MOBILE BODY, MOBILE BODY CONTROL METHOD, AND INFORMATION PROCESSING APPARATUS

Abstract

The present invention encourages the creation of a new operation of a moving body. This moving body is equipped with: a mixing parameter setting unit for setting a mixing parameter; an operation creation unit configured so as to be capable of creating a third operation by mixing a pre-set first operation as an operation of the moving body and a pre-set second operation as an operation of the moving body which differs from the first operation with one another according to a reflection degree corresponding to a mixing parameter set by the mixing parameter setting unit; and an operation control unit configured so to be capable of causing the operation actually implemented by the moving body to reflect the third operation created by the operation creation unit.

Description

[Technical Field]

[0001] The present disclosure relates to a mobile object, a method for controlling a mobile object, and an information processing device. [Background technology]

[0002] A mobile body is known in which executable actions are predetermined (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2019 / 087478 Summary of the Invention

[0004] When trying to make a moving object perform a new action, creating this new action from scratch is a time-consuming and laborious task, and for an inexperienced creator, it is also an expensive task. Furthermore, since there is a limit to the creator's own ideas, there is a limit to the actions that can be created.

[0005] The present disclosure aims to provide a moving body, a method for controlling a moving body, and an information processing device that can encourage the creation of new actions. [Means for solving the problem]

[0006] A moving body according to one embodiment of the present disclosure includes a mixing parameter setting unit that sets mixing parameters, an action creation unit that is configured to be able to create a third action by mixing a first action that is predetermined as a moving body action and a second action that is predetermined as a moving body action different from the first action with a degree of reflection according to the mixing parameter set by the mixing parameter setting unit, and an action control unit that is configured to be able to reflect the third action created by the action creation unit in an action actually performed by the moving body. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a perspective view showing the appearance of a moving object according to an embodiment of the present disclosure. [Figure 2] FIG. 2 is a schematic diagram showing the configuration of a movable part provided in the moving body according to the embodiment. [Figure 3] FIG. 3 is a block diagram showing the overall configuration of a control system for a moving body according to the embodiment. [Figure 4] FIG. 4 is a block diagram showing the configuration of a portion of the control system according to the embodiment, which is involved in setting a mixing parameter and creating a new operation based on the mixing parameter. [Figure 5] FIG. 5 is a block diagram showing the configuration of a part of the control system according to the embodiment that is involved in learning and proposing mixed parameters. [Figure 6] FIG. 6 is a schematic diagram showing the appearance of a user interface used to set mixing parameters in the control system according to the embodiment. [Figure 7] FIG. 7 is a schematic diagram showing the appearance of a user interface used to propose mixing parameters (present proposed parameters) in the control system according to the embodiment. [Figure 8] FIG. 8 is a schematic diagram showing the appearance of a user interface used for proposing mixing parameters and for the user to adjust the mixing parameters in the control system according to the embodiment. [Figure 9] FIG. 9 is a schematic diagram showing the appearance of a user interface according to a modified example, used for proposing mixing parameters and for the user to adjust the mixing parameters, in the control system according to the embodiment. [Figure 10] FIG. 10 is a flowchart showing the flow of setting the mixing parameters and controlling the operation based on the mixing parameters as the operation of the control system according to the embodiment. [Figure 11] FIG. 11 is a flowchart showing the flow of learning of the mixed parameters as an operation of the control system according to the embodiment. [Figure 12]FIG. 12 is a flowchart showing the flow of proposing mixing parameters and adjusting the mixing parameters by the user as an operation of the control system according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. The embodiment described below is a specific example of the present disclosure, and is not intended to limit the technology of the present disclosure to the specific aspects described below. Furthermore, the arrangement, dimensions, and dimensional ratios of each component in the following embodiment are not limited to the examples shown in the drawings.

[0009] The explanation will be given in the following order. 1.Basic configuration 1.1. Mobile Configuration 1.2.Configuration of moving parts 2. Control system configuration and operation 3. Flowchart explanation 4. Action and Effects 5. Summary

[0010] <1.Basic configuration> (1.1. Configuration of moving body) FIG. 1 is a perspective view showing the appearance of a moving object 1 according to an embodiment of the present disclosure.

[0011] The moving body 1 according to this embodiment is an autonomous quadrupedal walking robot modeled after an animal (specifically, a dog). The moving body 1 comprises a head 11, a neck 12, a torso 13, a tail 14, two front legs (right front leg 15r and left front leg 15l), and two rear legs (right rear leg 16r and left rear leg 16l). Joints, which are the movable parts of the moving body 1, are formed between these parts, and can be driven in predetermined directions by actuators installed at each joint. The number of actuators installed at each joint is determined depending on the direction in which the joint can be driven, and in this embodiment, electric motors (servo motors) are used as the actuators.

[0012] The head 11 of the moving object 1 is equipped with a display 11a, a speaker 11b, and a microphone 11c. The display 11a is provided in a portion of the head 11 that corresponds to the dog's eyes, and the speaker 11b is provided in a portion that corresponds to the mouth. FIG. 1 shows the general positions of the speaker 11b and the microphone 11c (described below) on the head 11. In this embodiment, the display 11a is configured to display a video or image representing the dog's eyes, and the display 11a and the speaker 11b are used as means for attaching media associated with the actions performed by the moving object 1. The microphone 11c is provided in or near a portion that corresponds to the dog's ears. Both the mouth and ears are equipped with actuators that can be operated, allowing the actions or gestures of the moving object 1 to resemble those of a real dog.

