CN108115678B - Robot and its motion control method and device - Google Patents

Abstract

The invention discloses a robot and an action control method and device thereof. The method comprises the following steps: receiving a control instruction, wherein the control instruction is used for instructing the robot to execute a preset action; inputting the control instruction into a preset model to obtain an output result, wherein the output result comprises a behavior emotion mode corresponding to a preset action executed by the robot, and the preset model is obtained by training at least based on emotion evaluation parameters fed back by a user; and controlling the robot to execute a preset action according to the behavior emotion mode according to the output result. By the method and the device, the problem that the robot cannot meet the requirements of users due to single behavior is solved.

Description

Robot and its motion control method and device

technical field

The present invention relates to the field of robots, and in particular, to a robot and its motion control method and device.

Background technique

Since the development of robots, the functions have become more and more perfect, and the requirements for robots have become higher and higher. The existing reinforcement learning technology is widely used in robots, but most of them focus on motion decision-making, such as controlling balance and controlling robot walking. However, the decision-making model has not yet been deeply developed. In addition, the reinforcement learning (RL) technology itself is also developing, and it is far from mature. In principle, reinforcement learning is inspired by behavioral psychology. It was first embodied in experiments such as Pavlov's dog and Skinner box. It is also supported by theories such as neuroplasticity in biology. However, the reinforcement learning method does not touch the emotional training model, and there is no similar implementation on the robot. It stays at the level of computer vision, natural language processing, etc., and the robot cannot have different behaviors according to the user's preferences.

Aiming at the problem that the robot's behavior in the related technologies cannot meet the needs of users, there is no effective solution yet.

SUMMARY OF THE INVENTION

The main purpose of the present invention is to provide a robot and its motion control method and device, so as to solve the problem that a single behavior performance cannot meet the needs of users.

In order to achieve the above object, according to an aspect of the present invention, there is provided a robot motion control method, the method comprising: receiving a control instruction, wherein the control instruction is used to instruct the robot to perform a predetermined action; inputting the control instruction into a preset model to obtain Output results, wherein the output results include behavioral emotion patterns corresponding to the predetermined actions performed by the robot, and the preset models are trained based on at least the emotion evaluation parameters fed back by the user; control the robot to perform predetermined actions according to the behavioral emotions according to the output results.

Further, the behavior emotion mode is used to determine the behavior emotion state of the robot for the predetermined action, and when the robot performs the predetermined action, the state of the robot is the behavior emotion state.

Further, before inputting the control instruction into a preset model and obtaining an output result, the method further includes: receiving behavioral emotion indexes corresponding to a plurality of things in the object set, wherein the behavioral emotion index is used to indicate the robot's attitude towards the object set. The degree of preference of multiple things; a preset model is established according to the behavioral emotion index corresponding to multiple things in the set of things.

Further, the multiple things in the thing set are things corresponding to the predetermined actions performed by the robot.

其中，w_x为预设的基础参数，w_i为事物集中的第i项事物的影响参数，x_i为事物集中的第i项事物当前被设置的开关状态参数，b_i为事物集中的第i项事物的评价结果数值。Further, the number of things in the thing set is n, and n is an integer greater than 1; the calculation formula of the preset model is

Among them, w _x is the preset basic parameter, wi is the influence parameter of the _i -th thing in the thing set, x _i is the current switch state parameter of the i-th thing in the thing set, and b _i is the ith thing in the thing set. The value of the evaluation result of the i item.

Further, the output result also includes the expected value of the predetermined action, and after controlling the robot to perform the predetermined action according to the behavioral emotion pattern according to the output result, the method further includes: receiving a feedback result of the robot performing the predetermined action; updating the preset according to the feedback result and the expected value Model.

Further, the control instructions include at least one of the following: image control instructions; voice control instructions; biological signal control instructions.

