the instruction description information based on the instruction description information, and determining a behavior strategy associated with the prompt word; and

[0100] Step S203, sending the prompt word and the behavior strategy associated with the prompt word to the cleaning robot, so that the cleaning robot executes the behavior strategy based on the prompt word.

[0101] It should be noted that the specific implementation process and technical effects of steps S201 to S203 can refer to the related description of the method embodiments in the above, which will not be described here.

[0102] According to some embodiments of the present disclosure, a control system of a cleaning robot is provided. FIG. 7 shows a structural schematic diagram of the control system according to some embodiments of the present disclosure. As shown in FIG. 7, the control system 1 can include a user terminal 20 and a cleaning robot 10. The user terminal 20 is in communication connection with the cleaning robot 10. For example, it can be connected through wireless communication modes such as Bluetooth or WIFI. The user terminal 20 is installed with a client, which can be used to control the cleaning robot 10 and show the status of the cleaning robot 10 to the user.

[0103] The user terminal 20 is configured to acquire instruction description information input by a user, generate a prompt word matched with the instruction description information based on the instruction description information, determine a behavior strategy associated with the prompt word, and send the prompt word and the associated behavior strategy to the cleaning robot 10. The cleaning robot 10 is configured to execute the behavior strategy associated with the prompt word based on the prompt word. The specific implementation process and effects can refer to the related description of the method embodiments above, and will not be repeated here.

[0104] According to some embodiments of the present disclosure, a cleaning robot is also provided, which includes a processor and a memory storing a computer program executable on the processor. The computer program, when executed by the processor, implements the steps of the prompt word generation method performed by the cleaning robot in the above method embodiments, and achieves the same technical effects. To avoid repetition, details will not be repeated here.

[0105] According to some embodiments of the present disclosure, a user terminal is also provided, which includes a processor and a memory storing a computer program executable on the processor. The computer program, when executed by the processor, implements the steps of the prompt word generation method performed by the user terminal in the above method embodiments, and achieves the same technical effects. To avoid repetition, details will not be repeated here. Exemplarily, the user terminal can be a mobile terminal such as a mobile phone, a tablet computer, or a wearable device (such as a smart watch or a bracelet).

[0106] According to some embodiments of the present disclosure, a computer readable storage medium is also provided. The computer readable storage medium includes computer instructions stored thereon, which, when executed by a processor, implement the steps of the prompt word generation method provided by any one of the above method embodiments, and achieve the same technical effects. To avoid repetition, details will not be repeated here. Exemplarily, the computer readable storage medium can be a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, etc.

[0107] According to some embodiments of the present disclosure, a computer program product is also provided, which includes computer instructions, which, when executed by a processor, implement the steps of the prompt word generation method provided by any one of the above method embodiments, and achieve the same technical effects. To avoid repetition, details will not be repeated here.

[0108] It should be noted that each embodiment in the present disclosure focuses on the difference from other embodiments, and the same or similar parts between various embodiments can be referred to each other. In the case of no conflict, the features in the same embodiment and different embodiments in the present disclosure can be combined with each other.

[0109] In a first aspect of the present disclosure, a prompt word generation method is provided, including: obtaining instruction description information; and generating a prompt word adapted to the instruction description information based on the instruction description information, and determining a behavior strategy associated with the prompt word, so that a cleaning robot executes the behavior strategy based on the prompt word.

[0110] In some embodiments, the obtaining of the instruction description information includes: obtaining user-inputted instruction description information; or collecting a picture or a video in a process in which the cleaning robot occurs a preset event, as the instruction description information, the preset event including a stuck event.

[0111] In some embodiments, the obtaining of the user-inputted instruction description information includes: presenting an input interface of the instruction description information to a user; and obtaining instruction description information inputted by the user in the input interface.

[0112] In some embodiments, the input interface provides one or more of the following input channels: a voice input channel, a text input channel, and an image input channel.

[0113] In some embodiments, the obtaining of the user-inputted instruction description information includes: collecting, by a voice module in the cleaning robot, voice information inputted by a user, and taking the voice information as the instruction description information.

[0114] In some embodiments, the generating of the prompt word adapted to the instruction description information based on the instruction description information includes: obtaining, based on the instruction description information and a pre-trained processing model, the prompt word adapted to the instruction description information.