[0013] (1.2. Configuration of moving parts) FIG. 2 is a schematic diagram showing the configuration of a movable part provided in the moving body 1 according to this embodiment.

[0014] In this embodiment, the joints that are movable include a joint j1 between the head 11 and the neck 12, a joint j2 between the neck 12 and the torso 13, a joint j3 between the front part (i.e., the chest) and the rear part (i.e., the waist) of the torso, a joint j41 between the thigh of the right front foot 15r and the torso 13, a joint j42 between the thigh of the right front foot 15r and the lower knee, a joint j51 between the thigh of the left front foot 15l and the torso 13, a joint j52 between the thigh of the left front foot 15l and the lower knee, a joint j61 between the thigh of the right hind foot 16r and the torso 13, a joint j62 between the thigh of the right hind foot 16r and the lower knee (no symbol shown in Figure 2), a joint j71 between the thigh of the left hind foot 16l and the torso 13, and a joint j72 between the thigh of the left hind foot 16l and the lower knee.

[0015] Actuators capable of imparting a predetermined degree of freedom to each of these joints are installed, enabling the moving object 1 to perform movements that mimic those of a real dog. For example, by imparting three degrees of freedom to the head 11, it is possible for the moving object 1 to perform movements including nodding and tilting its head, and by imparting two degrees of freedom to the thigh of the leg (for example, the right front leg 15r), it is possible for the moving object 1 to perform movements including walking and opening both legs outward. By combining the movements of these joints, it is possible to achieve more natural movements that are closer to those of a real dog.

[0016] 2. Control system configuration and operation FIG. 3 is a block diagram showing the overall configuration of a control system S for the moving body 1 according to this embodiment.

[0017] The control system S controls the movement of the moving object 1, specifically the rotation angle of the electric motor provided as an actuator in each joint. The control system S is roughly divided into a mechanism unit 101, a motion blender unit 102, and an actuator controller 103, and except for the actuator controller 103, it is provided separately from the moving object (i.e., the main body of the moving object) 1. In this embodiment, the moving object is composed of the main body 1 of the moving object and the control system S, and hereinafter, the term "moving object" refers to the main body 1 of the moving object. In addition to the above, the control system S also includes an interface device 104.

[0018] The mechanitage unit 101 sets blending parameters and creates new actions for the moving object 1 based on the set blending parameters. As an example, the mechanitage unit 101 and the motion blender unit 102 described below can be implemented in a computer that is a server SVR, and can be installed in a location remote from the moving object 1. The mechanitage unit 101 and the moving object 1 (in this embodiment, the actuator controller 103) can communicate wirelessly or via a network line.

[0019] The motion blender unit 102 learns the user's preferences regarding mixing parameters, and presents the learned mixing parameters (i.e., mixing parameters according to the user's preferences) to the user as suggested parameters. If the suggested parameters presented by the motion blender unit 102 are close to the user's preferences, the user can select the suggested parameters as mixing parameters and specify the mixing parameters.

[0020] The actuator controller 103 generates drive signals for the actuators to cause the moving body 1 to execute the new motion created by the mechanism unit 101. The actuator controller 103 can be separated from the mechanism unit 101 and the motion blender unit 102 and built into the moving body 1.

[0021] The interface device 104 displays the proposed parameters presented by the motion blender unit 102 and prompts the user to specify mixed parameters. The user can specify mixed parameters of their own choice via the interface device 104 for the presented proposed parameters, that is, mixed parameters that will cause the moving object 1 to perform an action that is closest to their preference. The interface device 104 can be embodied by a personal computer owned or used by the user, and the specification of mixed parameters can be promoted by displaying the proposed parameters on the screen of the personal computer.

[0022] FIG. 4 is a block diagram showing the configuration of the mechanism unit 101 according to this embodiment.

[0023] The mechanism unit 101 includes a mixed parameter setting unit B11, a motion creation unit B12, a motion control unit B13, and a media presentation unit B14.

[0024] The mixing parameter setting unit B11 sets a mixing parameter. The mixing parameter is an index indicating the degree to which a second existing action (i.e., second action) to be mixed is reflected in a first existing action (i.e., first action) that serves as the basis for creating a new action. The first action and the second action are mutually different actions, and are predetermined, for example, as a time-series set of rotation angles of an electric motor that is an actuator, stored in a memory unit of the mechanism unit 101, and can be read out appropriately by the mechanism unit 101. The first action is basically one action, but the second action may be one action or multiple actions. When there are multiple second actions, a mixing parameter is set for each of the multiple second actions.

[0025] In this embodiment, the first motion is a "body sway" motion in which the torso 13 is swayed from side to side, and the following two motions are used as the second motions: a "here, doggy doggy" motion in which the front legs 15r and 15l are used to dig in the ground, and a "butt shake" motion in which the rear half of the waist part of the torso 13 is swayed relative to the front half of the chest part. There is no particular limit to the number of second motions, and there may be two or more.

