JP7539851B2 - Bed Equipment - Google Patents

Description

This embodiment relates to a bed device.

A system has been proposed that detects dangerous events and abnormal behaviors of users of bed devices. For example, Patent Document 1 discloses an invention that can detect foreign objects between the main frame and the floor surface from below the main frame.

JP 2019-097914 A

The objective of this disclosure is to provide a bed apparatus that can properly determine abnormalities in the bed apparatus.

The bed device disclosed herein is characterized by having a drive unit that transmits a drive force to a mechanism in the bed body, an acquisition unit that acquires the load acting on the drive unit, and a determination unit that determines an abnormality based on the acquired load.

This disclosure makes it possible to provide a bed device and the like that can properly detect abnormal behavior.

1 is a diagram showing an entire bed system (bed apparatus) according to an embodiment of the present invention; 1 is a diagram for explaining the configuration of a bed system in this embodiment. FIG. 1 is a diagram for explaining the configuration of a bed system in this embodiment. FIG. 4 is an operational flow for explaining a main process in the present embodiment. 5 is a diagram for explaining a first abnormality determination process in the present embodiment. FIG. 10 is a diagram for explaining a second abnormality determination process in the present embodiment. FIG. FIG. 4 is a diagram for explaining the operation in the present embodiment. 11 is a diagram for explaining a third abnormality determination process in the present embodiment. FIG. FIG. 4 is a diagram for explaining the operation in the present embodiment. 11A and 11B are diagrams for explaining a response process in the present embodiment. FIG. 4 is a diagram for explaining the operation in the present embodiment. 5 is a graph illustrating a change in load in the present embodiment. 5 is a graph illustrating a change in load in the present embodiment.

Below, one embodiment for implementing the disclosed bed system will be described with reference to the drawings. Note that the contents of this disclosure merely show one example of an embodiment, and are not limited to the disclosed numerical values and configurations, but also include equivalents that can be conceived by a person skilled in the art.

A bed system including a bed apparatus is designed to detect dangerous events involving the user of the bed apparatus. The bed system detects abnormal behavior (abnormal movements) of the user or the bed apparatus as movements that can lead to dangerous events. Abnormal movements include the user, a trash can, an IV pole, etc., getting caught in the bed apparatus, a part of the user's body getting caught in a fence or a gap between the fence and a board, the user climbing over the fence of the bed apparatus, or assuming an upright position on the bed apparatus. Since abnormal movements of users can lead to risks such as tripping or falling, there is a need for bed systems to properly detect such movements.

To detect such abnormal behavior, bed systems typically provide various sensors in the bed device. For example, it is common to provide the bed device with a contact sensor, an IR sensor, or a camera. For example, the bed system can detect entrapment when it detects that the user has come into contact with a contact sensor.

However, adding sensors to a bed system increases costs accordingly. In addition, many sensors have a short service life compared to the components of the bed device itself, which reduces maintenance performance. Furthermore, depending on the location of the sensor, it may be difficult for the user to use the bed device. Also, depending on the location of the sensor, the location where the sensor can detect the user may be limited to the location, making it difficult to properly detect abnormal behavior.

In addition, the system disclosed in Patent Document 1 cannot function unless the frame is not hung between the sensors, which places significant constraints on the frame design. This creates a problem in that it is not possible to lower the bed height of the bed device to its lowest possible level. In addition, the system disclosed in Patent Document 1 has the problem that, depending on how it is applied, it may stop working just by inserting a power cable. In particular, in the case of a bed device that achieves a low bed height, the frame must be lowered to a very low position, which creates a problem in that the method of Patent Document 1 cannot be adopted.

Therefore, according to the bed system of this embodiment, it is possible to provide a system that can properly detect abnormal behavior without adding sensors. In addition, it is possible to provide a system that requires only a minimum number of additional sensors, even if sensors are added. In addition, it is possible to provide a system that can properly detect abnormal behavior without creating blind spots that cannot be detected by the sensors, even if sensors are added.

In this specification, the term "user" refers to a person who uses a bed device (mattress), and is not limited to those who are receiving treatment for an illness (e.g., patients), but can also be a person who receives care in a facility or is in a bed device (e.g., a person lying on their back), even if they are healthy.

In addition, in this specification, staff, etc. includes not only those who assist users, including medical professionals, staff at facilities, and household staff, but also those involved with users.

[1. Description of the entire system]
Fig. 1 is an overall view for explaining an outline of a system 1 in this embodiment. The system 1 is configured to include at least a bed apparatus 10. For example, the bed apparatus 10 has a control device. The control device may be connected to an operation device and a display device. Although not shown in Fig. 1, the system 1 may appropriately include the control device, the operation device, and the display device as separate devices.

The bed device 10 has one or more bottoms 20. The bottoms 20 are composed of a movable back section, a seat section, an upper leg section, a foot section, etc., and the back and legs can be raised by the operation of each bottom. For example, FIG. 1 shows a state in which the back section is raised and the back is raised. A mattress 30 may be placed on the bottom 20. The mattress 30 may be a mattress made of urethane, or may be an air cell made of multiple air cells. The configuration of only the bed device 10 (the configuration not including a mattress, etc.) is called the bed body.

