Information processing device, measurement system, information processing method, and program

Abstract

The present invention provides an information processing device that can accurately determine whether the tilt of a container is caused by a specific operation and estimate the amount of material that would be discharged as a result of that specific operation. [Solution] The information processing device 200 according to the present disclosure is characterized by comprising: an operation estimation unit 3 that extracts feature quantities in an operation that caused the tilt of the container 1 from posture information showing the change in the tilt angle of the container 1 over time, and uses the feature quantities to determine whether or not the tilt of the container 1 is caused by a specific operation; and an outflow amount estimation unit 4 that, if the tilt of the container 1 is caused by a specific operation, acquires correspondence information representing the correspondence between the contents of an object held in the container 1 and the tilt angle, and uses the correspondence information and the tilt angle to estimate the outflow amount of an object that flows out of the container 1.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus.

Background Art

[0002] When taking out an object such as liquid and powder contained in a container by tilting the container, the tilt angle of the container from the vertical direction and the content volume of the object held in the container tilted at the tilt angle have a one-to-one correspondence. Therefore, by detecting the tilt angle of the container, the content volume of the object in the container can be estimated. Further, by detecting the tilt angle of the container at the start time and end time of the outflow of the object and taking the difference between the content volume of the container at the start time and the content volume of the container at the end time, the outflow volume of the object from the container can be estimated.

[0003] Also, if it is possible to determine whether the tilt of the container is caused by a specific operation, the outflow volume of the object caused by the specific operation can be estimated. For example, in the invention described in Patent Document 1, in a beverage container, when a tilt within a predetermined range is continuously detected for a predetermined time or more, it is determined that the holder of the container has drunk the liquid in the container. That is, in the invention described in Patent Document 1, it can be determined that the tilt of the container is caused by a specific operation of drinking the liquid in the container. Therefore, by estimating the outflow volume when the tilt of the container is caused by drinking the liquid in the container, the outflow volume can be estimated as the drinking volume and applied to the control of the drinking volume.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the invention described in Patent Document 1, the determination of whether the container owner has drunk the liquid inside the container is based on whether or not a tilt within a predetermined range is detected continuously for a predetermined time or longer. For example, if the container owner has discarded the liquid inside the container, but a tilt within a predetermined range is detected continuously for a predetermined time or longer, it will be incorrectly determined that the container owner has drunk the liquid inside the container. Thus, the invention described in Patent Document 1 has a problem with the accuracy of determining whether or not the tilt of the container is due to a specific operation.

[0006] This disclosure is made to solve the above-mentioned problems and aims to provide an information processing device that can accurately determine whether the tilt of a container is caused by a specific operation and estimate the amount of material that flows out due to that specific operation. [Means for solving the problem]

[0007] The information processing device according to this disclosure is characterized by comprising: an operation estimation unit that extracts feature quantities in an operation that caused the tilt of a container from posture information showing the change in the tilt angle of the container over time, and uses the feature quantities to determine whether or not the tilt of the container is caused by a specific operation; and, if the tilt of the container is caused by a specific operation, an outflow volume estimation unit that acquires correspondence information representing the correspondence between the volume of contents of an object held in the container and the tilt angle, and uses the correspondence information and the tilt angle to estimate the outflow volume of an object flowing out of the container.

[0008] Furthermore, the measurement system relating to this disclosure is characterized by comprising the above-mentioned information processing device and a posture sensor that measures the inclination angle of the container to acquire posture information and outputs the posture information to the operation estimation unit and the outflow rate estimation unit.

[0009] Furthermore, the information processing method relating to this disclosure is characterized by comprising the steps of: an operation estimation unit extracting feature quantities in an operation that caused the tilt of the container from posture information showing the change in the tilt angle of the container over time, and using the feature quantities to determine whether or not the tilt of the container is due to a specific operation; and an outflow volume estimation unit, if the tilt of the container is due to a specific operation, obtaining correspondence information representing the correspondence between the volume of contents of the object held in the container and the tilt angle, and using the correspondence information and the tilt angle to estimate the outflow volume of the object flowing out of the container.

[0010] Furthermore, the program relating to this disclosure is characterized in that an operation estimation unit causes a computer to perform the following steps: an operation estimation unit extracts feature quantities in the operation that caused the tilt of the container from posture information showing the change in the tilt angle of the container over time, and uses the feature quantities to determine whether or not the tilt of the container is due to a specific operation; and an outflow volume estimation unit, if the tilt of the container is due to a specific operation, obtains correspondence information representing the correspondence between the volume of contents of the object held in the container and the tilt angle, and uses the correspondence information and the tilt angle to estimate the outflow volume of the object flowing out of the container. [Effects of the Invention]

[0011] According to this disclosure, the information processing device, measurement system, information processing method, and program can accurately determine whether the tilt of a container is caused by a specific operation and estimate the amount of material that will flow out as a result of that specific operation. [Brief explanation of the drawing]

[0012] [Figure 1] This is a diagram showing the configuration of the measurement system according to Embodiment 1. [Figure 2] This is a schematic diagram showing an example of a container according to Embodiment 1. [Figure 3] This is a schematic diagram illustrating the inclination angle of the container in Embodiment 1. [Figure 4] This figure shows an example of the change over time in the inclination angle of the container with respect to time in Embodiment 1. [Figure 5] This is a diagram showing the configuration of a learning device related to the information processing device of Embodiment 1. [Figure 6] This is a diagram showing an example of the neural network of Embodiment 1. [Figure 7] This is a flowchart regarding the learning process of the learning device of Embodiment 1. [Figure 8] This is a configuration diagram of the operation estimation unit regarding the information processing device of Embodiment 1. [Figure 9] This is a diagram showing an example of the output information output by the output unit of Embodiment 1. [Figure 10] This is a block diagram showing an example of the hardware configuration of each component of Embodiment 1. [Figure 11] This is a flowchart showing the processing flow of the information processing device of Embodiment 1. [Figure 12] This is a schematic diagram showing an example of the container of Embodiment 2.

Embodiments for Carrying out the Invention

[0013] Hereinafter, the information processing device according to the embodiment will be described with reference to the drawings. The following embodiments are merely examples, and it is possible to appropriately combine the embodiments and appropriately modify each embodiment. In the figures, the same components are denoted by the same reference numerals.

[0014] Embodiment 1. The measurement system 100 and the information processing device 200 in Embodiment 1 will be described with reference to FIG. 1. FIG. 1 is a configuration diagram of the measurement system 100 of Embodiment 1. As shown in FIG. 1, the measurement system 100 includes a container 1, an attitude sensor 2, and an information processing device 200.

[0015] Container 1 holds objects such as liquids and powders. Container 1 is not particularly limited as long as the objects inside Container 1 can be taken out by tilting Container 1. Fig. 2 is a schematic diagram showing an example of Container 1 in Embodiment 1. In Embodiment 1, Container 1 is for the holder of Container 1 to drink the liquid inside Container 1, and will be described, for example, as a cup as shown in Fig. 2A, a glass as shown in Fig. 2B, etc. Also, in Embodiment 1, the object held in Container 1 will be described as a liquid beverage.