In order to achieve the above object, according to another aspect of the present invention, a robot motion control device is further provided, the device comprising: a first receiving unit for receiving a control instruction, wherein the control instruction is used to instruct the robot to perform a predetermined action; The input unit is used for inputting the control instruction into the preset model to obtain the output result, wherein the output result includes the behavioral emotion pattern corresponding to the predetermined action performed by the robot, and the preset model is obtained by training at least based on the emotion evaluation parameters fed back by the user; The unit is used to control the robot to perform a predetermined action according to the behavioral emotion mode according to the output result.

Further, the behavior emotion mode is used to determine the behavior emotion state of the robot for the predetermined action, and when the robot performs the predetermined action, the state of the robot is the behavior emotion state.

Further, the device further includes: a second receiving unit, configured to receive the behavioral emotion index corresponding to a plurality of things in the thing set before inputting the control instruction into the preset model and obtaining the output result, wherein the behavioral emotion index is used to represent The robot's preference for multiple things in the thing set; the establishment unit is used to establish a preset model according to the behavioral emotion index corresponding to the multiple things in the thing set.

Further, the multiple things in the thing set are things corresponding to the predetermined actions performed by the robot.

其中，w_x为预设的基础参数，w_i为事物集中的第i项事物的影响参数，x_i为事物集中的第i项事物当前被设置的开关状态参数，b_i为事物集中的第i项事物的评价结果数值。Further, the number of things in the thing set is n, and n is an integer greater than 1; the calculation formula of the preset model is

Among them, w _x is the preset basic parameter, wi is the influence parameter of the _i -th thing in the thing set, x _i is the current switch state parameter of the i-th thing in the thing set, and b _i is the ith thing in the thing set. The value of the evaluation result of the i item.

Further, the output result also includes the expected value of the predetermined action, and the device further includes: a third receiving unit, configured to receive a feedback result for the robot to perform the predetermined action after controlling the robot to perform the predetermined action according to the behavioral emotion mode according to the output result; update the The unit is used to update the preset model according to the feedback results and expected values.

Further, the control instructions include at least one of the following: image control instructions; voice control instructions; biological signal control instructions.

In order to achieve the above object, according to another aspect of the present invention, a robot is also provided, and the robot includes: the robot motion control device according to the embodiment of the present invention.

The present invention receives control instructions, wherein the control instructions are used to instruct the robot to perform a predetermined action; input the control instructions into a preset model to obtain an output result, wherein the output result includes the behavioral emotion pattern relative to the predetermined action performed by the robot, and the preset It is assumed that the model is trained at least based on the emotional evaluation parameters fed back by the user; according to the output results, the robot is controlled to perform predetermined actions according to the behavioral emotion pattern, which solves the problem that the robot's single behavior cannot meet the user's needs, and then achieves different behaviors according to the user's needs. Effect.

Description of drawings

The accompanying drawings constituting a part of the present application are used to provide further understanding of the present invention, and the exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the attached image:

FIG. 1 is a flowchart of a robot motion control method according to a first embodiment of the present invention;

FIG. 2 is a flowchart of a robot motion control method according to a second embodiment of the present invention; and

FIG. 3 is a schematic diagram of a robot motion control apparatus according to an embodiment of the present invention.

Detailed ways

It should be noted that the embodiments in the present application and the features of the embodiments may be combined with each other in the case of no conflict. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

In order to make those skilled in the art better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only The embodiments are part of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the scope of protection of the present application.

It should be noted that the terms "first", "second", etc. in the description and claims of the present application and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances for the embodiments of the application described herein. Furthermore, the terms "comprising" and "having" and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those expressly listed Rather, those steps or units may include other steps or units not expressly listed or inherent to these processes, methods, products or devices.

Embodiments of the present invention provide a method for controlling robot motion.

Fig. 1 is a flow chart of a robot motion control method according to a first embodiment of the present invention. As shown in Fig. 1 , the method includes the following steps:

Step S102: receiving a control instruction, wherein the control instruction is used to instruct the robot to perform a predetermined action;

Step S104: Input the control instruction into the preset model to obtain an output result, wherein the output result includes the behavioral emotion pattern corresponding to the predetermined action performed by the robot, and the preset model is obtained by training at least based on the emotion evaluation parameters fed back by the user.