[0026] In this embodiment, the degree of reflection of the second action is gradually increased and gradually decreased at the start and end of the second action. Referring to FIG. 6, to achieve this operation, the following confusion parameters are used: a time TM for continuing the reflection of the second action; a ratio BR of the second action to the first action when the second action is reflected; a time Tfis for starting the reflection of the second action; a time Tfoe for ending the reflection of the second action; a time Tfie for ending the gradual increase in the degree of reflection; and a time Tfos for starting the gradual decrease in the degree of reflection. FIG. 6 shows the mixture parameters specified for the second second action BM22 (i.e., "butt shake") according to this embodiment. Here, the time Tfis for starting the reflection of the second action can be interpreted as the time when the gradual increase in the degree of reflection begins, and the time Tfoe for ending the reflection of the second action can be interpreted as the time when the gradual decrease in the degree of reflection ends.

[0027] The action creation unit B12 mixes the second action with the first action with a degree of reflection according to the confusion parameter set by the confusion parameter creation unit B11, to create a new action (third action) that is different from both the first and second actions.

[0028] The movement control unit B13 reflects the third movement created by the movement creation unit B12 in the movement actually performed by the moving object 1. Although not limited to this, in this embodiment, the reflection of the third movement is achieved by having the moving object 1 execute the third movement itself. A signal from the movement control unit B13 is input to the actuator controller 103 as a signal indicating a time series set of rotation angles of the actuators (electric motors), and the actuator controller 103 converts the input signal into a drive signal for each actuator and outputs it to the actuator to be controlled.

[0029] The media presentation unit B14 presents predetermined media perceivable by the user in association with the third action. In this embodiment, the predetermined media is visual and auditory media, specifically, image information imitating a dog's eyes and audio information imitating a dog's bark. The image information of the dog's eyes can be presented by the display 11a provided on the head 11, and the audio information of the dog's bark can be presented by the speaker 11b.

[0030] In this embodiment, media is set in advance in association with the first action that is the basis for creation and the second action that is the target of mixing, and the media set in association with the action that has a higher degree of reflection in the third action, that is, the action that has a greater degree of reflection in the third action, is presented. For example, if the ratio BR of the second action to the first action when reflecting it exceeds 1, the media set in association with the second action is presented.

[0031] The presentation of media is not limited to this, and it is also possible to present media set in association with the action of the first or second action that has a higher final reflection degree in the third action. For example, if the time Tfie at which the gradual increase in the reflection degree of the second action ends or the time Tfoe at which the reflection of the second action ends is later than the end time Te of the first action, the media set in association with the second action is presented.

[0032] FIG. 6 shows a specific example of a screen UI1 (hereinafter sometimes referred to as an "interface screen") displayed by the interface device 104 according to this embodiment for specifying mixing parameters.

[0033] The interface screen UI1 has a plurality of regions R11, R12, and R13 that are separated from one another. The regions R11 to R13 may be divided into different screen planes, or may be divided into different regions on the same screen plane by interposing black display areas between them (the same applies to the following description). Of the plurality of regions R11 to R13, the first is region R11 that shows the appearance of the moving object 1 and a demonstration of an action performed by the moving object 1. The second is region R12 that shows information related to media, and the third is region R13 that shows information related to a first action and a second action. As information related to the first action, it is possible to display the time Ts at which the first action starts and the time Te at which the first action ends, and as information related to the second action, it is possible to display the mixing parameters of the second action. The second region R12 includes, as display units for displaying information related to the media, a display unit Eye indicating the start and end times of displaying an image of a dog's eyes on the display 11a, and display units Snd1 and Snd2 indicating the start and end times of presenting the dog's bark on the speaker 11b. In this embodiment, multiple dog barks (display units Snd1 and Snd2) are set as the dog barks that can be presented. For the barks, it is also possible to display the waveform of the bark. As described above, the mixing parameters include the time TM for continuing the reflection of the second action, the ratio BR of the second action to the first action when reflecting the second action, the time Tfis for starting the reflection of the second action, the time Tfoe for ending the reflection of the second action, the time Tfie for ending the gradual increase in the reflection level, and the time Tfos for starting the gradual decrease in the reflection level. The third region R13 includes, as display units for displaying information related to the first and second actions, a display unit BM1 for showing a profile of the first action, and display units BM21 and BM22 for showing profiles of the second action. The user can change the mixing parameters by changing the motion profile displayed in the third region R13 by operating a cursor on the screen, etc. The time for which the media is presented can also be changed by operating a cursor on the screen, etc. In this embodiment, the first region R11 is further provided with a display section Da1 that displays the content of the media, specifically, an image of a dog's eyes that is actually displayed on the display 11a.

[0034] 5 is a block diagram showing the configuration of the motion blender unit 102 according to this embodiment. For convenience, an interface device 104 is also shown in FIG.

[0035] The motion blender unit 102 includes a designation facilitating unit B21, a designation recognizing unit B22, a learning processing unit B23, and an operation mode selecting unit B24.

[0036] The specification promotion unit B21 prompts the user to specify the mixing parameters. In this embodiment, the specification of the mixing parameters is promoted by displaying a screen for specifying the mixing parameters on the interface device 104. The user can specify the mixing parameters via the screen displayed on the interface device 104.

[0037] In this embodiment, the specification promotion unit B21 displays mixing parameters that are candidates for specification by the user as suggested parameters on the screen of the interface device 104, and prompts the user to specify the mixing parameters. The specification of the mixing parameters is achieved by the user selecting the suggested parameters. The number of suggested parameters displayed may be one or more. It is also possible to prompt the user to directly input the mixing parameters, regardless of the display of the suggested parameters.

[0038] The designation promoting unit B21 further presents the proposed parameters by displaying the motions created in accordance with the candidate mixed parameters on the screen of the interface device 104. In other words, a demonstration image of the candidate mixed parameters is displayed.

