Accompanying robot control method and accompanying robot

Abstract

The invention relates to the technical field of robots, in particular to an accompanying robot control method and an accompanying robot. The accompanying robot control method comprises the following steps: presetting a series of mood values based on different expressions; a series of interactive behaviors are preset, and different interactive behaviors correspond to different mood values; obtaining the current performance of the target, and automatically adjusting the current mood value according to the obtained current performance; outputting a corresponding interactive behavior according to the current mood value; in the scheme, the accompanying robot can adjust the own mood value by sensing the external emotion, that is, the accompanying robot can adjust the own emotion by sensing the external emotion and make the interactive behavior corresponding to the current emotion, so that not only can the target emotion be captured, but also the own emotion can be owned, and real bidirectional emotion interaction is realized; and the deep condition sharing between the accompanying robot and the target is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of robots, and in particular to a control method for a companion robot and a companion robot. Background Art

[0002] A companion robot is an intelligent device designed specifically to provide emotional support, daily companionship, and social interaction, especially suitable for the elderly, children, people living alone, or those in need of emotional comfort. It generally has functions such as human-computer interaction, health and safety monitoring, etc. However, existing companion robots can usually only perform command interactions with users, and they do not have emotional changes themselves, unable to carry out two-way emotional interactions with users, and it is difficult to achieve deep empathy between the companion robot and the user.

[0003] Therefore, designing a control method for a companion robot and a companion robot that can achieve two-way emotional interaction is crucial for those skilled in the art. Summary of the Invention

[0004] Embodiments of the present invention provide a control method for a companion robot and a companion robot that can achieve two-way emotional interaction to solve the problem in the prior art that it is difficult to actively understand and respond to users.

[0005] A control method for a companion robot, the solution is that it includes the following steps:

[0006] Preset a series of mood values based on different expressions;

[0007] Preset a series of interaction behaviors and correspond different interaction behaviors to different mood values;

[0008] Obtain the current expression of the target and automatically adjust the current mood value according to the obtained current expression;

[0009] Output the corresponding interaction behavior according to the current mood value.

[0010] Optionally, obtain the current expression of the target through a camera, a sensor, or remote input;

[0011] Take the current expression of the target, the interaction record, and the environmental data as feature inputs to the transform model at both ends of the multimodal transform algorithm, and calculate and analyze the output emotional state in combination with its MOE architecture;

[0012] Automatically adjust the current mood value according to the emotional state analysis.

[0013] Optionally, it further includes the following steps:

[0014] According to the emotional state analysis, determine whether the current expression is a positive expression or a negative expression,

[0015] If the current performance is a positive performance, the current mood value is automatically increased; if the current performance is a negative performance, the current mood value is automatically decreased.

[0016] Optionally, the method further includes the following steps:

[0017] Establish a pairing mechanism and perform friend binding through the pairing mechanism.

[0018] Optionally, the method further includes the following steps:

[0019] Establish a communication protocol and perform remote communication with the bound friend according to the communication protocol.

[0020] Optionally, the interaction behavior includes one or more of a language output behavior, a visual output behavior, an action feedback behavior, a tactile feedback behavior, and a biological feedback behavior.

[0021] Optionally, the method further includes the following steps:

[0022] Preset an optimal temperature threshold;

[0023] Obtain the current temperature. If the current temperature exceeds the optimal temperature threshold, the current mood value is automatically decreased.

[0024] To solve the problems existing in the prior art, the present invention further provides a companion robot. The solution is that the companion robot is used to implement the companion robot control method described in any one of the above, and includes: a camera, a sensing module, a voice module, a main control module, and an output module. The camera, the sensing module, the voice module, and the output module are all connected to the main control module. The camera and the sensing module are used to obtain the current performance of the target, the voice module is used to collect the target sound, the main control module is used to connect to an external server to upload the target current performance and the target sound to the external server, and control the output module to output corresponding interaction behaviors according to the server instructions.

[0025] Optionally, the output module includes one or more of a display, a speaker, a light, a fan, a heating element, and a robotic arm, and the sensing module includes one or more of a temperature sensor, a biological sensor, and an infrared sensor.

[0026] Optionally, it further includes a base, and a data transmission interface is provided on the base for connecting to an external server.