[0115] In some embodiments, the obtaining of the prompt word adapted to the instruction description information based on the instruction description information and the pre-trained processing model includes: processing, by the processing model, the instruction description information, generating a candidate prompt word, and presenting the candidate prompt word to a user; and obtaining, in response to an operation performed by the user on the candidate prompt word, the prompt word adapted to the instruction description information.

[0116] In some embodiments, the obtaining of the prompt word adapted to the instruction description information based on the instruction description information and the pre-trained processing model includes: processing, by the processing model, the instruction description information, generating a candidate keyword, and presenting the candidate keyword to a user; and obtaining, based on a keyword filtered by the user from the candidate keyword, the prompt word.

[0117] In some embodiments, the prompt word is obtained based on the keyword selected by the user from the candidate keywords, including: in response to the user's selection operation on the keyword, retaining the keyword selected by the user from the candidate keywords; if the user triggers a keyword regeneration instruction, processing the instruction description information again through the processing model to generate new candidate keywords, and displaying the new candidate keywords and the retained keyword to the user for the user to select again; and if the user triggers a keyword confirmation instruction after selecting the keyword, obtaining the prompt word based on the keyword selected by the user.

[0118] In some embodiments, the prompt word is obtained based on the keyword selected by the user, including: concatenating the keyword selected by the user into a coherent description sentence, or processing the keyword selected by the user through a sentence model to convert the keyword selected by the user into a coherent description sentence; and obtaining the prompt word based on the description sentence.

[0119] In some embodiments, the prompt word is obtained based on the description sentence, including: displaying the description sentence to the user as a candidate prompt word; in response to a prompt word confirmation instruction triggered by the user, determining the candidate prompt word as the prompt word; or in response to a modification operation performed by the user on the candidate prompt word, obtaining a candidate prompt word modified by the user; and in response to a prompt word confirmation instruction triggered by the user, determining the modified candidate prompt word as the prompt word.

[0120] In some embodiments, the behavior policy associated with the prompt word is determined, including: displaying selectable behavior policies of the prompt word to the user; and determining the behavior policy selected by the user from the selectable behavior policies as the behavior policy associated with the prompt word.

[0121] In some embodiments, the selectable behavior policies of the prompt word are displayed to the user, including: in a case where the instruction description information is a picture or a video, displaying the selectable behavior policies of the prompt word to the user in a preset arrangement order.

[0122] In some embodiments, the selectable behavior policies of the prompt word are displayed to the user, including: in a case where the instruction description information is a voice or text, if there is a target behavior policy in the selectable behavior policies that matches the instruction description information, arranging the target behavior policy in front of the remaining behavior policies in the selectable behavior policies when displaying the selectable behavior policies of the prompt word to the user.

[0123] In some embodiments, the selectable behavior policies include at least two of the following policies: obstacle avoidance, selected area cleaning, deep cleaning, sorting and induction, and moving position.

[0124] In some embodiments, when the instruction description information is voice or text, the behavior strategy associated with the prompt word is determined by: if a target behavior strategy matching the instruction description information exists in the preset selectable behavior strategy, the target behavior strategy is determined as the behavior strategy associated with the prompt word; or the user is prompted to confirm whether the target behavior strategy is determined as the behavior strategy associated with the prompt word, if the user performs a confirmation operation, the target behavior strategy is determined as the behavior strategy associated with the prompt word, if the user performs a denial operation, the user is prompted to select a behavior strategy from the selectable behavior strategy, and the selected behavior strategy is determined as the behavior strategy associated with the prompt word.

[0125] In some embodiments, the instruction description information includes one or more of the following: text, voice, picture, and video.

[0126] In some embodiments, the prompt word is a prompt word of an object category related to the instruction description information.

[0127] In a second aspect of the present disclosure, a prompt word generation method is provided, including: obtaining instruction description information input by a user; generating a prompt word adapted to the instruction description information based on the instruction description information, and determining a behavior strategy associated with the prompt word; and sending the prompt word and the behavior strategy associated with the prompt word to a cleaning robot, so that the cleaning robot executes the behavior strategy based on the prompt word.