[0016] Fig. 3 is a schematic diagram explaining the tilt angle of Container 1 in Embodiment 1. As described above, Container 1 can take out the object inside Container 1 by tilting Container 1. Here, the tilt angle is represented by the angle by which Container 1 tilts from the vertical direction, as shown in Fig. 3. That is, the tilt angle is the angle formed by the plane perpendicular to the bottom surface of Container 1 and the vertical plane. Here, when the bottom surface of Container 1 is a plane, the bottom surface of Container 1 is a surface that coincides with the horizontal plane when Container 1 is placed stationary on a table, etc. Note that when the bottom surface of Container 1 can be approximated as a plane, the bottom surface of Container 1 is not limited to a plane and may be a curved surface or the like. In Fig. 3A, the tilt angle of Container 1 is the first tilt angle θ1, and in Fig. 3B, the tilt angle of Container 1 is the second tilt angle θ2. Since Container 1 in Fig. 3B tilts at a larger angle from the vertical direction than Container 1 in Fig. 3A, the second tilt angle θ2 is larger than the first tilt angle θ1.

[0017] Figure 4 shows an example of the change over time of the tilt angle of container 1 in Embodiment 1. In Figure 4, the vertical axis represents the tilt angle of container 1, and the horizontal axis represents time. The unit of the tilt angle of container 1 is radians (rad), and the unit of time is seconds (sec). The solid line in Figure 4 shows the change over time of the tilt angle of container 1. Peak A in Figure 4 is when container 1 is lifted and the holder drinks three sips of the liquid inside while holding container 1. Peak B in Figure 4 is when container 1 is lifted and the holder drinks all of the liquid inside container 1. The range C in Peak B shows a sharper rise in the tilt angle compared to the range D in Peak A, because the holder tilted container 1 forcefully in order to drink all of the liquid inside. A method for identifying the operations that caused the tilt of container 1 based on the peaks in Figure 4 will be described later.

[0018] The attitude sensor 2 acquires attitude information showing the change in the tilt angle of container 1 over time. That is, the attitude sensor 2 measures the tilt angle of container 1 and acquires attitude information of container 1 as shown in Figure 4 by representing it chronologically. The attitude sensor 2 is, for example, an accelerometer and a gyroscope, and may be a single sensor or a combination of multiple sensors. Alternatively, a camera may be used as the attitude sensor 2, in which case the attitude information may be acquired using optical flow, SLAM (Simultaneous Localization and Mapping), and environmental markers. If the attitude sensor 2 is fixed to container 1, such as an accelerometer and a gyroscope, and does not need to acquire information about the external environment of container 1, the restrictions on attachment to container 1 are reduced compared to an attitude sensor 2 that acquires information about the external environment of container 1, which is advantageous in terms of cost and maintenance. Also, although the attitude sensor 2 is attached to container 1 in Figure 2, the position where the attitude sensor 2 is attached is not limited to this, and the attitude sensor 2 may be attached to any position where the tilt angle of container 1 can be measured.

[0019] Furthermore, although the attitude sensor 2 is attached to the container 1 in Figure 2, it is not limited to this. That is, the attitude sensor 2 may not be directly attached to the container 1, but may acquire attitude information from outside the container 1. For example, the attitude sensor 2 may be configured such that an infrared light source is provided on the container 1, and tracking is performed by an infrared camera provided on the outside of the container 1. The attitude sensor 2 outputs the acquired attitude information of the container 1 to the operation estimation unit 3 and the outflow rate estimation unit 4 of the information processing device 200.

[0020] The information processing device 200 includes an operation estimation unit 3, an outflow volume estimation unit 4, a storage unit 5, and an output unit 6.

[0021] The operation estimation unit 3 acquires posture information from the posture sensor 2 that shows the change in the tilt angle of the container 1 over time. Then, the operation estimation unit 3 acquires operation information from the posture information of the tilt angle of the container 1 that shows the operation that caused the tilt of the container 1, and outputs it to the outflow rate estimation unit 4 and the output unit 6.

[0022] At this time, the operation estimation unit 3 extracts a feature quantity representing the change in the tilt angle over time in the operation that caused the tilt of container 1 from the orientation information of the tilt angle of container 1, and uses this feature quantity to obtain operation information indicating the operation that caused the tilt of container 1. In the following, as a specific example of this feature quantity, it will be explained assuming that this feature quantity is the amount of change in the tilt angle per unit time. The amount of change in the tilt angle per unit time is the ratio of the difference between the tilt angle at one point in time and the tilt angle at another point in time. Note that the feature quantity extracted by the operation estimation unit 3 from the orientation information of the tilt angle of container 1 is not limited to the amount of change in the tilt angle per unit time, but can be any waveform pattern characteristic of each operation that caused the tilt of container 1, as long as it can identify the operation.

[0023] The method for obtaining specific operation information by the operation estimation unit 3 will now be explained. For example, in Figure 4, range E at peak B represents the time when the user lifts container 1 to drink the liquid inside, and range F at peak B represents the time when the user is drinking the liquid inside container 1. When the user is drinking the liquid inside container 1, they need to swallow, so they need to tilt container 1 more slowly than when they lift container 1 and bring it to their mouth. Therefore, the rate of change of the tilt angle per unit time in range F is smaller than the rate of change of the tilt angle per unit time in range E. When the operation estimation unit 3 observes the above characteristics in the posture information, it can determine that the user has performed a drinking operation to drink the beverage.

[0024] Furthermore, for example, if the user tilts container 1 to discard the liquid inside, the rate of change in the tilt angle per unit time will differ from the rate of change in the tilt angle per unit time shown by peak B in Figure 4, because swallowing is not required. For example, when the user tilts container 1 to discard the liquid inside, it is characterized by tilting it sharply only once and maintaining that tilt angle, and this characteristic is thought to be reflected in the rate of change in the tilt angle per unit time. Also, for example, if the user performs a washing operation to wash container 1, the tilt angle of container 1 changes rapidly in a shorter time compared to when drinking the liquid inside container 1, and this characteristic is thought to be reflected in the rate of change in the tilt angle per unit time.

[0025] Therefore, the operation estimation unit 3 can obtain operation information indicating the operation that caused the tilt of container 1 using the amount of change in the tilt angle per unit time. If container 1 is a cup or glass as shown in Figure 2, the operation information could include, for example, a drinking operation such as drinking a beverage, a disposal operation such as discarding the liquid, and a washing operation such as washing container 1.

[0026] Next, the operation estimation unit 3 determines whether the operation that caused the tilt of container 1, which is the acquired operation information, matches a specific operation. That is, the operation estimation unit 3 determines from the posture information whether the tilt of container 1 is due to a specific operation. A specific operation is an operation that is pre-set to estimate the amount of liquid spilled. For example, the operation estimation unit 3 can set the specific operation as a drinking operation, and the amount of liquid spilled due to operations other than drinking operations, such as disposal operations, will not be estimated by the amount of liquid spilled by the owner of container 1, thereby estimating the amount of liquid consumed by the owner of container 1. When the operation estimation unit 3 outputs the operation information to the amount of liquid spilled by container 1, it includes information in the operation information indicating whether the operation that caused the tilt of container 1 matches a specific operation.

[0027] Furthermore, operation information and specific operations are information that indicates the attributes of the operations that caused the tilting of container 1. In other words, the operation estimation unit 3 may acquire as operation information the attributes common to each operation that caused the tilting of container 1. For example, if the operation that caused the tilting of container 1 was sipping or gulping, the operation estimation unit 3 acquires as operation information the attribute of drinking operation, which is common to both operations. Even in this case, the operation estimation unit 3 can estimate the amount of liquid consumed by the owner of container 1 by setting a specific operation as a drinking operation and not estimating the amount of liquid spilled due to operations other than drinking operations, such as disposal operations, by not having the spillage amount estimation unit 4 estimate it.