[0039] Here, if a motion created according to a candidate mixing parameter places a load exceeding the allowable range on an actuator installed at any joint of the moving object 1, the designation promoting unit B21 restricts the mixing parameter related to the motion of that joint. For example, if an excessively large change in rotation angle occurs in the joint j1 between the head 11 and the neck 12 in a motion created by mixing a second motion, causing acceleration exceeding the allowable range at the joint j1 and placing a load exceeding the allowable range on the actuator installed at the joint j1, the designation promoting unit B21 sets the mixing parameter to 0 and excludes the second motion from the motions to be mixed. An example of a second motion that places a load exceeding the allowable range on the actuator can be a first motion in a sitting position compared to a second motion in a standing position. The restriction of the mixing parameter is not limited to excluding the problematic second motion from the motions to be mixed. It is also possible to restrict the mixing parameter by reducing the ratio BR of the second motion to the first motion when reflecting the second motion from the original ratio, or by extending the time from the start of reflection of the second motion to the end of the gradual increase in the reflection degree (= Tfie - Tfis) from the original time, thereby reflecting the second motion more gradually.

[0040] The operation mode selection unit B22 sets an operation mode according to the user's selection. The operation mode can be selected by displaying a screen for operation mode selection on the interface device 104, and the user can select the operation mode via the screen displayed on the interface device 104. The specification promotion unit B21 changes the proposed parameters according to the selected operation mode.

[0041] Examples of operational modes include operational modes with functional significance, such as an operational mode that is less likely to cause malfunctions and an operational mode that reduces battery consumption, and operational modes with personality or emotional significance, such as an angry operational mode and a happy operational mode.

[0042] As for the former operating mode, an operating mode that is less likely to cause malfunctions can be realized as an operating mode that reduces the acceleration occurring in the joints, and an operating mode that reduces battery consumption can be realized as an operating mode that reduces posture changes and reduces the load on the actuator.

[0043] Furthermore, as the latter motion mode, the angry motion mode can be realized as a motion mode that suppresses joint movement and makes the motion feel sluggish, and the happy motion mode can be realized as a motion mode that increases joint movement and makes the motion feel aggressive.

[0044] FIG. 7 shows a specific example of an interface screen UI2 displayed by the interface device 104 according to this embodiment regarding the proposal of mixing parameters (presentation of proposed parameters).

[0045] The interface screen UI2, like the interface screen UI1 described above, has multiple regions R21, R22, and R23 separated from one another. Of these multiple regions R21 to R33, the first is region R21, which indicates selectable operation modes for a specific combination of first and second operations. The second is region R22, which indicates a demonstration of an operation by the moving object 1 for each proposed parameter presented. The third is region R23, which allows the user to select a proposed parameter. In this embodiment, the operation mode can be selected from three options. Meanwhile, for example, the first region R21 has a "Mode A" button for selecting a normal operation mode, a "Mode B" button for selecting an operation mode that is less likely to cause malfunctions, and a "Mode C" button for selecting an operation mode that reduces battery consumption. The second region R22 is divided into three display sections. The first is a display unit MM1 that displays a motion based on a first proposed parameter presented by the motion blender unit 102. The second is a display unit MM2 that displays a motion based on a second proposed parameter. The third is a display unit MM3 that displays a motion based on a third proposed parameter. The first of the proposed parameters that can be employed can be the average value of all blending parameters that can be specified for a specific combination of the first and second motions. The second can be a blending parameter obtained via a trained model (described later). The third can be any of the blending parameters that can be specified for a specific combination of the first and second motions. The user can compare the proposed parameters by observing the motions of the moving object 1 displayed in sync with each other in the second area R22. The third area R23 has a "Select" button associated with each proposed parameter. The user can select a proposed parameter that best suits their preferences by pressing the "Select" button.In this embodiment, instead of being limited to single selection, pressing the "Select" button for multiple proposed parameters is permitted. The order in which the "Select" button is pressed can be used to rank the user's selections (specifically, first, second, and third). It is also possible to suspend selection by pressing the "Suspend" button or to select no proposed parameters by pressing the "No Contest" button. By ranking the user's selections, it is possible to label the mixed parameters used in machine learning, i.e., the training data, based on this ranking. This reduces the burden of selecting only one of the presented proposed parameters as training data, shortens the time required for labeling, and simplifies the labeling process itself.

[0046] The designation recognition unit B23 recognizes, as a designated parameter, a mixing parameter designated by the user via the interface device 104. The recognized designated parameter is input to the mixing parameter setting unit B11 of the mechanism unit 101 and set as the mixing parameter.

[0047] The learning processing unit B24 learns user preferences regarding the designation of mixture parameters. Specifically, machine learning is performed to determine training data regarding mixture parameters for a second action to be mixed with a first action, and a trained model of the mixture parameters is created. After the trained model is created, the designation promoting unit B21 presents the mixture parameters obtained by the learning processing unit B24 via the trained model as suggested parameters. As a result, the mixture parameters obtained via the trained model are included in the targets for designation by the user.