[0027] Compared with the prior art, the beneficial effects of the companion robot control method provided by the embodiments of the present invention are as follows: By designing a companion robot control method including the following steps: preset a series of mood values based on different expressions; preset a series of interaction behaviors and correspond different interaction behaviors to different mood values; obtain the current expression of the target and automatically adjust the current mood value according to the obtained current expression; output the corresponding interaction behavior according to the current mood value; the companion robot can adjust its own mood value by sensing the external emotion, that is, adjust its own emotion by sensing the external emotion and make interaction behaviors corresponding to the current emotion, which can not only capture the target emotion, but also have its own emotion, realize true two-way emotional interaction, and achieve deep empathy between the companion robot and the target. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The following will further describe the solution of the present invention in detail with reference to the drawings. In the drawings:

[0029] Figure 1 is the flowchart of the companion robot control method provided by the embodiments of the present invention Figure One ;

[0030] Figure 2 is the flowchart of the companion robot control method provided by the embodiments of the present invention Figure Two ;

[0031] Figure 3 is the flowchart of the companion robot control method provided by the embodiments of the present invention Figure Three ;

[0032] Figure 4 is the flowchart of the companion robot control method provided by the embodiments of the present invention Figure Four ;

[0033] Figure 5 is the structural block diagram of the companion robot provided by the embodiments of the present invention.

[0034] The reference numerals in the drawings are as follows:

[0035] 100, camera; 200, sensing module; 300, voice module; 400, main control module; 500, output module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. Now, with reference to the drawings, the preferred embodiments of the present invention will be described in detail.

[0037] As Figures 1 to 4 shown, the present invention provides a specific embodiment of a companion robot control method.

[0038] A control method for a companion robot, refer to Figure 1 , the control method for the companion robot includes the following steps:

[0039] S1. Preset a series of mood values based on different expressions;

[0040] S2. Preset a series of interaction behaviors and correspond different interaction behaviors to different mood values;

[0041] S3. Obtain the current expression of the target and automatically adjust the current mood value according to the obtained current expression;

[0042] S4. Output the corresponding interaction behavior according to the current mood value.

[0043] Specifically, a series of mood values matching the companion robot are set based on different external expressions. The emotions of external expressions can be recognized by using natural language processing technologies such as sentiment analysis algorithms, or the emotional expressions of the external world can be recognized by facial expression recognition algorithms and their emotions can be inferred. Voice analysis technologies such as speech emotion recognition algorithms can also be used to recognize the emotions of external expressions. Additionally, dialogue generation, machine learning algorithms, and models can be used to train the companion robot to recognize the emotions corresponding to different external expressions. After obtaining a large number of external expressions and recognizing the corresponding emotions, mood values with corresponding scores increasing from low to high are set according to the progressive levels of emotions.

[0044] Furthermore, the mood values correspond to different emotions of the companion robot, such as extremely low, low, flat, pleasant, extremely pleasant, etc.; corresponding interaction behaviors are set for the companion robot according to different mood values. For example, when the companion robot is in a pleasant mood, it can perform interaction behaviors such as laughing, flashing, jumping, clapping, making a heart gesture, or raising its own temperature to the human body temperature. When the companion robot is in a low mood, it can perform interaction behaviors such as crying, covering its head, wiping tears, or lowering its temperature.

[0045] After presetting a series of rules, the companion robot can obtain the current expression of the target in real time under the current mood value, and adjust its own mood value according to the current expression of the target, so as to perform the interaction behavior corresponding to the new mood value, making the interaction of the robot more consistent with the current emotion of the target, thereby realizing two-way emotional interaction.

[0046] Although existing companion robots also have functions such as human-computer interaction, health and safety monitoring, etc., existing companion robots usually can only perform command interactions with users, and they do not have emotional changes themselves, so they cannot perform two-way emotional interactions with users, and it is difficult to achieve deep empathy between the companion robot and the user.

[0047] In this embodiment, a control method for a companion robot includes the following steps: preset a series of mood values based on different performances; preset a series of interaction behaviors and correspond different interaction behaviors to different mood values; obtain the current performance of the target and automatically adjust the current mood value according to the obtained current performance; output the corresponding interaction behavior according to the current mood value. The companion robot can adjust its own mood value by sensing external emotions, that is, adjust its own emotions by sensing external emotions and make interaction behaviors corresponding to the current emotions. It can not only capture the target's emotions but also have its own emotions, realizing true two-way emotional interaction and achieving deep empathy between the companion robot and the target.