[0028] Furthermore, the operation estimation unit 3 obtains information from the posture information indicating the tilt angle of the container 1 at the time a specific operation is started and at the time a specific operation is completed, and includes this information in the operation information.

[0029] The operation estimation unit 3 is equipped with a function to acquire operation information indicating the operation that caused the tilt of container 1, using machine learning or a computational algorithm. First, we will explain the case in which the function of the operation estimation unit 3 is implemented by machine learning. Figure 5 is a configuration diagram of the learning device 10 relating to the information processing device 200 of Embodiment 1. The learning device 10 comprises a data acquisition unit 10a, a model generation unit 10b, and a trained model storage unit 10c.

[0030] The data acquisition unit 10a of the learning device 10 acquires posture information and correct operation information corresponding to the posture information as training data.

[0031] The model generation unit 10b learns operation information indicating the operation that caused the tilt of container 1, based on training data created from the combination of posture information output from data acquisition unit 10a and correct operation information corresponding to the posture information. In other words, the model generation unit 10b generates a learning model that infers the optimal operation information from the posture information acquired by posture sensor 2 and correct operation information corresponding to the posture information. Here, the training data is data in which the correct operation information of posture information and the correct operation information corresponding to the posture information are related to each other.

[0032] The learning device 10 and the operation estimation unit 3 are used to learn operation information output by the information processing device 200, but they may be connected to the information processing device 200 via a network and be separate devices from the information processing device 200. Furthermore, the learning device 10 and the operation estimation unit 3 may be built into the information processing device 200. Additionally, the learning device 10 and the operation estimation unit 3 may reside on a cloud server.

[0033] The learning algorithm used by the model generation unit 10b can be any known algorithm, such as supervised learning or unsupervised learning. As an example, the case in which a neural network is applied will be described.

[0034] The model generation unit 10b learns operation information, for example, by supervised learning following a neural network. Here, supervised learning is a method in which pairs of input and result (label) data are provided to the learning device 10, thereby learning features in the training data and inferring the result from the input.

[0035] A neural network consists of an input layer made up of multiple neurons, an intermediate layer (hidden layer) made up of multiple neurons, and an output layer made up of multiple neurons. The intermediate layer can be one or more layers.

[0036] Figure 6 shows an example of a neural network according to Embodiment 1. For example, in a three-layer neural network like the one shown in Figure 6, when multiple inputs are input to the input layer (X1-X3), these values ​​are multiplied by weights W1 (W11-W16) and input to the hidden layer (Y1-Y2), and the result is further multiplied by weights W2 (W21-W26) and output from the output layer (Z1-Z3). This output result varies depending on the values ​​of weights W1 and W2.

[0037] In this application, the neural network learns operation information indicating the operation that caused the tilt of the container 1 through so-called supervised learning, according to training data created based on the combination of posture information and operation information corresponding to the posture information acquired by the data acquisition unit 10a.

[0038] In other words, a neural network learns by inputting pose information into the input layer and adjusting the weights W1 and W2 so that the output from the output layer approaches the correct operation information.

[0039] The model generation unit 10b generates and outputs a trained model by performing the training described above.

[0040] The trained model storage unit 10c stores the trained model output from the model generation unit 10b.

[0041] Next, we will explain the learning process performed by the learning device 10 using Figure 7. Figure 7 is a flowchart of the learning process of the learning device 10 in Embodiment 1.

[0042] In step S10, the data acquisition unit 10a of the learning device 10 acquires correct data for posture information and operation information corresponding to the posture information. Although the correct data for posture information and operation information corresponding to the posture information are acquired simultaneously, it is sufficient if the correct data for posture information and operation information corresponding to the posture information are input in association, and the correct data for posture information and operation information corresponding to the posture information may be acquired at different times.

[0043] In step S11, the model generation unit 10b learns operation information indicating the operation that caused the tilt of the container 1 through so-called supervised learning, according to training data created based on the combination of posture information acquired by the data acquisition unit 10a of the learning device 10 and the correct data of operation information corresponding to the posture information, and generates a trained model.

[0044] In step S12, the trained model storage unit 10c stores the trained model generated by the model generation unit 10b.

[0045] Figure 8 is a diagram showing the configuration of the operation estimation unit 3 of the information processing device 200 of Embodiment 1. The operation estimation unit 3 comprises a data acquisition unit 3a and an inference unit 3b.

[0046] The data acquisition unit 3a of the operation estimation unit 3 acquires attitude information from the attitude sensor 2.

[0047] The inference unit 3b infers operation information obtained using a trained model. That is, by inputting the posture information acquired by the data acquisition unit 3a into this trained model, it is possible to output operation information inferred from the posture information.

[0048] In this embodiment, operation information is output using a trained model learned by the model generation unit 10b of the information processing device 200. However, a trained model may be obtained from an external source, such as another information processing device 200, and operation information may be output based on this trained model.

[0049] Furthermore, the model generation unit 10b may learn operation information according to training data created for multiple information processing devices 200. The model generation unit 10b may also acquire training data from multiple information processing devices 200 used in the same area, or it may learn operation information using training data collected from multiple information processing devices 200 operating independently in different areas. It is also possible to add or remove information processing devices 200 from the target midway through the process. Additionally, a learning device 10 that has learned operation information for one information processing device 200 may be applied to another information processing device 200, and the learning device 10 may relearn and update the operation information for that other information processing device 200.

[0050] Furthermore, the learning algorithm used in the model generation unit 10b may be deep learning, which learns to extract the features themselves, or machine learning may be performed according to other known methods, such as genetic programming, inductive logic programming, and support vector machines.

[0051] Furthermore, the functions of the operation estimation unit 3 may be implemented by a rule-based calculation algorithm as described above. In this case, an index is designed that can explain the characteristics of the posture information when each operation is performed. For example, in the case of a drinking operation, the amount of change per unit time in the posture information of the tilt angle of container 1 differs during the operation from lifting container 1 to bringing it close to the mouth, the operation until the liquid surface touches the mouth, the operation of the person holding container 1 drinking the liquid and pouring it into their mouth, the operation of moving the liquid surface away from the mouth, and the operation of returning container 1, and each action exhibits its own characteristics. These characteristics are then classified into operations according to the amount of change per unit time in the posture information of the tilt angle of container 1. The operation estimation unit 3 identifies the operation corresponding to the input amount of change per unit time in the posture information of the tilt angle of container 1 from among the classified operations and outputs it as operation information.

[0052] The outflow rate estimation unit 4 estimates the amount of liquid that has flowed out of container 1 based on the tilt of container 1, using the attitude information of container 1 acquired from the attitude sensor 2, and outputs the estimated outflow rate to the output unit 6 as outflow rate information. As described above, there is a one-to-one correspondence between the tilt angle of container 1 from the vertical and the volume of the object held in container 1 when tilted at that angle. For example, suppose θ1 in Figure 3A is the tilt angle of container 1 at the start of a particular operation, and θ2 in Figure 3B is the tilt angle of container 1 at the end of a particular operation. The outflow rate estimation unit 4 can estimate that the volume of liquid held in container 1 when the tilt angle is θ1 is uniquely determined in Figure 3A. Similarly, the outflow rate estimation unit 4 can estimate that the volume of liquid held in container 1 when the tilt angle is θ2 in Figure 3B is uniquely determined. Therefore, the outflow rate estimation unit 4 can estimate the amount of liquid outflow when the inclination angle of container 1 is changed from θ1 to θ2 by calculating the difference between the amount of liquid held in container 1 when the inclination angle is θ1 and the amount of liquid held in container 1 when the inclination angle is θ2, as shown in Figure 3.