[0048] In this embodiment, supervised learning using a neural network is performed as machine learning. A typical neural network applicable to learning may be a neural network having a plurality of nodes arranged in an input layer, a plurality of nodes arranged in a hidden layer, and a plurality of nodes arranged in an output layer. The number of nodes arranged in the output layer is equal to the number of mixture parameters. The number of hidden layers may be one or more. Typically, a network having two hidden layers is adopted, with a plurality of nodes arranged in a first hidden layer and a plurality of nodes arranged in a second hidden layer. Weights indicating the strength of synaptic connections between the input layer nodes and the hidden layer nodes, and between the hidden layer nodes and the output layer nodes, are set, respectively. Based on this neural network configuration, the learning processing unit B24 learns the correlation between the first and second operations, which are input layer variables, and the mixture parameters, which are output layer variables. Specifically, the output layer node values ​​corresponding to the input layer node values ​​are calculated using a general output layer calculation method using an activation function, and the calculated output layer node values ​​are compared with the values ​​of the training data to calculate the error between them. The learning processing unit B24 adjusts the weighting of synaptic connections to reduce this error. The weighting can be adjusted, for example, by the backpropagation method. The above process is repeated a predetermined number of times, or learning is terminated when the error falls within a predetermined range, and the trained model is determined using the weightings finally obtained.

[0049] In this embodiment, in creating a trained model, predetermined mixed parameters are presented to the user, and the user can adjust the mixed parameters to match their preferences. In this case, the learning processing unit B24 performs machine learning using the adjusted mixed parameters as training data.

[0050] FIG. 8 shows a specific example of an interface screen UI3 displayed by the interface device 104 according to this embodiment regarding the proposal of mixing parameters and the adjustment of mixing parameters by the user.

[0051] The interface screen UI3 has multiple regions R31, R32 separated from one another. The first of these multiple regions R31, R32 is region R31, which shows a demonstration of an action by the moving object 1 for each proposed parameter to be presented, and includes a display unit MM2, similar to the second region R22 in the interface screen UI2, that displays an action based on mixed parameters obtained via a trained model. The second is region R32, which shows information about the first action and the second action, and includes a display unit BM1, similar to the third region R13 in the interface screen UI1, that displays the start time Ts and end time Te of the first action, and display units BM21, BM22 that display mixed parameters for the second action. In this embodiment, profiles of the first and second movements related to the proposed parameters selected in the first area R31 are displayed in the second area R32, and the user can observe and compare the posture of the moving body 1 at each moment in the movements based on each proposed parameter in the first area R31 by moving the scroll bar B shown in the second area R32 back and forth.

[0052] FIG. 9 shows an interface screen UI4 according to a modified example, which is displayed by the interface device 104 according to the present embodiment in relation to the proposal of mixing parameters and the adjustment of mixing parameters by the user.

[0053] The interface screen UI4 has multiple regions R41 to R43 separated from one another. Among these regions R41 to R43, the first is region R41, which displays a demonstration of an action based on the selected proposed parameters. The second is region R42, which displays information about the first and second actions, and is similar to the third region R13 in the interface screen UI1 and the second region R32 in the interface screen UI3. The third is region R43, which displays the proposed parameters presented by the motion blender unit 102 in two dimensions. The multidimensional proposed parameters can be converted into two dimensions using, for example, t-SNE (t-SNE). In this embodiment, the third region R43 can display, as proposed parameters, an average value A of all mixture parameters that can be specified for a specific combination of the first and second actions, a mixture parameter B obtained via a trained model, and an arbitrary mixture parameter C among the mixture parameters that can be specified for a specific combination of the first and second actions. This allows the user to understand the relationship between the multiple proposed parameters presented based on the two-dimensional parameter display and select the one that best suits their taste. The arbitrary mixture parameter C can be completely arbitrary, or it can be extracted from a vector space close to an arbitrary mixture parameter previously presented (proposed parameter C), or it can be a mixture parameter C' with an extreme value.

[0054] <3. Explanation using a flowchart> 10 is a flowchart showing the flow of setting mixing parameters and controlling operations based on the mixing parameters as an operation of the control system according to this embodiment. The process according to the flowchart in FIG. 10 is executed by the mechanism unit 101 of the control system S each time a third operation is generated.

[0055] In S101, a first existing action (first action) that is the basis for creation is selected.

[0056] In S102, a second existing action (second action) to be mixed is selected.

[0057] In S103, the mixing parameters are set.

[0058] In S104, media to be attached to the newly created third action is selected.

[0059] In S105, the third action is created by mixing the second action with the first action at a reflection degree according to the mixing parameter.

[0060] In S106, based on the third action and the media associated therewith, drive signals for the actuators provided in the respective joints are generated, and drive signals for the display 11a and the speaker 11b are also generated.

[0061] 11 is a flowchart showing the flow of learning of blending parameters as an operation of the control system according to this embodiment. The processing according to the flowchart in FIG. 10 is executed by the motion blender unit 102 of the control system S each time blending parameters are learned.

[0062] In S201, a data set for learning is input. The data set includes the types of first motion (e.g., "body swaying") and second motion (e.g., "butt shaking") used for learning, and mixing parameters selected by the user for the first and second motions.

[0063] In S202, machine learning is performed using the types of the first and second actions as input layer variables and the mixture parameters selected by the user as training data.

[0064] In S203, a trained model is generated or a trained model generated in the past is updated.

[0065] 12 is a flowchart showing the operation of the control system according to this embodiment, which illustrates the flow of proposing mixing parameters and adjusting the mixing parameters by the user. The processing according to the flowchart in FIG. 12 is executed by the motion blender unit 102 of the control system S each time a mixing parameter is specified, that is, each time a third motion is created.