Step S106: Control the robot to perform a predetermined action according to the behavioral emotion mode according to the output result.

In this embodiment, the control instruction is received, and then the control instruction is input into the preset model to obtain the output result, and according to the output result, the robot is controlled to perform the predetermined action according to the behavioral emotion pattern. Since the preset model is obtained by training based on the emotion evaluation parameters fed back by the user, It can determine the behavioral emotion pattern of the robot when it performs a predetermined action. Therefore, controlling the robot to perform the predetermined action according to the behavioral emotion pattern according to the model output can solve the problem that the robot's behavior can not meet the needs of users, and then achieve different behaviors according to the needs of users. effect of behavior.

The robot in the embodiment of the present invention may be an intelligent service robot, a companion robot, etc. Optionally, the control instruction includes at least one of an image control instruction, a voice control instruction, and a biological signal control instruction, and the control instruction may also be of other types Instructions, receiving control instructions can be the robot detects the action of the owner through the camera, receives the image control instruction, or receives the voice control instruction, for example, receives the "dance" voice control instruction from the owner to the robot, the robot also It can receive wearable bio-signal control instructions. For example, some bio-signal detection equipment worn by the owner detects that the host's physical state is abnormal and sends out control instructions, and the robot receives bio-signal control instructions. The control command controls the robot to perform a predetermined action. The predetermined action can be the robot dancing, sweeping, turning in circles, etc. After receiving the control command, the control command is input into the preset model, and the model is calculated to obtain the output result, and control according to the obtained output result. The robot performs predetermined actions according to the behavioral emotion. Through the technical solutions of the embodiments of the present invention, the robot can interact with the user in multiple emotional modes.

Optionally, before inputting the control instruction into the preset model and obtaining the output result, the behavior emotion index corresponding to the multiple things in the thing set is received, wherein the behavior emotion index is used to indicate the robot's preference for the multiple things in the thing set. ; Establish a preset model according to the behavioral emotion index corresponding to multiple things in the thing set.

The establishment of the preset model requires the indices corresponding to the objects in the object set. The multiple objects in the object set can be things that can reflect the behavior and emotion of the robot, for example, things such as singing and dancing, sweeping the floor, etc. The indices corresponding to the objects in the object set can be It is the user's preference for the thing. For example, when the user cares more about a certain thing, the index is relatively high, and when the user does not care much about a certain thing, the index is relatively low. After receiving indices corresponding to multiple things in the thing set, a preset model is established according to the corresponding behavioral emotion indices. The behavioral emotion mode is used to determine the behavioral emotional state of the robot for a predetermined action. When the robot performs a predetermined action, the state of the robot is the behavioral emotional state, and the behavioral emotional state can correspond to the behavioral emotion mode. In a behavioral emotion mode, Can have a behavioral affective state.

Optionally, the output result includes the expected value of the predetermined action, and after controlling the robot to perform the predetermined action according to the behavior and emotion according to the output result, the feedback result of the robot's action is received; the preset model is updated according to the feedback result and the expected value.

The output result obtained by inputting the control instruction into the preset model includes the expected value of the predetermined action, and the expected value of the predetermined action can be used to represent the expected value of the behavior and emotion of the robot when performing the predetermined action, for example, whether the robot is happy when performing the predetermined action, After the robot performs a predetermined action according to the behavior and emotion, it receives the feedback result of the robot's action, and the feedback result is sent by the user. For example, the owner scores the robot's behavior, receives the feedback result of the robot's action, and then according to the feedback result The preset model is updated with the expected value of the predetermined action. For example, the preset model can be updated according to the difference between the feedback result and the expected value of the predetermined action.

其中，w_x为预设的基础参数，w_i为事物集中的第i项事物的影响参数，x_i为事物集中的第i项事物当前被设置的开关状态参数，b_i为事物集中的第i项事物的评价结果数值。Optionally, the number of things in the thing set is n, where n is an integer greater than 1, and the calculation formula of the preset model is:

Among them, w _x is the preset basic parameter, wi is the influence parameter of the _i -th thing in the thing set, x _i is the current switch state parameter of the i-th thing in the thing set, and b _i is the ith thing in the thing set. The value of the evaluation result of the i item.