[0053] The memory unit 5 stores correspondence information representing the correspondence between the volume of the object held in the container 1 and the tilt angle of the container 1. The outflow volume estimation unit 4 then retrieves this correspondence information from the memory unit 5 and uses the tilt angle included in this correspondence information and the attitude information obtained from the attitude sensor 2 to estimate the outflow volume of the liquid object that has flowed out of the container 1. In Figure 1, the memory unit 5 is located within the information processing device 200, but is not limited to this. For example, the memory unit 5 may be connected to the information processing device 200 via a network, or it may be a separate device from the information processing device 200.

[0054] The function of the outflow volume estimation unit 4 is implemented by a calculation algorithm or machine learning. When the function of the outflow volume estimation unit 4 is implemented by a calculation algorithm, the storage unit 5 stores a function as corresponding information for calculating the amount of contents that can be held in container 1 when it is tilted at a given angle, based on the tilt angle of container 1. Since the values ​​of the tilt angle and the amount of contents that can be held change depending on the shape of container 1, this function is determined analytically in advance. Alternatively, this function may be obtained experimentally in advance from a table that associates tilt angle with content volume. The outflow volume calculation unit obtains this function as corresponding information from the storage unit 5, inputs the tilt angle in the posture information into the function, and obtains the amount of contents that can be held in container 1 when it is tilted at that angle as output. As a result, the outflow volume estimation unit 4 can estimate the amount of liquid that has flowed out of container 1 due to the tilt of container 1 by calculating the difference between the content volume at the start of a particular operation and the content volume at the end of a particular operation. Furthermore, the function may also take into account the error in the volume of contents corresponding to the tilt angle of container 1 and output a range of possible values ​​for the volume of contents corresponding to that tilt angle. Even in this case, the outflow rate estimation unit 4 can calculate the range of possible values ​​for the outflow rate of the liquid, and therefore can estimate the outflow rate of the liquid, which is the substance that has flowed out of container 1 due to the tilt of container 1.

[0055] Furthermore, if the function of the outflow volume estimation unit 4 is implemented using machine learning, a trained model is generated that takes the tilt angle in the posture information as input and outputs the amount of contents held by the container 1 tilted at that angle, and this model is stored in the storage unit 5 as corresponding information. The outflow volume estimation unit 4 obtains this model as corresponding information from the storage unit 5, takes the tilt angle in the posture information as input, and obtains the amount of contents held by the container 1 tilted at that angle as output, thereby estimating the amount of liquid, which is the object that has flowed out of the container 1 due to the tilt of the container 1.

[0056] The discharge volume estimation unit 4 estimates the discharge volume based on the posture information as described above when the operation that caused the tilt of container 1 matches a predetermined specific operation in the operation information acquired from the operation estimation unit 3. That is, if the tilt of container 1 is caused by a specific operation, the discharge volume estimation unit 4 estimates the amount of liquid discharged from container 1 using the tilt angle, but if the tilt of container 1 is not caused by a specific operation, it does not estimate the amount of liquid discharged from container 1 using the tilt angle. As a result, as described above, for example, the discharge volume estimation unit 4 can estimate the amount of liquid consumed by the owner of container 1 by estimating the discharge volume during a drinking operation, which is set as a specific operation, and not estimating the amount of discharge due to operations other than drinking operations, such as disposal operations.

[0057] The output unit 6 outputs output information which includes at least one of the operation information obtained from the operation estimation unit 3 and the outflow amount information obtained from the outflow amount estimation unit 4, and presents the output information to the person holding the container 1. The output method of the output unit 6 is not particularly limited. For example, the output unit 6 may output visual information using a lamp and a display. Alternatively, for example, the output unit 6 may output auditory information using a buzzer and a speaker. Alternatively, for example, the output unit 6 may output tactile information using a vibration motor and electrical stimulation. Furthermore, the output unit 6 may be equipped with communication means and configured to output output information to external information devices such as servers and mobile terminals, or to acquire information from external information devices. In this case, the information processing device 200 can be integrated with a management system in which multiple information devices cooperate.

[0058] A specific example of how the output unit 6 presents output information will be explained. For example, suppose the owner of container 1 is an athlete who needs to strictly manage their physical condition, and a drinking operation is set as a specific operation, and the outflow rate estimation unit 4 estimates the outflow rate as the amount to be consumed at the time of the drinking operation. In this case, the output unit 6 presents the outflow rate as the amount to be consumed at the time of the drinking operation to the athlete, enabling the athlete to understand the amount to be consumed. The output unit 6 may also present information as output information indicating whether the amount to be consumed is sufficient or excessive compared to a pre-set appropriate amount to be consumed. The appropriate amount to be consumed may be calculated from the athlete's heart rate, weight, and exercise amount indicated by calories. The output unit 6 may also compare the actual amount to be consumed with a pre-set appropriate amount to be consumed, and if the actual amount to be consumed is insufficient, it may suggest performing a drinking operation, and may include information in the output information indicating the timing of performing a drinking operation.

[0059] Figure 9 shows an example of output information output by the output unit 6 of Embodiment 1. In Figure 9, the output information is displayed on the display unit of a mobile terminal or the like. As shown in Figure 9A, the output unit 6 presents the user with the amount of liquid dispensed during the drinking operation as "this time's amount of liquid," the elapsed time between this drinking operation and the previous drinking operation as "interval," and the total amount of liquid dispensed on the day the drinking operation was performed as "today's amount of liquid." Alternatively, as shown in Figure 9B, the output unit 6 may calculate the amount of liquid dispensed for each day of the week for the most recent week and present it as output information. In this case, the output unit 6 presents the user with the total amount of liquid dispensed on the day the output information is presented as "today's amount of liquid," and the difference between the pre-set appropriate amount of liquid dispensed and "today's amount of liquid" as "shortfall to target." This allows the athlete or other user who possesses container 1 to understand and appropriately control their liquid consumption.

[0060] Furthermore, suppose the owner of container 1 is a sick person, an infant, or an elderly person, and there is a guardian who manages the owner's health in addition to the owner. Also, suppose a drinking operation is set as a specific operation, and the outflow rate estimation unit 4 estimates the outflow rate as the amount consumed at the time of the drinking operation. In this case, the output unit 6 presents the outflow rate as the amount consumed at the time of the drinking operation to the guardian, so that the guardian can understand the remaining amount of beverage in container 1. This allows the guardian to efficiently replenish the beverage in container 1. The output unit 6 may also present operation information as output information to the guardian, indicating the operation that caused the tilting of container 1. This allows the guardian to understand, for example, whether the owner of container 1 is performing a drinking operation or a disposal operation, and to manage the owner's health by supporting appropriate drinking operations.

[0061] Furthermore, for example, the output unit 6 may include information indicating the elapsed time since the last operation estimation unit 3 obtained confirmation that a cleaning operation had been performed, and present this information to the user. This allows the user of container 1 to know the elapsed time since the last cleaning operation of container 1 and to determine whether container 1 is being kept clean.