[0048] In one embodiment, the mood value is used to represent the companion robot's own emotions. Multiple emotions can be set, and the emotions can include but are not limited to plain, joy, sadness, anger, fear, disgust, surprise, anxiety, shame, guilt, excitement, and the multiple emotions are corresponded to mood values with scores from 0 to 100.

[0049] In the actual use process, the initial mood value of the companion robot can be preset to 50 points, so that the companion robot is in a plain mood level initially. During the contact with the outside world, the companion robot can automatically increase or decrease its own mood value according to the external performance. When the companion robot encounters positive performances, it will automatically increase its own mood value, so that it enters positive emotions such as joy and excitement from the plain emotion, and then makes more enthusiastic interaction behaviors. When the companion robot encounters negative performances, it will automatically decrease its own mood value, so that it enters emotions such as sadness, anger, fear, disgust, surprise, anxiety, shame, guilt from the plain emotion, and thus makes relatively cold interaction behaviors.

[0050] In one embodiment, referring to Figure 2 ,"obtaining the current performance of the target" includes the following steps:

[0051] S31. Obtain the current performance of the target through a camera, a sensor or remote input;

[0052] S32. Use the current performance of the target, interaction records and environmental data as feature inputs for the multi-modal transform algorithm to the transform model, and calculate and analyze the output emotional state in combination with its MOE architecture;

[0053] S33. Automatically adjust the current mood value according to the emotional state analysis.

[0054] Specifically, a camera can be configured for the companion robot to obtain the current performance images of the user or other companion robots through the camera, and detect and recognize the current postures and facial expressions of the other party through computer vision algorithms, so as to obtain the current performance of the user or other companion robots; the current performance of the user or other companion robots can also be obtained through sensor induction. When two companion robots approach each other face to face, they can be sensed and recognized through sensors, so as to obtain the current performance of the corresponding companion robot. The current performance of the companion robot can be vibration, light emission, sound, etc. The sound can be human language, or other simulated speaking sounds. The ambient temperature, light brightness, user body temperature, user distance, etc. can also be sensed through sensors to obtain the current external performance. The current performance of the user or other companion robots can also be obtained through remote input. For example, multiple companion robots can be connected to the same local network. When multiple companion robots are not in the same area, group chat can be established, and text transmission or induction transmission can be carried out in the group chat mode; the scenario description or performance description remotely input by the user can also be obtained through the communication module to obtain the current performance.

[0055] After the companion robot obtains the current performance of the target, it combines multi-modal data and large models for emotional state analysis, that is, integrates various types of data inputs, and uses deep learning models to understand and predict the emotional state of the target. In this embodiment, the obtained current performance, previous interaction records, and environmental data are used as key feature data and input into the transform model, and the emotional state is calculated and analyzed through the MOE architecture in the transform model, and the emotional state is output.

[0056] Among them, the transform model is a deep learning architecture based on the self-attention mechanism. After the current performance, previous interaction records, and environmental data are used as key feature data and input into the transform model, the transform model will extract features from images, texts, and sensors respectively, and learn the interaction between the three through the cross-modal attention mechanism. For example, the weight of the user frowning in a high-noise environment is higher, etc. Finally, possible emotion state classifications are output, such as plain, joy, sadness, anger, fear, disgust, surprise, anxiety, shame, guilt, excitement, etc.

[0057] Among them, the MOE architecture is a sparse neural network architecture that allows different input data to be processed by different sub-networks through dynamic routing, so as to greatly improve the model capacity while maintaining computational efficiency, and reduce the cost of training and inference.

[0058] After analyzing the target's emotions, the companion robot automatically adjusts its own mood value and outputs corresponding interactive behaviors based on the adjusted mood value. For example, after sensing the target's joy, the companion robot's mood value will increase and it will exhibit positive interactive behaviors. After sensing the target's sadness, the companion robot's mood value will decrease and it will exhibit negative interactive behaviors. When two companion robots are close to each other and face each other, they can communicate through sensing, such as vibration, light, and sound. The sound can be human language or other simulated speech sounds. During the communication process, the two companion robots will synchronously adjust their own mood values by recognizing each other's performance. For example, when the two companion robots engage in positive interaction, the mood values of both companion robots will increase, thereby exhibiting more enthusiastic interaction. After the two companion robots engage in negative interaction, the mood values of both companion robots will decrease, thereby exhibiting more indifferent interaction.