其中，w_x为预设的基础参数，可以是一个定值，由用户设置或者出厂前设置，w_i为事物集中的第i项事物的影响参数，第i项事物的影响参数可以是第i项事物对主人的重要程度，x_i为事物集中的第i项事物当前被设置的开关状态参数，x_i具有两个数值，在取第一数值时，例如，x_i＝1，表示第i项事物对于主人的影响状态为开启，在取第二数值时，例如，x_i＝0，表示第i项事物对于主人的影响状态为关闭。在第i项事物对于主人的影响状态为关闭时，机器人对该事物的行为情感不会对主人产生影响。b_i为事物集中的第i项事物的评价结果数值，可以是主人对于机器人执行该事物时的行为情感的打分。The calculation formula of the preset model is

Among them, w _x is a preset basic parameter, which can be a fixed value, set by the user or set before leaving the factory, w _i is the influence parameter of the ith thing in the thing set, and the influence parameter of the ith thing can be the ith thing The importance of the item to the owner, x _i is the current switch state parameter of the i-th item in the transaction set, and x _i has two values. When taking the first value, for example, x _i =1, indicating that the i-th item is The influence state of the item on the owner is ON, and when the second value is taken, for example, x _i =0, it means that the influence state of the item i on the owner is OFF. When the influence state of the i-th thing on the owner is off, the behavior and emotion of the robot will not affect the owner. b _i is the value of the evaluation result of the i-th thing in the thing set, which can be the score of the owner's behavior and emotion when the robot executes the thing.

In an optional application scenario, after receiving the owner's voice command "sweep the floor", the robot obtains the behavioral emotion when sweeping the floor according to the preset model, and then performs the sweeping action. If the behavioral emotion mode output by the preset model is happy Mode, the robot shows a happy sweeping, for example, the action is brisk, and music is played at the same time, etc., the owner is very satisfied with the robot's behavior and emotion, and the score is high, then the robot can judge according to the owner's score, the owner likes to sweep the floor. Happy behavior and emotional state. Therefore, when the robot sweeps the floor next time, it still shows happy behavior and emotion. If the owner is tired of the happy behavior and emotion shown by the robot when sweeping the floor, the robot will be given a low score, and the robot will judge according to the score. If the owner does not like the happy behavioral emotional state shown when sweeping the floor, he will show an unhappy behavioral emotional state when sweeping the floor next time.

Fig. 2 is a flow chart of a robot motion control method according to a second embodiment of the present invention, which can be used as a preferred implementation of the above-mentioned first embodiment. As shown in Fig. 2, the robot motion control method includes the following steps:

Step S201: Receive a control instruction.

The control instruction may be one or more of a visual input signal (Vision input), a language input (Language input) and a wearable biosignal input (Wearable biosignal input).

Step S202: online reinforcement training.

Online reinforcement training (Soul Model Reinforcement Learning Core) can input control instructions into a preset model for calculation, obtain output results, and determine the behavior output of the robot according to the output results. Moreover, the preset model can be updated according to the feedback of the reward and punishment mechanism to realize online reinforcement training.

Step S203: Behavior output.

After the output result is obtained according to the preset model, the behavior output of the robot is controlled according to the output result.

Step S204: User evaluation feedback.

After receiving the user evaluation feedback (Human evaluation feedback), for example, the user scores the behavior of the robot.

Step S205: feedback, reward and punishment mechanism.

Through the introduction of reward and punishment mechanism feedback (Feedback Reward/punishment block), the preset model is revised to realize online reinforcement training.

Step S206: Expected value.

The expectation value is obtained according to the preset model, and the expectation value and the user feedback evaluation are used as the basis for the feedback reward and punishment mechanism, and the difference between the user feedback evaluation and the expectation value can be used as the feedback parameter.