[0066] In S301, the types of the first and second actions selected by the user are input.

[0067] In S302, the operation mode selected by the user is input.

[0068] In S303, the proposed parameters are calculated using the trained model.

[0069] In S304, it is determined whether or not the loads applied to the actuators installed on the respective joints of the moving object 1 are within the allowable ranges in the operation calculated using the proposed parameters as mixed parameters. If all the loads are within the allowable ranges, the process proceeds to S306, and if any of the loads exceeds the allowable range, the process proceeds to S305.

[0070] In S305, the proposed parameters are excluded from the targets of actual proposals.

[0071] In S306, the proposed parameters are presented to the user, urging the user to specify the mixing parameters.

[0072] In S307, the mixing parameters designated by the user are recognized.

[0073] In S308, it is determined whether or not the user has adjusted the specified mixing parameters. If the user has adjusted the parameters, the process proceeds to S309; ​​if the user has not adjusted the parameters, the process proceeds to S310.

[0074] In S309, the mixing parameters are changed to the adjusted mixing parameters.

[0075] In S310, a dataset is created and saved. The created dataset is reflected in subsequent learning.

[0076] <4. Action and Effects> When a user (e.g., a creator or designer) wants a mobile object to perform a new action, creating this new action from scratch is a time-consuming and laborious task, and for an inexperienced creator, this can be costly. Furthermore, since there are limitations to the creator's own ideas, there is a limit to the actions that can be created. The problem becomes even more pronounced when trying to give a certain value to the new action. The value that is given to an action is a value that brings substantial benefits to the user of the mobile object (i.e., the consumer). Examples of such valuable actions include carrying heavy luggage to help the user, monitoring the room to reassure the user, or behaving in a cute manner to impress the user.

[0077] According to this embodiment, a new third motion is created by combining predetermined first and second motions, and this third motion can be reflected in the motion actually performed by the moving object 1, thereby reducing the effort and time required to create and realize a new motion and saving costs. Furthermore, it is possible to encourage the user to create motions that are not limited by the limits of their own imagination, in other words, that go beyond the limits of their own imagination.

[0078] By making it possible to present media in association with the third action, it is possible to encourage the user to understand the value that is intended to be attributed to the third action and to clarify the value that is intended to be attributed to the third action.

[0079] By prompting the user to specify mixing parameters, recognizing the mixing parameters specified by the user as specified parameters and making it possible to reflect them in the mixing of the second movements (i.e., the creation of the third movement), it becomes possible to reflect the user's preferences in the creation of the third movement through the specification of mixing parameters. Here, by making it possible to create the third movement by mixing the second movements with a reflection degree according to the mixing parameters specified by the user, it becomes possible to reduce the burden on the user required to create the third movement and save time because it is not necessary to specify the movements of all moving parts individually.

[0080] By presenting candidate mixture parameters as suggested parameters and allowing the user to specify them, the third action created based on the specified mixture parameters (specified parameters) can be modified as necessary to create a third action that suits the user's preferences, further reducing the burden on the user.

[0081] By performing machine learning for the first and second actions using the mixture parameters specified by the user as training data and making it possible to calculate the mixture parameters using the created trained model, it is possible to propose mixture parameters that reflect the user's preferences by simply specifying the first and second actions (in other words, mixture parameters for creating a third action that is close to the user's preferences), thereby further reducing the effort and time required to create the third action.

[0082] <5. Summary> The embodiments of the present disclosure have been described above in detail with reference to the drawings. According to the embodiments of the present disclosure, it is possible to prompt the user to create a new action for a moving object.

[0083] The technology of the present disclosure is not limited to the above specific embodiments, and various modifications are possible, and combinations of modifications are also possible. In the above description, an animal (specifically, a dog)-type moving body is used as the moving body, but the moving body that can be used may be a moving body that imitates the form of an animal or living thing other than a dog, including a human-type moving body, and may be a moving body that is movable not only by the function of an animal or living thing's organs (e.g., legs) but also by wheels. In this specification, wheels are positioned as a means for enabling movement, and are not limited to ring-shaped bodies configured to rotate around an axle, but are understood as a concept that includes axle-less rolling bodies such as ball casters and band-shaped bodies such as caterpillars.

[0084] Furthermore, in the above explanation, the mobile object (i.e., the main body of the mobile object) 1 and the control unit S are configured separately, but this configuration is not limiting, and the functions of the control unit S can be integrated into the main body 1 of the mobile object, and the learning, proposing and setting of the mixed parameters, as well as the creation of the third action, can be performed by a computer provided in the main body 1 of the mobile object. In this case, the interface device through which the user specifies or selects the mixed parameters can be embodied by a personal computer.

[0085] Furthermore, in the above description, a plurality of proposed parameters or a plurality of actions based on presented proposed parameters are displayed in parallel on the interface device 104. However, it is also possible to display actions based on individual proposed parameters in an overlapping manner, which allows the user to easily understand the difference in actions for each proposed parameter.

[0086] Furthermore, in the above description, a moving body with a physical existence, such as a robot, is used as the object to be operated, and the third action is reflected in a change in the orientation or posture of the moving body itself (i.e., the movement of the moving body). However, the object that can be used is not limited to this, and the object may be a virtual object. For example, a display unit may be provided in the interface device, and the object and its movement may be displayed on this display unit. In this case, the interface device may be realized by a smartphone or tablet computer, and the third action is reflected in the movement of an image (i.e., a change in the image) displayed on the screen of the smartphone or tablet computer. The displayed image, i.e., the object, may be a moving body, specifically an image imitating various autonomous mobile robots.