[0062] Next, we will describe hardware configuration examples for each component in Embodiment 1, namely the operation estimation unit 3, the outflow volume estimation unit 4, the storage unit 5, the output unit 6, the data acquisition unit 10a of the learning device 10, the model generation unit 10b, the trained model storage unit 10c, the data acquisition unit 3a of the operation estimation unit 3, and the inference unit 3b. Figure 10 is a block diagram showing hardware configuration examples for each component in Embodiment 1. Each component in Embodiment 1 may be a dedicated hardware processing circuit 90 as shown in Figure 10A, or a processor 92 that executes a program stored in memory 94 as shown in Figure 10B.

[0063] As shown in Figure 10A, when each configuration in Embodiment 1 is dedicated hardware, the processing circuit 90 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-programmable Gate Array), or a combination thereof. Each function of each configuration in Embodiment 1 may be realized by the processing circuit 90, or the functions of each part may be combined and realized by a single processing circuit 90.

[0064] As shown in Figure 10B, when each configuration in Embodiment 1 is a processor 92, the functions of each part are realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in memory 94. The processor 92 realizes each function of each configuration in Embodiment 1 by reading and executing the program stored in memory 94. In other words, each configuration in Embodiment 1 is equipped with memory 94 for storing a program that, when executed by the processor 92, will result in the execution of each step shown in Figure 7 and Figure 11, which will be described later. These programs can also be said to cause the computer to execute the procedures or methods of each configuration in Embodiment 1.

[0065] Here, processor 92 refers to, for example, a CPU (Central Processing Unit), processing unit, arithmetic unit, processor, microprocessor, microcomputer, or DSP (Digital Signal Processor). Memory 94 may be, for example, a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable ROM), or EEPROM (Electrically EPROM), or a magnetic disk such as a hard disk or flexible disk, or an optical disk such as a MiniDisc, CD (Compact Disc), or DVD (Digital Versatile Disc).

[0066] Furthermore, for each function of each configuration in Embodiment 1, some may be implemented with dedicated hardware, and some may be implemented with software or firmware. In this way, the processing circuit 90 in Embodiment 1 can implement the above-mentioned functions by hardware, software, firmware, or a combination thereof.

[0067] Next, the processing flow of the information processing device 200 in Embodiment 1 will be described. Figure 11 is a flowchart of the processing flow of the information processing device 200 in Embodiment 1. In the following, we will describe the case in which the function of the operation estimation unit 3 is implemented by machine learning as shown in Figure 8, and will omit the explanation of the case in which machine learning is not used. Even if the function of the operation estimation unit 3 is not implemented by machine learning, the operation estimation unit 3 takes the posture information acquired from the posture sensor 2 as input, acquires operation information indicating the operation that caused the tilt of the container 1, includes information in the operation information indicating whether or not the operation that caused the tilt of the container 1 matches a specific operation, and outputs the operation information to the outflow rate estimation unit 4 and the output unit 6.

[0068] In step S101, the data acquisition unit 3a of the operation estimation unit 3 acquires attitude information from the attitude sensor 2.

[0069] In step S102, the operation estimation unit 3 extracts feature quantities from the posture information for the operations that caused the tilt of container 1, and uses these feature quantities to obtain operation information indicating the operations that caused the tilt of container 1. For example, the operation estimation unit 3 obtains operation information indicating the operations that caused the tilt of container 1 using the amount of change in the tilt angle per unit time from the posture information. The inference unit 3b of the operation estimation unit 3 obtains a trained model from the trained model storage unit 10c, inputs the posture information into the obtained trained model, and obtains operation information.

[0070] In step S103, the inference unit 3b of the operation estimation unit 3 outputs the operation information obtained by the trained model to the outflow volume estimation unit 4 and the output unit 6. The operation estimation unit 3 also determines whether the tilt of container 1 is due to a specific operation. That is, when the inference unit 3b of the operation estimation unit 3 outputs the operation information to the outflow volume estimation unit 4, it includes information in the operation information indicating whether the operation that caused the tilt of container 1 matches a specific operation.

[0071] In step S104, the outflow rate estimation unit 4 determines whether the operation information obtained from the operation estimation unit 3 includes information indicating that the operation that caused the tilt of container 1 matches a specific operation that has been set in advance.

[0072] In step S105, if the operation information obtained from the operation estimation unit 3 contains information indicating that the operation that caused the tilt of container 1 matches a specific pre-set operation (step S104: YES), the outflow volume estimation unit 4 obtains correspondence information from the storage unit 5 that represents the correspondence between the volume of the contents of the object held in container 1 and the tilt angle of container 1. If the operation information obtained from the operation estimation unit 3 does not contain information indicating that the operation that caused the tilt of container 1 matches a specific pre-set operation (step S104: NO), the outflow volume estimation unit 4 proceeds to step S107.

[0073] In step S106, the outflow rate estimation unit 4 uses the tilt angle included in the attitude information of the container 1 obtained from the attitude sensor 2 and the corresponding information obtained from the storage unit 5 to estimate the amount of liquid that has flowed out of the container 1 due to the tilt of the container 1, and outputs the estimated outflow rate information to the output unit 6. In other words, if the tilt of the container 1 is caused by a specific operation, the outflow rate estimation unit 4 uses the tilt angle to estimate the amount of liquid that has flowed out of the container 1, but if the tilt of the container 1 is not caused by a specific operation, it does not use the tilt angle to estimate the amount of liquid that has flowed out of the container 1.

[0074] In step S107, the output unit 6 outputs output information which includes at least one of the operation information obtained from the operation estimation unit 3 and the outflow amount information obtained from the outflow amount estimation unit 4, and presents the output information to the person holding the container 1. This concludes the explanation of the processing flow of the information processing device 200 in Embodiment 1.

[0075] As described above, the information processing device 200 in Embodiment 1 includes an operation estimation unit 3 that extracts feature quantities in the operation that caused the tilt of the container 1 from posture information showing the change in the tilt angle of the container 1 over time, and uses these feature quantities to determine whether or not the tilt of the container 1 is due to a specific operation, and an outflow amount estimation unit 4 that, if the tilt of the container 1 is due to a specific operation, acquires correspondence information representing the correspondence between the volume of the contents of the object held in the container 1 and the tilt angle, and uses the correspondence information and the tilt angle to estimate the outflow amount of the object flowing out of the container 1. With the above configuration, the information processing device 200 in Embodiment 1 can distinguish features in the posture information for each operation that caused the tilt of the container 1, accurately determine whether or not the tilt of the container 1 is due to a specific operation, and estimate the outflow amount of the object due to that specific operation.

[0076] Furthermore, in Embodiment 1, the feature quantity used by the operation estimation unit 3 when acquiring operation information is the amount of change in the tilt angle per unit time in the posture information that shows the change in the tilt angle of the container 1 over time. With the above configuration, the information processing device 200 in Embodiment 1 can distinguish the features in the posture information for each operation that caused the tilt of the container 1, accurately determine whether or not the tilt of the container 1 is caused by a particular operation, and estimate the amount of material that will flow out due to that particular operation.

[0077] Furthermore, the operation estimation unit 3 in Embodiment 1 includes a data acquisition unit 3a that acquires posture information, and an inference unit 3b that outputs operation information indicating the operation that caused the tilt of the container 1 from the posture information acquired by the data acquisition unit 3a. With the above configuration, the information processing device 200 in Embodiment 1 can distinguish the characteristics of the posture information for each operation that caused the tilt of the container 1 based on machine learning or a computational algorithm, accurately determine whether the tilt of the container 1 is caused by a specific operation, and estimate the amount of material that will flow out due to that specific operation.