The technical solution of the embodiment of the present invention adopts the principle of behavioral psychology in principle, and was first applied in experiments such as Pavlov's dog and Skinner box, and is also supported by theories such as neuroplasticity in biology. However, the reinforcement learning of the prior art does not involve the training model of emotion, and there is no similar implementation on the robot, which stays at the level of computer vision, natural language processing, etc. The embodiment of the present invention proposes the construction and implementation of the emotion model for the first time.

The present invention is about how to make the machine life better communicate with human beings, using online reinforcement learning to add psychological concepts of positive and negative reinforcement and positive and negative punishment for long-term training, the purpose is that the user can feel the interaction process with the robot. to close to human emotional feedback.

The reinforcement punishment theory of behavioral psychology covers positive and negative reinforcement and positive and negative punishment. The following is a brief introduction:

1. Positive reinforcement: Give a good stimulus. In order to establish an adaptive behavior pattern, use the reward method to make this behavior pattern recur and maintain. For example, companies award bonuses to employees who actively put forward reasonable suggestions.

2. Negative reinforcement: remove a bad stimulus. Set up to induce the appearance of the desired behavior. For example, the company does not allow personal phone calls during working hours. An employee who has this habit will be blamed as soon as this behavior occurs, but once he stops this behavior, the blame should be stopped immediately.

3. Positive Punishment: Applying a bad stimulus. This is a way of giving penalties when inappropriate behavior occurs.

4. Negative punishment: remove a good stimulus. This punishment is more commonly used than the positive punishment. When inappropriate behavior occurs, the original reward is no longer given.

Based on the basic idea of reinforcement learning, the feedback loop is reward-driven (referred to as RL), and the expectation can be expressed as wx+b, where b is the reward parameter, which is determined and assigned by the evaluation function. However, psychology believes that human behavior and emotion come from two aspects, namely good stimuli and bad stimuli. Human behavior can be understood as the multiplication and addition effect of complex good stimuli and complex bad stimuli, but the existing RL does not add this. Class classification mechanism, the technical solutions of the embodiments of the present invention add this mechanism and are applied to human-computer interaction robots. So the desired model of RL is w _x +(w ₁ x ₁ +b ₁ )+(w ₂ x ₂ +b ₂ )+…. On the basis of the original reinforcement learning, the dimension is extended. Regarding the newly added dimensions, part of it comes from presets. According to the formula, w ₁ x ₁ +b ₁ is expressed as the expected contribution model under the thing, w ₁ is a fixed value, preset by the designer, which is conceptually equivalent to the degree of human preference for a certain thing, and b ₁ represents is the reward and punishment parameter of the thing. If mopping the floor is a kind of thing, it is likely to be disgusting for children, then w ₁ is a negative number, x ₁ is 0 or 1 to represent the on-off state of the thing, and b ₁ represents the reward and punishment degree of the current thing. The set of things can be increased, but the preference of the thing needs to be preset. By synthesizing the expected contribution model of all things, the degree of reward and punishment of the individual at the current time point can be obtained and fed back to the input of the individual. The change of the model directly affects the training method of the model. Although the human-computer interaction robot can use the pre-trained transfer model, it is always criticized for its humanization. Therefore, the technical solution of the embodiment of the present invention uses the pre-trained transfer model. The transfer model is changed to an online learning method, allowing the user to personally train the robot, so that the robot's behavior pattern is gradually closer to the user, and finally serves the user.

It should be noted that the steps shown in the flowcharts of the accompanying drawings may be executed in a computer system, such as a set of computer-executable instructions, and, although a logical sequence is shown in the flowcharts, in some cases, Steps shown or described may be performed in an order different from that herein.

The embodiment of the present invention provides a robot motion control apparatus, and the apparatus can be used to execute the robot motion control method of the embodiment of the present invention.

FIG. 3 is a schematic diagram of a robot motion control device according to an embodiment of the present invention. As shown in FIG. 3 , the device includes:

a first receiving unit 10, configured to receive a control instruction, wherein the control instruction is used to instruct the robot to perform a predetermined action;

The input unit 20 is configured to input the control instruction into a preset model to obtain an output result, wherein the output result includes a behavioral emotion pattern corresponding to the predetermined action performed by the robot, and the preset model is at least based on the emotion evaluation fed back by the user parameters are trained;

The control unit 30 is configured to control the robot to perform the predetermined action according to the behavioral emotion mode according to the output result.