[0059] It should be noted that when the companion robot obtains the current performance of the user or other companion robots through the camera, it can simulate the same actions or expressions as the user or other companion robots; when abnormal behavior is identified, such as falling or suddenly falling to the ground, special measures can be taken to respond.

[0060] In one embodiment, reference Figure 3 Automatically adjusting the current mood value based on the current performance includes the following steps:

[0061] S331. According to the emotional state analysis, determine whether the current performance is positive or negative.

[0062] S332. If the current performance is positive, the current mood value is automatically increased; if the current performance is negative, the current mood value is automatically decreased.

[0063] Specifically, after analyzing the target emotion through the large model, the current performance is determined to be positive or negative based on the target emotion. When the target emotion is positive, it can be considered that the target's current performance is positive, thereby automatically increasing the current mood value. When the target emotion is negative, it can be considered that the target's current performance is negative, thereby automatically decreasing the current mood value.

[0064] A series of performance standards can also be preset within the companion robot as a basis for judging positive and negative performances. If the current performance of the target obtained does not exceed the performance standard, the current performance of the target is determined to be positive. If the current performance of the target obtained exceeds the performance standard, the current performance of the target is determined to be negative. For example, touch force standards, speech standards, task completion standards, energy standards, moving distance standards, etc. can be set.

[0065] When the touch is relatively gentle and the intensity does not exceed the touch intensity standard, it is determined that the current performance is a positive performance. When the touch is relatively rough and the intensity exceeds the touch intensity standard, it is determined that the current performance is a negative performance. When the words are relatively friendly and meet the speech standard, it is determined that the current performance is a positive performance. When the words are relatively rude and do not meet the speech standard, it is determined that the current performance is a negative performance; when the actual learning duration does not reach the set learning duration task after setting the learning duration task, it does not meet the task completion standard, that is, it is determined that the current performance is a negative performance. When the learning duration reaches the pre-set learning duration, that is, after completing the learning duration task, it is determined that the current performance is a positive performance.

[0066] In one embodiment, the interaction behavior includes one or more of language output behavior, visual output behavior, motion feedback behavior, tactile feedback behavior, and biological feedback behavior.

[0067] Specifically, the language output behavior can be an active greeting or daily chatting. For example, it can trigger voices such as "Good morning", "Good evening", "Good night" according to time or specific scenarios, or analyze the target's living habits based on the target performance and make relevant greetings that conform to the target's living habits, and adjust the tone and manner of language output in combination with the current mood level of the companion robot.

[0068] The visual output behavior includes screen display, lighting, projection, etc. The screen display can make anthropomorphic expressions, such as blinking, smiling, frowning, pupil dilation, etc. The screen display can also play celebration special effects and display large subtitle text content, etc. The lighting can be indicator light changes or emotional light effects, etc., showing different indicator light states and different light effects at different mood levels.

[0069] The motion feedback behavior includes head expressions, upper limb movements, lower limb movements, emotional movements, etc. When making an instruction confirmation or a selection query, it can nod to indicate agreement or correctness and shake the head to indicate refusal or error. When it recognizes that the target is speaking, it can tilt the head to indicate normal listening or thinking and quickly tilt the head when feeling curious. When saying hello or goodbye, it can swing the upper limb from side to side when raising it. When a pointing action is needed, it can use the finger to guide. When it needs to move along with the target, it can walk or jump with the lower limbs.

[0070] Among them, the emotional movements include celebration movements, comfort movements, withdrawal reactions, etc. The celebration movements can be raising both arms in combination with lighting flashes and verbal praise, quickly spinning around and playing happy sound effects. The comfort movements can be slowly leaning forward and stretching out the upper limb to gently pat the other person's shoulder or gently hug the other person, etc. The withdrawal reaction generally occurs in a negative or extremely negative state and can be lowering the head and hanging the arms, etc.