[0078] Furthermore, the output unit 6 in Embodiment 1 presents output information that includes at least one of the following: operation information indicating the operation that caused the tilt of the container 1, and discharge amount information indicating the amount of liquid discharged. With the above configuration, the information processing device 200 in Embodiment 1 can, for example, present to the owner or guardian of the container 1 information such as what operation was performed on the container 1, or the amount of liquid discharged in a specific operation. This allows the owner or guardian of the container 1 to appropriately manage the amount of liquid consumed.

[0079] Furthermore, in Embodiment 1, the object held in container 1 is a beverage, and the specific operation for which the outflow rate estimation unit 4 estimates the outflow rate is the drinking operation of the beverage. With the above configuration, the information processing device 200 in Embodiment 1 can estimate the amount of beverage consumed by the holder of container 1 by estimating the outflow rate only during the drinking operation performed by the holder of container 1, among the operations that caused the tilt of container 1.

[0080] Furthermore, in Embodiment 1, the operation estimation unit 3 extracts feature quantities for the operations that caused the tilt of container 1 from the posture information showing the change in the tilt angle of container 1 over time, obtains operation information indicating the operations that caused the tilt of container 1 using these feature quantities, and determines whether the tilt of container 1 is caused by a specific operation by determining whether the operations that caused the tilt of container 1 match a specific operation that has been set in advance. With the above configuration, the information processing device 200 in Embodiment 1 can distinguish the features in the posture information for each operation that caused the tilt of container 1, accurately determine whether the tilt of container 1 is caused by a specific operation, and estimate the amount of material that will flow out due to that specific operation.

[0081] Furthermore, the measurement system of Embodiment 1 includes an information processing device 200 and a posture sensor 2 that measures the tilt angle of the container 1 to acquire posture information and outputs the posture information to the operation estimation unit 3 and the outflow amount estimation unit 4. With the above configuration, the measurement system of Embodiment 1 can distinguish the characteristics of the posture information for each operation that caused the tilt of the container 1, accurately determine whether the tilt of the container 1 is caused by a particular operation, and estimate the amount of material outflow caused by that particular operation.

[0082] Furthermore, the information processing method of Embodiment 1 includes the steps of: an operation estimation unit 3 extracting feature quantities in the operation that caused the tilt of container 1 from posture information showing the change in the tilt angle of container 1 over time, and using these feature quantities to determine whether or not the tilt of container 1 is due to a specific operation; and an outflow amount estimation unit 4, if the tilt of container 1 is due to a specific operation, acquiring correspondence information representing the correspondence between the volume of the contents of the object held in container 1 and the tilt angle, and using the correspondence information and the tilt angle to estimate the outflow amount of the object flowing out of container 1. Thus, the information processing method of Embodiment 1 can distinguish features in the posture information for each operation that caused the tilt of container 1, accurately determine whether or not the tilt of container 1 is due to a specific operation, and estimate the outflow amount of the object due to that specific operation.

[0083] Furthermore, the program of Embodiment 1 causes the computer to perform the following steps: the operation estimation unit 3 extracts characteristic quantities from the posture information showing the change in the tilt angle of the container 1 over time, and uses these characteristic quantities to determine whether the tilt of the container 1 is due to a specific operation; and the outflow amount estimation unit 4, if the tilt of the container 1 is due to a specific operation, obtains correspondence information representing the correspondence between the volume of the contents of the object held in the container 1 and the tilt angle, and uses the correspondence information and the tilt angle to estimate the outflow amount of the object flowing out of the container 1. As a result, the program of Embodiment 1 can distinguish the characteristics in the posture information for each operation that caused the tilt of the container 1, accurately determine whether the tilt of the container 1 is due to a specific operation, and estimate the outflow amount of the object due to that specific operation.

[0084] It should be noted that the container 1 in Embodiment 1 is designed so that an object can be removed from inside the container 1 by tilting the container 1. Therefore, it does not include containers that use pressure, such as baby bottles, straws, and pumps, to remove objects, or containers that use surface tension, such as space glass, to remove objects.

[0085] Furthermore, the operation estimation unit 3 and the outflow rate estimation unit 4 may be a single configuration that incorporates both functions. In this case, the single configuration takes posture information acquired from the posture sensor 2 as input and outputs operation information indicating the operation that caused the tilt of the container 1, and outflow rate information indicating the outflow rate at a specific operation, to the output unit 6.

[0086] Although the description of Embodiment 1 assumed that the owner of container 1 was a human, the invention is not limited to this, as long as the container 1 is tilted to remove the object inside, the target may be a living organism other than a human.

[0087] Furthermore, the outflow rate estimation unit 4 does not estimate the amount of material flowing out of container 1 using the tilt angle if the tilt of container 1 is not caused by a specific operation. However, it is not limited to this as long as it can estimate the amount of material flowing out when the tilt of container 1 is caused by a specific operation. For example, the outflow rate estimation unit 4 may estimate separately the amount of material flowing out when the tilt of container 1 is not caused by a specific operation and the amount of material flowing out when the tilt of container 1 is caused by a specific operation. Even in this case, the outflow rate estimation unit 4 can estimate the amount of material flowing out when a specific operation is caused. For example, by estimating the amount of material flowing out during a drinking operation set as a specific operation, it is possible to estimate the amount of liquid consumed by the person holding container 1.

[0088] Embodiment 2. The measurement system 101 and information processing device 200 in Embodiment 2 will now be described. In Embodiment 1, the container 1 is for the holder of the container 1 to drink the liquid inside, and the object held in the container 1 is a beverage. The container 1 in the measurement system 101 of Embodiment 2 differs from the measurement system 100 of Embodiment 1 in that the container 1 is not for the holder of the container 1 to drink the liquid inside, and the object held in the container 1 is not a beverage. Components similar to those in Embodiment 1 are denoted by the same reference numerals. Detailed explanations of components similar to those in Embodiment 1 will be omitted, and the explanation will mainly focus on components that differ from those in Embodiment 1. In Embodiment 2 as well, the feature quantities extracted by the operation estimation unit 3 from the orientation information of the tilt angle of the container 1 are not limited to the amount of change in the tilt angle per unit time, but are waveform patterns characteristic of each operation that caused the tilt of the container 1, as long as they can identify the operation.

[0089] As described above, the container 1 in the measurement system 101 of Embodiment 2 is not for the purpose of the person holding the container 1 to drink the liquid inside, and the object held in the container 1 is also not a beverage. Figure 12 is a schematic diagram showing an example of the container 1 of Embodiment 2. In Embodiment 2, the container 1 is described as being for holding seasonings, such as a soy sauce dispenser as shown in Figure 12A and a salt shaker as shown in Figure 12B. Also, in Embodiment 2, the object held in the container 1 is described as a seasoning, and the seasoning is not limited to a liquid such as soy sauce, but may be a powder such as salt.