In this embodiment, a first receiving unit is used to receive a control instruction, wherein the control instruction is used to instruct the robot to perform a predetermined action; an input unit is used to input the control instruction into a preset model to obtain an output result, wherein the preset model uses In order to determine the behavioral emotion pattern when the robot performs a predetermined action, the preset model is obtained by training at least based on the emotion evaluation parameters fed back by the user; the control unit is used to control the robot to perform the predetermined action according to the behavioral emotion pattern according to the output result, thereby solving the behavioral performance of the robot. A single problem that cannot meet the needs of users, and then achieves the effect of showing different behaviors according to the needs of users.

Optionally, the behavioral emotion mode is used to determine the behavioral emotion state of the robot for the predetermined action, and when the robot performs the predetermined action, the state of the robot is the behavioral emotion state.

Optionally, the device further includes: a second receiving unit, configured to receive the behavior emotion index corresponding to a plurality of things in the thing set before inputting the control instruction into a preset model and obtaining an output result, wherein the behavior The emotion index is used to indicate the degree of preference of the robot to multiple things in the thing set; the establishment unit is used to establish the preset model according to the behavior emotion index corresponding to the multiple things in the thing set.

Optionally, the multiple things in the thing set are things corresponding to the predetermined action performed by the robot.

其中，w_x为预设的基础参数，w_i为事物集中的第i项事物的影响参数，x_i为事物集中的第i项事物当前被设置的开关状态参数，b_i为事物集中的第i项事物的评价结果数值。Optionally, the number of things in the thing set is n, and n is an integer greater than 1. The calculation formula of the preset model is:

Among them, w _x is the preset basic parameter, wi is the influence parameter of the _i -th thing in the thing set, x _i is the current switch state parameter of the i-th thing in the thing set, and b _i is the ith thing in the thing set. The value of the evaluation result of the i item.

Optionally, the output result includes an expected value of the predetermined action, and the device further includes: a third receiving unit, configured to receive a feedback result on the action of the robot after controlling the robot to perform the predetermined action according to the behavioral emotion mode according to the output result; an updating unit; , which is used to update the preset model based on the feedback results and expected values.

Optionally, the control instructions include at least one of the following: image control instructions; voice control instructions; biological signal control instructions.

An embodiment of the present invention further provides a robot, and the robot includes the robot motion control device of the embodiment of the present invention.

In the above-mentioned embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device, and they can be centralized on a single computing device or distributed in a network composed of multiple computing devices Alternatively, they can be implemented with program codes executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or they can be integrated into The multiple modules or steps are fabricated into a single integrated circuit module. As such, the present invention is not limited to any particular combination of hardware and software.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (13)

1. a robot motion control method, is characterized in that, comprises: receiving a control instruction, wherein the control instruction is used to instruct the robot to perform a predetermined action; Inputting the control instruction into a preset model to obtain an output result, wherein the output result includes a behavioral emotion pattern corresponding to the predetermined action performed by the robot, and the preset model is at least based on an emotional evaluation fed back by a user parameters are trained; Controlling the robot to perform the predetermined action according to the behavioral emotion pattern according to the output result; Before inputting the control instruction into a preset model and obtaining an output result, the method further includes: Receiving behavioral emotion indexes corresponding to multiple things in the thing set, wherein the behavioral emotion index is used to indicate the degree of preference of the robot to the multiple things in the thing set; The preset model is established according to the behavioral emotion index corresponding to a plurality of things in the thing set, Wherein, the behavioral emotion mode output by the preset model includes a happy mode and a behavioral emotion mode corresponding to an unhappy behavioral emotion state, and the predetermined action includes sweeping the floor. If the behavioral emotion mode output by the preset model is the happy mode , then the robot shows a happy sweeping, and the robot judges whether the owner likes the happy behavioral emotion pattern when sweeping the floor according to the owner's score. If so, it still shows a happy behavioral emotion pattern when sweeping the floor next time. No, show an unhappy behavioral emotional state the next time you sweep the floor. 2 . The method according to claim 1 , wherein the behavioral emotion mode is used to determine the behavioral emotion state of the robot for the predetermined action, and when the robot performs the predetermined action, the robot The state is the behavioral affective state. 3 . The method according to claim 1 , wherein a plurality of things in the thing set are things corresponding to the predetermined action performed by the robot. 4 . 4. The method according to claim 1, wherein the number of things in the thing set is n, and n is an integer greater than 1; the calculation formula of the preset model is