[0071] In one embodiment, special states such as being sick or depressed can also be additionally set. For example, if the companion robot is in an extremely low or low mood level for a long time, special states such as being sick or depressed can be triggered. For example, when the mood value of the companion robot is lower than the value corresponding to the neutral mood and lasts for a certain period of time, it enters the sick state. When the mood value of the companion robot is lower than the value corresponding to the neutral mood and lasts for a certain period of time, it enters the depressed state. After the companion robot enters the sick or depressed state, more negative interaction behaviors will occur, such as standby.

[0072] In one of the embodiments, by establishing a pairing mechanism and a communication protocol, the companion robot can be paired with other companion robots through the pairing mechanism and perform friend binding. After the two companion robots are bound as friends, they can have close contact and can also perform remote communication through the communication protocol. During the close contact and remote communication processes, both companion robots can adjust their own mood values according to the current performance of the other party, so as to output different interaction behaviors to achieve the interaction between the two companion robots.

[0073] Furthermore, a group can be established using the communication protocol and multiple companion robots can be added to the same group to achieve remote communication between multiple companion robots.

[0074] In one of the embodiments, referring to Figure 4 , the following steps are further included:

[0075] S51. Preset the optimal temperature threshold;

[0076] S52. Obtain the current temperature. If the current temperature exceeds the optimal temperature threshold, automatically reduce the current mood value.

[0077] Specifically, the optimal temperature threshold can be the comfortable temperature range that the human body can accept. When the temperature exceeds this range or is lower than this range, the companion robot will automatically reduce its current mood value to simulate the mood change of the human body in an extreme temperature environment. For example, when it is recognized that the external environment is too cold, the companion robot will automatically reduce its current mood value and automatically restore its own mood value after putting on clothes or wrapping a scarf and sensing that the temperature has returned to the optimal temperature threshold; when it is recognized that the external environment is too hot, the companion robot will automatically reduce its current mood value and automatically restore its own mood value after turning on the air conditioner or fan and sensing that the temperature has dropped and returned to the optimal temperature threshold.

[0078] As Figure 5 shown, the present invention also provides a specific embodiment of a companion robot.

[0079] A companion robot, referring to Figure 5, the companion robot is used to implement the above-mentioned companion robot control method. The companion robot includes a camera 100, a sensing module 200, a voice module 300, a main control module 400, and an output module 500. The camera 100, the sensing module 200, the voice module 300, and the output module 500 are all connected to the main control module 400. The camera 100 and the sensing module 200 are used to obtain the current performance of the target, the voice module 300 is used to collect the target sound, and the main control module 400 is used to connect to the external server 10 to upload the target current performance and the target sound to the external server 10, and control the output module to output corresponding interaction behaviors according to the instructions of the external server 10.

[0080] Specifically, the camera 100, the sensing module 200, and the voice module 300 are all used to obtain the external performance. The camera 100 can capture images of the user's body movements and expressions, and can also capture images of the body movements and expressions of other companion robots. The sensing module 200 includes one or more of a temperature sensor, a biological sensor, and an infrared sensor. The sensing module 200 can sense the touch, temperature, etc. of the user or other companion robots. The voice module 300 can capture the words of the user or other companion robots.

[0081] The main control module 400 is mainly used to connect to the external server 10 and upload the target current performance and the target sound to the external server 10. The external server 10 can extract the emotional characteristics of the target through methods such as facial action coding, voice emotion mapping, or behavior pattern matching. When there are multiple performances, the final emotion can be determined through multi-modal emotion fusion. In a simple scenario, the weighted voting method can be used to combine expressions, language, and sensing to determine the final emotion label. In a complex scenario, deep learning fusion can be used to process and analyze multi-modal data using the transform model to output the final emotion label. After analyzing and outputting the emotion label, the external server will output the corresponding control instruction to the main control module 400; the main control module 400 automatically increases or decreases the own mood value of the companion robot according to the control instruction. After the mood value of the companion robot is adjusted, the main control module 400 controls the current mood value of the companion robot to the output module to output corresponding interaction behaviors.