[0087] Furthermore, not all of the configurations and operations described in each embodiment are necessarily essential to the configurations and operations of the present disclosure. For example, among the components in each embodiment, any component that is not recited in an independent claim that represents the highest concept of the present disclosure is understood to be an optional component.

[0088] Terms used throughout this specification and the appended claims should be interpreted as "open-ended" terms. For example, the terms "including" or "including" should be interpreted as "limited to the aspects described as including," and the term "having" should be interpreted as "limited to the aspects described as having."

[0089] The terms used in this specification are used merely for the convenience of description and are not intended to limit the configuration, operation, etc. For example, terms such as "right," "left," "upper," and "lower" merely indicate directions in the drawings to which reference should be made. Furthermore, terms such as "inner" and "outer" indicate directions toward and away from the center of a focused element, respectively. The same applies to terms similar to or having the same meaning as these terms.

[0090] The technology of the present disclosure may have the following configuration. According to the technology of the present disclosure having the following configuration, a new motion can be created by mixing a plurality of predetermined motions, and this new motion can be reflected in the motion actually performed by the moving object, thereby reducing the effort and time required to create a new motion and saving costs. Furthermore, it is possible to encourage the user to create a new motion that is not limited to their own ideas. The effects achieved by the technology of the present disclosure are not necessarily limited to these, and may be any of the effects described in this specification. (1) A moving body comprising: a mixing parameter setting unit that sets mixing parameters; an action creation unit configured to be able to create a third action by mixing a first action that is predetermined as a moving body action and a second action that is predetermined as a moving body action different from the first action with a degree of reflection according to the mixing parameters set by the mixing parameter setting unit; and an action control unit configured to be able to reflect the third action created by the action creation unit in the action actually performed by the moving body. (2) The mobile body according to (1) above, further comprising a media presentation unit configured to be able to present predetermined media perceptible from outside the mobile body in association with the third action. (3) The predetermined media is a moving body as described in (2) above, which is a visual media. (4) The predetermined media is an auditory media, and the moving body is the above-mentioned (3). (5) A moving body of any one of (2) to (4) above, having media set in association with each of the first and second actions, wherein the specified media is media set in association with the action of the first and second actions that has a greater degree of reflection in the third action. (6) A moving body that is any one of (2) to (4) above, having media set in association with each of the first and second actions, and the specified media is media set in association with the action, of the first and second actions, that has a higher final degree of reflection in the third action. (7) a designation prompting unit configured to prompt a user to designate the mixing parameters; The mobile body is any one of (1) to (6) above, further comprising a designation recognition unit configured to be able to recognize the mixed parameters designated by the user as designated parameters, and the mixed parameter setting unit sets the designated parameters recognized by the designation recognition unit as the mixed parameters. (8) The mobile body of (7) above, wherein the designation promoting unit presents candidate mixed parameters for designation by the user as suggested parameters to promote designation of the mixed parameters. (9) The mobile body of (8) above, wherein the designation promoting unit displays an action created in accordance with the candidate mixed parameters to present the suggested parameters. (10) The mobile body according to (8) or (9) above, wherein the designation promoting unit presents, as the proposed parameters, a plurality of mixed parameters that are candidates for the designation. (11) The designation promotion unit is any one of the moving bodies described in (8) to (10) above, which limits the mixed parameters related to the operation of a moving part of the moving body when the load on the moving part of the moving body exceeds an allowable range in an operation created according to the mixed parameters that are candidates for designation. (12) A mobile body according to any one of (8) to (11) above, further comprising an operation mode selection unit configured to allow the user to select an operation mode of the mobile body, and the designation promotion unit changes the proposed parameters according to the operation mode selected by the user. (13) A moving body according to any one of (8) to (12) above, further comprising a learning processing unit configured to perform machine learning to determine training data regarding the mixture parameters for the second action to be mixed with the first action, and to create a learned model of the mixture parameters, and the designation promotion unit presents the mixture parameters calculated by the learned model based on the first and second actions as the proposed parameters. (14) The mobile body of (13) above, wherein the designation promotion unit presents a plurality of mixture parameters to be designated as the proposed parameters and prompts the user to designate one or some of the mixture parameters, and the learning processing unit performs the machine learning using the designated one or some of the mixture parameters as training data. (15) The learning processing unit is a mobile body as described in (14) above, which obtains the ranking of each mixed parameter when some of the mixed parameters are commanded, and performs the machine learning based on the ranking. (16) The learning processing unit is a mobile body of (13) above, which presents predetermined mixture parameters in creating the trained model and is configured so that the presented mixture parameters can be adjusted by the user, and performs the machine learning using the mixture parameters adjusted by the user as the training data. (17) A method for controlling a moving body, which prompts a user to specify a mixing parameter, mixes a first action, which is predetermined as a moving body action, with a second action, which is predetermined as a moving body action different from the first action, with a degree of reflection according to the mixing parameter specified by the user, to create a third action, and reflects the third action in the action actually performed by the moving body. (18) An information processing device comprising: a command generation unit configured to output a command to an interface device to cause the interface device to perform an action prompting a user to specify a mixing parameter; a mixing parameter input unit configured to input the mixing parameter specified by the user; an action creation unit configured to create a third action by mixing a first action predetermined as an action of an object with a second action predetermined as an action of the object that is different from the first action, with a degree of reflection according to the mixing parameter input to the mixing parameter input unit; and an action control unit configured to reflect the third action created by the action creation unit in an action actually performed by the object. (19) The information processing device according to (18), wherein the target object is a moving object. (20) The information processing device according to (18) above, wherein the object has a display unit, and the movement of the object is accompanied by the movement of an image displayed on the display unit.