[0128] In a third aspect of the present disclosure, a control system of a cleaning robot is provided, including: a user terminal and a cleaning robot. The user terminal is configured to obtain instruction description information input by a user, generate a prompt word adapted to the instruction description information based on the instruction description information, determine a behavior strategy associated with the prompt word, and send the prompt word and the behavior strategy associated with the prompt word to the cleaning robot; and the cleaning robot is configured to execute the behavior strategy based on the prompt word.

[0129] In a fourth aspect of the present disclosure, a cleaning robot is provided, including a processor and a memory, the memory storing a computer program executable on the processor, the computer program implementing the steps of the prompt word generation method provided in the first aspect when executed by the processor.

[0130] In a fifth aspect of the present disclosure, a user terminal is provided, including a processor and a memory, the memory storing a computer program executable on the processor, the computer program implementing the steps of the prompt word generation method provided in the first aspect or the second aspect when executed by the processor.

[0131] In a sixth aspect of the present disclosure, a computer-readable storage medium is provided, including computer instructions stored thereon, which, when executed by a processor, implement the steps of the prompting word generation method provided in the first aspect or the second aspect.

[0132] The prompting word generation method of some embodiments of the present disclosure can generate an adapted prompting word and determine a behavior strategy associated with the prompting word according to the obtained instruction description information, so that the cleaning robot executes the behavior strategy associated with the prompting word, which can facilitate the user to issue instructions to the cleaning robot and is conducive to enriching the functions of the cleaning robot.

[0133] The above description is only a summary of the technical solutions provided by the present disclosure. In order to enable one of ordinary skill in the art to better understand the technical means of the present disclosure and implement it according to the content of the description, and in order to make the above and other characteristics and effects of the present disclosure more apparent and easy to understand, the following embodiments of the present disclosure are described.

[0134] Those skilled in the art should understand that the above discussion of any of the embodiments is only exemplary and is not intended to imply that the scope of the present disclosure is limited to these examples. Under the idea of the present disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes of different aspects of one or more embodiments of the present disclosure as described above. In order to be brief, they are not provided in details.

[0135] Although exemplary embodiments of the present disclosure have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the exemplary embodiments and all changes and modifications falling within the scope of the present disclosure.

Claims

1. A prompt word generation method, comprising: obtaining instruction description information; and generating a prompt word adapted to the instruction description information based on the instruction description information, and determining a behavior strategy associated with the prompt word, so that a cleaning robot executes the behavior strategy based on the prompt word. The obtaining of the instruction description information comprises:

2. The method of claim 1, wherein, obtaining user-inputted instruction description information; or collecting a picture or a video in a process in which the cleaning robot occurs a preset event as the instruction description information, the preset event comprising a trapped event. The obtaining of the user-inputted instruction description information comprises:

3. The method of claim 2, wherein, showing an input interface of the instruction description information to a user; and obtaining instruction description information inputted by the user in the input interface. The input interface provides one or more of the following input channels:

4. The method of claim 3, wherein, a voice input channel, a text input channel, and an image input channel. The obtaining of the user-inputted instruction description information comprises:

5. The method of claim 2, wherein, collecting voice information inputted by the user through a voice module in the cleaning robot, and taking the voice information as the instruction description information. The generating of the prompt word adapted to the instruction description information based on the instruction description information comprises:

6. The method of claim 1, wherein, obtaining the prompt word adapted to the instruction description information based on the instruction description information and a pre-trained processing model. The obtaining of the prompt word adapted to the instruction description information based on the instruction description information and the pre-trained processing model comprises:

7. The method of claim 6, wherein, processing the instruction description information through the processing model to generate candidate prompt words and show the candidate prompt words to a user; and obtaining the prompt word adapted to the instruction description information in response to an operation performed by the user on the candidate prompt words. The obtaining of the prompt word adapted to the instruction description information based on the instruction description information and the pre-trained processing model comprises:

8. The method of claim 6, wherein, processing the instruction description information through the processing model to generate candidate keywords and show the candidate keywords to a user; and obtaining the prompt word based on a keyword filtered by the user from the candidate keywords. The obtaining of the prompt word based on the keyword filtered by the user from the candidate keywords comprises:

9. The method of claim 8, wherein, in response to a filtering operation of the keyword by the user, retaining the keyword filtered by the user from the candidate keywords; if the user triggers a keyword regeneration instruction, processing the instruction description information through the processing model again to generate new candidate keywords, and showing the new candidate keywords and the retained keyword to the user for the user to filter again; and if the user triggers a keyword confirmation instruction after filtering the keyword, obtaining the prompt word based on the keyword filtered by the user. The obtaining of the prompt word based on the keyword filtered by the user comprises:

10. The method of claim 9, wherein, splicing the keyword filtered by the user into a coherent description sentence, or processing the keyword filtered by the user through a sentence model to convert the keyword filtered by the user into a coherent description sentence; and obtaining the prompt word based on the description sentence. The obtaining of the prompt word based on the description sentence comprises:

11. The method of claim 10, wherein, showing the description sentence to the user as a candidate prompt word; and ​ determine the candidate prompt word as the prompt word in response to a prompt word confirmation instruction triggered by the user; or obtain a candidate prompt word modified by the user in response to a modification operation performed by the user on the candidate prompt word; and determine the modified candidate prompt word as the prompt word in response to a prompt word confirmation instruction triggered by the user.

12. The method of any one of claims 1-11, wherein, determine the behavior strategy associated with the prompt word, including: present the user with selectable behavior strategies of the prompt word; and determine the behavior strategy selected by the user from the selectable behavior strategies as the behavior strategy associated with the prompt word.

13. The method of claim 12, wherein, present the user with selectable behavior strategies of the prompt word, including: present the user with selectable behavior strategies of the prompt word in a preset arrangement order in a case where the instruction description information is a picture or a video.

14. The method of claim 12, wherein, present the user with selectable behavior strategies of the prompt word, including: if there is a target behavior strategy matching the instruction description information in the selectable behavior strategies in a case where the instruction description information is a voice or text, arrange the target behavior strategy in front of the remaining behavior strategies in the selectable behavior strategies when presenting the user with the selectable behavior strategies of the prompt word.

15. The method of claim 12, wherein, the selectable behavior strategies include at least two of the following strategies: obstacle avoidance, region selection cleaning, deep cleaning, sorting and induction, and position moving.

16. The method of any one of claims 1-11, wherein, determine the behavior strategy associated with the prompt word in a case where the instruction description information is a voice or text, including: if there is a target behavior strategy matching the instruction description information in the preset selectable behavior strategies, determine the target behavior strategy as the behavior strategy associated with the prompt word; or confirm with the user whether to determine the target behavior strategy as the behavior strategy associated with the prompt word, if the user performs a confirmation operation, determine the target behavior strategy as the behavior strategy associated with the prompt word, and if the user performs a denial operation, present the user with the selectable behavior strategies and determine the behavior strategy selected by the user from the selectable behavior strategies as the behavior strategy associated with the prompt word.

17. The method of any one of claims 1-11, wherein, the instruction description information includes one or more of the following information: text, voice, picture, and video.

18. The method of any one of claims 1-11, wherein, the prompt word is a prompt word of an object category involved in the instruction description information.

19. A prompt word generation method, including: obtaining instruction description information input by a user; generating a prompt word adapted to the instruction description information based on the instruction description information, and determining a behavior strategy associated with the prompt word; and sending the prompt word and the behavior strategy associated with the prompt word to a cleaning robot, so that the cleaning robot executes the behavior strategy based on the prompt word.

20. A control system for a cleaning robot, comprising: a user terminal and a cleaning robot, the user terminal is configured to obtain instruction description information input by a user, generate a prompt word adapted to the instruction description information based on the instruction description information, determine a behavior strategy associated with the prompt word, and send the prompt word and the behavior strategy associated with the prompt word to the cleaning robot; and the cleaning robot is configured to execute the behavior strategy based on the prompt word. ​ 21. A cleaning robot comprising a processor and a memory, the memory storing a computer program executable on the processor, the computer program when executed by the processor implementing the steps of the method of any one of claims 1-18.

22. A user terminal comprising a processor and a memory, the memory storing a computer program executable on the processor, the computer program when executed by the processor implementing the steps of the method of any one of claims 1-19.

23. A computer readable storage medium comprising computer instructions stored thereon, the computer instructions when executed by a processor implementing the steps of the method of any one of claims 1-19.

Images