[0082] The output module 500 is mainly used for the output of interactive behaviors. The output module 500 includes one or more of a display, a speaker, lights, a fan, a heating element, and a robotic arm. The display can convey emotions through dynamic facial expressions (such as blinking, smiling, etc.), and can also visualize information, display weather, schedules, health data, etc., and can also provide interactive guidance, providing a touch operation interface or a teaching demonstration; the speaker can give voice reminders, answer questions, and can also convey emotions by playing laughter, soothing music, etc., and can also simulate natural sounds to relieve the loneliness of the target, such as bird chirping, rain sounds, etc.; the lights can be used for status indication, can also intensify the expression of emotions, and can also create a scene atmosphere; the fan can blow a gentle breeze in coordination with the rhythm of the voice to simulate a "breathing feeling"; the heating element can adjust the tactile temperature to simulate body temperature when being hugged or actively hugging; the robotic arm can be used to transfer items, perform switch operations, wave to say hello, applaud to celebrate, provide support when detecting a fall, etc.

[0083] In one embodiment, the companion robot further includes a base. The base can provide electrical energy for the companion robot to ensure that the companion robot can be charged when installed on the base to ensure that the companion robot has sufficient power; the base can also be connected to the local network. The companion robot can join the local network by itself or through the base. The base stores the data generated by the local network. When other companion robots are placed on the base, they can automatically access the network and obtain the data retained in the local network.

[0084] Furthermore, a data transmission interface is provided on the base for connecting to an external server. When the companion robot is placed on the base, data can be synchronized, software can be updated, information can be backed up, etc.

[0085] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0086] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0087] It should be understood that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Those skilled in the art can modify the technical solutions recorded in the above embodiments or equivalently replace some of the technical features; and all such modifications and replacements should fall within the protection scope of the appended claims of the present invention.

Claims

1. A control method for a companion robot, characterized in that, It includes the following steps: Preset a series of mood values based on different performances; Preset a series of interaction behaviors and correspond different interaction behaviors to different mood values; Obtain the current performance of the target and automatically adjust the current mood value according to the obtained current performance; Output the corresponding interaction behavior according to the current mood value.

2. The companion robot control method according to claim 1, wherein, It also includes the following steps: Obtain the current performance of the target through a camera, a sensor or remote input; Use the current performance, interaction record and environmental data of the target as feature inputs to the transform model at the multi-modal transform algorithm end, and calculate and analyze the output emotional state in combination with its MOE architecture; Automatically adjust the current mood value according to the emotional state analysis.

3. The companion robot control method according to claim 2, wherein It also includes the following steps: Judge whether the current performance is a positive performance or a negative performance according to the emotional state analysis, If the current performance is a positive performance, automatically increase the current mood value; if the current performance is a negative performance, automatically decrease the current mood value.

4. The companion robot control method according to claim 1, characterized in that, It also includes the following steps: Establish a pairing mechanism and perform friend binding through the pairing mechanism.

5. The companion robot control method according to claim 4, characterized in that, It also includes the following steps: Establish a communication protocol and perform remote communication with the bound friend according to the communication protocol.

6. The companion robot control method according to claim 1, characterized in that, It also includes the following steps: Preset the optimal temperature threshold; Obtain the current temperature. If the current temperature exceeds the optimal temperature threshold, automatically decrease the current mood value.

7. The companion robot control method according to claim 1, wherein The interaction behavior includes one or more of a language output behavior, a visual output behavior, an action feedback behavior, a tactile feedback behavior, and a biological feedback behavior.

8. A companion robot, characterized in that, The companion robot is used to implement the companion robot control method as described in any one of claims 1-7, and includes: a camera, a sensing module, a voice module, a main control module, and an output module. The camera, the sensing module, the voice module, and the output module are all connected to the main control module. The camera and the sensing module are used to obtain the current performance of the target. The voice module is used to collect the target sound. The main control module is used to connect to an external server to upload the target current performance and the target sound to the external server, and control the output module to output the corresponding interaction behavior according to the server instruction.

9. The companion robot according to claim 8, characterized in that, The output module includes one or more of a display, a speaker, a light, a fan, a heating element, and a robotic arm. The sensing module includes one or more of a temperature sensor, a biological sensor, and an infrared sensor.

10. The companion robot according to claim 8, wherein, It also includes a base, and a data transmission interface is arranged on the base for connecting to an external server.

Images