[0090] The method for obtaining specific operation information by the operation estimation unit 3 will now be explained. For example, when the owner of container 1 performs an operation to use seasonings in cooking, etc., there is a characteristic that the container 1 is tilted slowly in order to prevent using too much seasoning, and it is thought that this characteristic is reflected in the amount of change in the tilt angle per unit time. Also, for example, when the owner performs a disposal operation to discard the seasonings in container 1, such as by replacing the contents of container 1, there is no need to consider using too much seasoning compared to when performing an operation to use, so there is a characteristic that the container is tilted sharply once and then maintained at that tilt angle, and it is thought that this characteristic is reflected in the amount of change in the tilt angle per unit time. Also, for example, when the owner performs a washing operation to wash container 1, there is a characteristic that the tilt of container 1 changes in a short time in order to wash the inside of container 1 compared to when performing an operation to use, and it is thought that this characteristic is reflected in the amount of change in the tilt angle per unit time.

[0091] Therefore, the operation estimation unit 3 can obtain operation information indicating the operation that caused the tilt of container 1 using the amount of change in the tilt angle per unit time. If container 1 is for holding seasonings as shown in Figure 12, the operation information could include, for example, the use operation of seasonings in cooking, the disposal operation of discarding seasonings, and the washing operation of washing container 1.

[0092] Next, the operation estimation unit 3 determines whether the acquired operation information, which is the operation that caused the tilting of container 1, matches a specific operation. For example, by setting a specific operation as a usage operation and not estimating the amount of spillage caused by operations other than usage operations, such as disposal operations, the operation estimation unit 3 can estimate the amount of seasoning used by the owner of container 1. When the operation estimation unit 3 outputs the operation information to the spillage estimation unit 4, it includes information in the operation information indicating whether the operation that caused the tilting of container 1 matches a specific operation.

[0093] Some containers 1 for condiments have small outlets, which limit the amount of liquid that flows out when the container 1 is tilted. In this case, it takes time for the contents of container 1 to reach the amount of liquid that is expected to be held in container 1 at a certain tilt angle. Therefore, when the flow rate estimation unit 4 estimates the flow rate from the orientation information, it may use time in addition to the tilt angle to estimate the flow rate. For example, the flow rate estimation unit 4 may determine that the contents of container 1 have reached the amount of liquid that is expected to be held in container 1 when container 1 has been tilted at a certain tilt angle for a certain period of time or longer, and then estimate the flow rate.

[0094] Furthermore, if the object held in container 1 is a powder, the powder may solidify due to moisture or other factors, limiting the amount that can be dispensed. In this case, the dispensing amount estimation unit 4 may estimate the amount of dispensing using a sensor that measures impact, such as an acceleration sensor. If the powder held in container 1 solidifies and does not come out, the person holding container 1 is likely to try to get the powder out by shaking container 1. Therefore, the acceleration sensor measures the acceleration applied to container 1, and if the absolute value of the measured acceleration is far from 1G, it determines that an acceleration other than gravity is applied to container 1. In other words, the acceleration sensor determines that the solidification of the powder has been resolved by the person holding container 1 shaking container 1. The dispensing amount estimation unit 4 then estimates the amount of powder that has been dispensed since the acceleration sensor determined that the solidification of the powder had been resolved. Note that if an acceleration sensor is used for the attitude sensor 2, the dispensing amount estimation unit 4 can estimate the amount of dispensing without requiring an additional sensor.

[0095] In Embodiment 2, a specific example of how the output unit 6 presents output information will be described. For example, suppose the owner of container 1 is a sick person who requires strict health management, and a usage operation is set as a specific operation, and the discharge amount estimation unit 4 estimates the discharge amount as the amount used during that usage operation. In this case, the output unit 6 presents the discharge amount as the amount used during that usage operation to the sick person, so that the sick person can understand the amount used. The output unit 6 may also present information as output information, such as whether the amount used is excessive compared to a pre-set appropriate amount. The appropriate amount used may be calculated from the sick person's height, weight, age, blood pressure, and gender, etc.

[0096] Furthermore, for example, the output unit 6 may display the amount used to the patient's attending physician or family doctor. This allows the attending physician or family doctor to understand the amount of seasoning used by the patient and to accurately diagnose the patient's condition, manage their health, and provide guidance on health management methods.

[0097] Furthermore, for example, if the amount of seasoning used by the container 1 approaches the original capacity of the container 1, that is, if the owner of the container 1 is about to run out of seasoning, the output unit 6 may include information in the output information prompting the owner of the container 1 to purchase additional seasoning, and present this information to the owner of the container 1. In this case, the output unit 6 may also place an order for additional seasoning via an external information device.

[0098] Furthermore, for example, if foreign matter contamination occurs in the seasoning held in container 1, the output unit 6 may acquire that information via an external information device. In this case, the output unit 6 includes information in the output information urging the user to stop using the seasoning and presents it to the user of container 1. The output unit 6 may also present the user of container 1 with the amount of the contaminated seasoning used so that the user can keep track of the amount used.

[0099] Furthermore, for example, if the output unit 6 obtains information via an external information device that foreign matter contamination or the like has occurred in the seasoning held in container 1 after the owner of container 1 has started using the seasoning held in container 1, it may include information in the output information indicating the timing of when the owner of container 1 used the seasoning held in container 1 in the past and present it to the owner of container 1. This allows the owner of container 1 to track the usage history of the seasoning in which foreign matter contamination or the like occurred, and to estimate when the seasoning was used or who may have eaten a dish in which the seasoning was used.

[0100] The information processing device 200 in Embodiment 2, similar to Embodiment 1, includes an operation estimation unit 3 that extracts feature quantities in the operation that caused the tilt of the container 1 from posture information showing the change in the tilt angle of the container 1 over time, and uses these feature quantities to determine whether the tilt of the container 1 is due to a specific operation; and an outflow amount estimation unit 4 that, if the tilt of the container 1 is due to a specific operation, acquires correspondence information representing the correspondence between the volume of the contents of the object held in the container 1 and the tilt angle, and uses the correspondence information and the tilt angle to estimate the outflow amount of the object flowing out of the container 1. With the above configuration, the information processing device 200 in Embodiment 2 can distinguish features in the posture information for each operation that caused the tilt of the container 1, accurately determine whether the tilt of the container 1 is due to a specific operation, and estimate the outflow amount of the object due to that specific operation.

[0101] Furthermore, the output unit 6 in Embodiment 2, similar to Embodiment 1, presents output information that includes at least one of the following: operation information indicating the operation that caused the tilt of container 1, and discharge amount information indicating the amount of discharge. With the above configuration, the information processing device 200 in Embodiment 2 can, for example, present information such as the amount of seasoning used, which is the amount of discharge in a specific operation, to the owner of container 1 or their attending physician. This allows the owner of container 1 or their attending physician to know the amount of seasoning used by the owner of container 1.

[0102] Furthermore, in Embodiment 2, the object other than the beverage held in the container 1 is, for example, a seasoning, and the specific operation for which the outflow amount estimation unit 4 estimates the outflow amount is the operation of using the seasoning. With the above configuration, the information processing device 200 in Embodiment 2 can estimate the amount of seasoning used by the owner of the container 1 by estimating the outflow amount only when the owner of the container 1 uses the seasoning, among the operations that caused the container 1 to tilt.

[0103] In the second embodiment, the container 1 is designed so that the object inside can be removed by tilting the container 1. Therefore, it does not include containers that use pressure, such as pressing a pump, to remove the object, or containers that allow the object to be removed directly from the container 1 with a spoon or the like.