Wherein, w _x is a preset basic parameter, _wi is the influence parameter of the i-th thing in the thing set, x _i is the switch state parameter currently set for the i-th thing in the thing set, and b _i is The value of the evaluation result of the i-th thing in the thing set. 5 . The method according to claim 1 , wherein the output result further includes an expected value of the predetermined action, and after the robot is controlled to perform the predetermined action according to the behavioral emotion pattern according to the output result. 6 . , the method also includes: receiving a feedback result for the robot to perform the predetermined action; The preset model is updated according to the feedback result and the expected value. 6. The method according to claim 1, wherein the control instruction comprises at least one of the following: image control instructions; voice control commands; Biosignal control commands. 7. A robot motion control device, characterized in that, comprising: a first receiving unit, configured to receive a control instruction, wherein the control instruction is used to instruct the robot to perform a predetermined action; an input unit, configured to input the control instruction into a preset model to obtain an output result, wherein the output result includes a behavioral emotion pattern corresponding to the predetermined action performed by the robot, and the preset model is based on at least The emotional evaluation parameters of user feedback are trained; a control unit, configured to control the robot to perform the predetermined action according to the behavioral emotion pattern according to the output result, The second receiving unit is configured to receive the behavior emotion index corresponding to a plurality of things in the object set before inputting the control instruction into the preset model and obtaining the output result, wherein the behavior emotion index is used to represent the robot pair the degree of preference for a plurality of things in the collection of things; a establishing unit for establishing the preset model according to the behavioral emotion index corresponding to a plurality of things in the thing set, Wherein, the behavioral emotion mode output by the preset model includes a happy mode and a behavioral emotion mode corresponding to an unhappy behavioral emotion state, and the predetermined action includes sweeping the floor. If the behavioral emotion mode output by the preset model is the happy mode , then the robot shows a happy sweeping, and the robot judges whether the owner likes the happy behavioral emotion pattern when sweeping the floor according to the owner's score. If so, it still shows a happy behavioral emotion pattern when sweeping the floor next time. No, show an unhappy behavioral emotional state the next time you sweep the floor. 8 . The device according to claim 7 , wherein the behavioral emotion mode is used to determine the behavioral emotion state of the robot for the predetermined action, and when the robot performs the predetermined action, the robot The state is the behavioral affective state. 9 . The apparatus according to claim 7 , wherein the plurality of things in the thing set are things corresponding to the predetermined actions performed by the robot. 10 . 10 . The device according to claim 7 , wherein the number of things in the thing set is n, and n is an integer greater than 1; the calculation formula of the preset model is: 10 .

Wherein, w _x is a preset basic parameter, _wi is the influence parameter of the i-th thing in the thing set, x _i is the switch state parameter currently set for the i-th thing in the thing set, and b _i is The value of the evaluation result of the i-th thing in the thing set. 11. The apparatus according to claim 7, wherein the output result further comprises an expected value of the predetermined action, and the apparatus further comprises: a third receiving unit, configured to receive a feedback result for the robot to perform the predetermined action after controlling the robot to perform the predetermined action according to the behavioral emotion pattern according to the output result; An update unit, configured to update the preset model according to the feedback result and the expected value. 12. The apparatus according to claim 7, wherein the control instruction comprises at least one of the following: image control instructions; voice control commands; Biosignal control commands. 13. A robot, comprising the robot motion control device according to any one of claims 7 to 12.

Images