[0091] This application claims priority based on Japanese Patent Application No. 2020-152309, filed on September 10, 2020, in the Japan Patent Office, the entire contents of which are incorporated herein by reference.

[0092] Those skilled in the art will recognize that various modifications, combinations, subcombinations, and variations may occur depending on design requirements and other factors, and are intended to be within the scope of the appended claims and their equivalents.

Claims

1. a mixing parameter setting unit that sets mixing parameters; a prompting unit configured to prompt a user to specify the mixing parameters; a motion creation unit configured to be able to create a third motion by mixing a first motion, which is predetermined as a motion of a moving object, with a second motion, which is predetermined as a motion of the moving object and is different from the first motion, with a reflection degree according to the mixing parameter set by the mixing parameter setting unit; a movement control unit configured to be able to reflect the third movement created by the movement creation unit in a movement actually performed by the moving object; Equipped with the designation facilitating unit restricts the mixed parameters related to the operation of the moving part when a load applied to the moving part of the moving object exceeds an allowable range in the operation created according to the mixed parameters that are candidates for designation; Mobile object.

2. a media presentation unit configured to present predetermined media perceivable from outside the moving object in association with the third action; The moving body according to claim 1 .

3. the predetermined media is a visual media; The moving body according to claim 2 .

4. the predetermined media is an auditory media; The moving body according to claim 2 .

5. media associated with each of the first and second actions; the predetermined media is media set in association with an action having a higher degree of reflection in the third action out of the first and second actions; The moving body according to claim 2 .

6. media associated with each of the first and second actions; the predetermined media is media set in association with an action having a high final degree of reflection in the third action, of the first and second actions; The moving body according to claim 2 .

7. A designation recognition unit configured to be able to recognize the mixing parameters designated by the user as designated parameters, the mixed parameter setting unit sets the designated parameter recognized by the designated parameter recognition unit as the mixed parameter. The moving body according to claim 1 .

8. the designation promoting unit presents, as suggested parameters, mixture parameters that are candidates for designation by the user, and promotes the designation of the mixture parameters; The moving body according to claim 7.

9. the designation facilitating unit displays an action created according to the candidate mixture parameters to be designated, and presents the proposed parameters; The moving body according to claim 8.

10. the designation promoting unit presents, as the proposed parameters, a plurality of mixed parameters that are candidates for the designation; The moving body according to claim 8.

11. an operation mode selection unit configured to allow the user to select an operation mode of the moving object; The designation promoting unit changes the proposed parameters in accordance with the operation mode selected by the user. The moving body according to claim 8.

12. a learning processing unit configured to perform machine learning to determine teacher data related to the mixing parameters for the second motion to be mixed with the first motion, and to create a learned model of the mixing parameters; The designation promotion unit presents, as the proposed parameters, mixture parameters calculated by the trained model based on the first and second actions. The moving body according to claim 8.

13. the designation promoting unit presents, as the proposed parameters, a plurality of mixture parameters that are candidates for the designation, and promotes designation of one or some of the plurality of mixture parameters; the learning processing unit performs the machine learning using the specified one or some of the mixture parameters as training data. The moving body according to claim 12.

14. the learning processing unit acquires a ranking of each of the mixture parameters when the part of the mixture parameters is commanded, and performs the machine learning based on the ranking. The moving body according to claim 13.

15. The learning processing unit presents predetermined mixture parameters in creating the trained model, and is configured to allow the user to adjust the presented mixture parameters, and performs the machine learning using the mixture parameters adjusted by the user as the training data. The moving body according to claim 12.

16. Prompts the user to specify mixing parameters, creating a third action by mixing a first action, which is predetermined as an action of a moving object, with a second action, which is predetermined as an action of the moving object and is different from the first action, with a reflection degree according to a mixing parameter designated by the user; reflecting the third motion on an actual motion of the moving object; limiting the mixing parameters related to the operation of the moving part when a load applied to the moving part of the moving object exceeds an allowable range in the operation created in accordance with the specified candidate mixing parameters; A method for controlling a moving object.

17. a command generating unit configured to be capable of outputting a command to the interface device to execute an operation of prompting a user to specify a mixing parameter; a prompting unit configured to prompt a user to specify the mixing parameters; a mixing parameter input unit configured to be able to input mixing parameters designated by the user; a movement creating unit configured to be able to create a third movement by mixing a first movement, which is predetermined as a movement of an object, with a second movement, which is predetermined as a movement of the object and is different from the first movement, with a reflection degree according to the mixing parameter input to the mixing parameter input unit; a movement control unit configured to be able to reflect the third movement created by the movement creation unit in an actual movement performed by the object; Equipped with the designation facilitating unit restricts the mixing parameters related to the operation of the moving part when a load applied to the moving part of the object exceeds an allowable range in the operation created according to the designation candidate mixing parameters; An information processing device comprising:

18. The object is a moving object. The information processing device according to claim 17.

19. the interface device has a display unit, The object and its movement are displayed on the display unit. The information processing device according to claim 17.

Images