[0104] Furthermore, although the container 1 in the measurement system 101 of Embodiment 2 was described as being for holding condiments such as soy sauce dispensers and salt shakers, it is not limited to this, and although the object held in the container 1 was described as being a condiment, it is not limited to this. In other words, the container 1 in the measurement system 101 of Embodiment 2 can be any container from which an object can be removed by tilting the container 1. For example, the container 1 in the measurement system 101 of Embodiment 2 may be the cargo bed of a truck such as a dump truck used to transport objects, and the object held in the container 1 may be transported material such as soil and sand. Even in this case, the information processing device 200 can estimate the amount of outflow only at the time of a specific operation among the operations that caused the tilting of the container 1. In addition, since the information processing device 200 does not need to directly measure the amount of soil and sand outflow, it can easily suppress the effect of contamination of the sensor by soil and sand compared to the case in which a sensor that measures the amount of outflow by coming into contact with the soil and sand is used.

[0105] Furthermore, for example, the container 1 in the measurement system 101 of Embodiment 2 may be steelmaking equipment such as a converter, and the object held in the container 1 may be pig iron used in steel production. Even in this case, the information processing device 200 can estimate the amount of outflow only during a specific operation among the operations that caused the tilting of the container 1. In addition, since the information processing device 200 does not need to directly measure the amount of pig iron outflow, it can easily estimate the amount of outflow even when the pig iron is at a high temperature and it is difficult to bring the sensor into contact with it.

[0106] (Note 1) An operation estimation unit extracts characteristic quantities related to the operation that caused the tilt of the container from posture information showing the change in the tilt angle of the container over time, and uses these characteristic quantities to determine whether or not the tilt of the container is due to a specific operation. If the tilt of the container is caused by the specific operation, the outlet amount estimation unit acquires correspondence information representing the correspondence between the volume of the contents of the object held in the container and the tilt angle, and estimates the amount of the object that flows out of the container using the correspondence information and the tilt angle. An information processing device equipped with the following features. (Note 2) The aforementioned feature quantity is the amount of change in the slope angle per unit time. The information processing device described in Appendix 1. (Note 3) The operation estimation unit comprises a data acquisition unit that acquires the posture information, and an inference unit that outputs operation information indicating the operation that caused the tilt of the container from the posture information acquired by the data acquisition unit. The information processing device described in Appendix 1 or Appendix 2. (Note 4) The system further includes an output unit that presents output information which includes at least one of the following: operation information indicating the operation that caused the tilt of the container, and discharge amount information indicating the amount of discharge. An information processing device as described in any one of the items from Appendix 1 to Appendix 3. (Note 5) The aforementioned object is a beverage, and the aforementioned specific operation is a drinking operation of the beverage. An information processing device as described in any one of the items from Appendix 1 to Appendix 4. (Note 6) The aforementioned object is a seasoning, and the aforementioned specific operation is an operation that uses the said seasoning. An information processing device as described in any one of the items from Appendix 1 to Appendix 4. (Note 7) The operation estimation unit obtains operation information from the posture information that indicates the operation that caused the tilt of the container, and determines whether the operation that caused the tilt of the container matches the predetermined specific operation. An information processing device as described in any one of the items from Appendix 1 to Appendix 6. (Note 8) An information processing device described in any one of the items from Appendix 1 to Appendix 7, A posture sensor that measures the tilt angle of the container to acquire posture information and outputs the posture information to the operation estimation unit and the outflow rate estimation unit, A measurement system equipped with the following features. (Note 9) The operation estimation unit extracts feature quantities related to the operation that caused the tilt of the container from the posture information showing the change in the tilt angle of the container over time, and uses these feature quantities to determine whether or not the tilt of the container is caused by a specific operation. The outflow volume estimation unit, when the tilt of the container is caused by the specific operation, acquires correspondence information representing the correspondence between the volume of the object held in the container and the tilt angle, and estimates the amount of the object flowing out of the container using the correspondence information and the tilt angle. An information processing method comprising the following: (Note 10) The operation estimation unit extracts feature quantities related to the operation that caused the tilt of the container from the posture information showing the change in the tilt angle of the container over time, and uses these feature quantities to determine whether or not the tilt of the container is caused by a specific operation. The outflow volume estimation unit, when the tilt of the container is caused by the specific operation, acquires correspondence information representing the correspondence between the volume of the object held in the container and the tilt angle, and estimates the amount of the object flowing out of the container using the correspondence information and the tilt angle. A program that causes a computer to execute something. [Explanation of Symbols]

[0107] 100, 101 Measurement system, 200 Information processing device, 1 Container, 2 Attitude sensor, 3 Operation estimation unit, 3a Data acquisition unit for operation estimation unit, 3b Inference unit, 4 Outflow volume estimation unit, 5 Storage unit, 6 Output unit, 10 Learning device, 10a Data acquisition unit for learning device, 10b Model generation unit, 10c Stored model unit, 90 Processing circuit, 92 Processor, 94 Memory

Claims

1. An operation estimation unit extracts characteristic quantities related to the operation that caused the tilt of the container from posture information showing the change in the tilt angle of the container over time, and uses these characteristic quantities to determine whether or not the tilt of the container is due to a specific operation. If the tilt of the container is caused by the specific operation, the outlet amount estimation unit acquires correspondence information representing the correspondence between the volume of the contents of the object held in the container and the tilt angle, and estimates the amount of the object that flows out of the container using the correspondence information and the tilt angle. An information processing device equipped with the following features.

2. The aforementioned feature quantity is the amount of change in the slope angle per unit time. The information processing apparatus according to claim 1.

3. The operation estimation unit comprises a data acquisition unit that acquires the posture information, and an inference unit that outputs operation information indicating the operation that caused the tilt of the container from the posture information acquired by the data acquisition unit. The information processing apparatus according to claim 1.

4. The system further includes an output unit that presents output information which includes at least one of the following: operation information indicating the operation that caused the tilt of the container, and discharge amount information indicating the amount of discharge. The information processing apparatus according to claim 1.

5. The aforementioned object is a beverage, and the aforementioned specific operation is a drinking operation of the beverage. The information processing apparatus according to claim 1.

6. The aforementioned object is a seasoning, and the aforementioned specific operation is an operation that uses the said seasoning. The information processing apparatus according to claim 1.

7. The operation estimation unit obtains operation information from the posture information that indicates the operation that caused the tilt of the container, and determines whether the operation that caused the tilt of the container matches the predetermined specific operation. The information processing apparatus according to claim 1.

8. An information processing device according to any one of claims 1 to 6, A posture sensor that measures the tilt angle of the container to acquire posture information and outputs the posture information to the operation estimation unit and the outflow rate estimation unit, A measurement system equipped with the following features.

9. The operation estimation unit extracts feature quantities related to the operation that caused the tilt of the container from the posture information showing the change in the tilt angle of the container over time, and uses these feature quantities to determine whether or not the tilt of the container is caused by a specific operation. The outflow volume estimation unit, when the tilt of the container is caused by the specific operation, acquires correspondence information representing the correspondence between the volume of the object held in the container and the tilt angle, and estimates the amount of the object flowing out of the container using the correspondence information and the tilt angle. An information processing method comprising the following:

10. The operation estimation unit extracts feature quantities related to the operation that caused the tilt of the container from the posture information showing the change in the tilt angle of the container over time, and uses these feature quantities to determine whether or not the tilt of the container is caused by a specific operation. The outflow volume estimation unit, when the tilt of the container is caused by the specific operation, acquires correspondence information representing the correspondence between the volume of the object held in the container and the tilt angle, and estimates the amount of the object flowing out of the container using the correspondence information and the tilt angle. A program that causes a computer to execute something.

Images