As will be described later, the bed apparatus 10 is equipped with a lifting mechanism, and the height (floor height) of the bottom part of the bed apparatus 10 can be changed. In this specification, the floor height of the bed apparatus refers to the distance from the floor on which the bed apparatus 10 is placed to the bottom (the bottom when in a horizontal state). The floor height of the bed apparatus 10 may be the distance from the floor to the bottom, or the distance from the floor to the upper frame, or the distance from the floor to the top of the mattress, other than the distance from the floor to the bottom. The floor height of the bed apparatus may be based on the lower frame instead of the floor.

Furthermore, the bed apparatus 10 can perform an operation of tilting the entire bed apparatus 10 (tilt operation). The bed apparatus 10 may be a bed apparatus in which none of its parts are movable.

The bed apparatus 10 can also be equipped with a removable board. As shown in FIG. 1, the board can have a headboard 50 installed on the head side and a footboard 55 installed on the foot side. The bed apparatus 10 can also be equipped with a removable side rail 52. The side rail 52 can be installed on the left and right sides of the bed apparatus. A plurality of side rails 52 can also be installed in the longitudinal direction of the bed apparatus 10. For example, two side rails 52 can be installed on the head side and foot side of the bed apparatus 10.

The bed device 10 may detect the attachment or detachment of the board (headboard 50 or footboard 55) or the attachment or detachment of the side rails 52 based on the detected load. The bed device 10 may also detect an abnormality when the headboard 50, footboard 55, or side rails 52 are detached.

[2. Functional configuration]
Fig. 2 and Fig. 3 are diagrams for explaining the configuration of the bed system 1. Fig. 2 is a diagram for explaining the functional configuration of the bed system 1, and Fig. 3 is a diagram for typically explaining the relationship between the bottom 20 and each driving unit. As shown in Fig. 3, the bottom 20 has a back bottom 22, a waist bottom 24, a knee bottom 26, and a foot bottom 28 under the user P.

The control unit 100 is a functional unit for controlling the entire bed system 1. The control unit 100 realizes various functions by reading and executing various programs stored in the memory unit 110, and is composed of one or more control devices (e.g., a CPU (Central Processing Unit)). The control unit 100 may also function as a bottom control unit 102, a height control unit 104, and an erroneous detection determination unit 106.

The bottom control unit 102 controls the drive unit 120 to operate the bottom 20 and perform back-raising and knee-raising operations. The height control unit 104 also controls the drive unit 120 to control the height of the bed device 10.

The erroneous detection determination unit 106 determines whether or not a detection has occurred erroneously based on the load-related information (e.g., changes in the load value) output by the load detection unit 130.

For example, when the user turns over or moves on the bed device 10, or changes from sitting on the edge to standing, causing a change in the user's posture, the detected load changes. The control unit 100 may detect this load change as an abnormal movement. The erroneous detection determination unit then determines whether the load change is an erroneous detection. The erroneous detection determination unit 106 may predetermine a range for determining an erroneous detection, or may determine an erroneous detection when there is a change of a predetermined amount or more. For example, the erroneous detection determination unit 106 may determine an erroneous detection when there is a change of preferably 8 to 12 kg. In order to issue an alert early for the entire system, the range for determining an erroneous detection is narrowed, making it possible to issue an alert early.

For example, the control unit 100 does not use the load value output by the load detection unit 130 as is, but uses the load value from which erroneous detections have been removed by the erroneous detection determination unit 106. For example, the erroneous detection determination unit 106 may determine that a load value that significantly deviates from the average value of the load values detected immediately before is an erroneous detection. Furthermore, the control unit 100 (erroneous detection determination unit 106) may use the average value of the load values continuously output from the load detection unit 130 as the load value. Furthermore, the erroneous detection determination unit 106 uses the load change in an event in which a load change occurs due to something other than pinching, and determines that the load change is due to something other than pinching if it corresponds to any of the patterns.

The erroneous detection determination unit 106 may also determine erroneous detection by utilizing changes in the load values and using artificial intelligence (machine learning, etc.). Here, the erroneous detection determination unit 106 can use one or a combination of a plurality of the following load values as input data.
A load value detected by the load detection unit 130 An average value of the load values detected by the load detection unit 130 A normalized value of the load value detected by the load detection unit 130

In this embodiment, the erroneous detection determination unit 106 performs an erroneous detection determination, for example, by using the load value detected by the load detection unit 130 and the normalized value. At this time, the erroneous detection determination unit 106 may perform multiple determinations other than the two-category determination of whether or not there is an erroneous detection.

The load change may also be determined at predetermined time intervals. For example, the erroneous detection determination unit 106 may obtain the load value preferably every 0.5 to 4 seconds, and more preferably every 1 to 2 seconds, and perform a erroneous detection determination.

The storage unit 110 is a functional unit that stores various programs and various data necessary for the operation of the bed system 1. The storage unit 110 is composed of, for example, a semiconductor memory, a solid state drive (SSD), a hard disk drive (HDD) which is a magnetic disk device, etc. The storage unit 110 may be a non-transient storage medium or a temporary storage medium.

The memory required by the control unit 100 during execution may be secured in the storage unit 110, or a storage area may be provided within the control unit 100.

The memory unit 110 stores a load value table 112 and a trained model 114. The load value table 112 is a table that stores load values output by the load detection unit 130 described below. The trained model 114 is a trained model that is the result of machine learning in this embodiment. The trained model 114 is pre-stored in the memory unit 110, but may be trained (updated) by the control unit 100 as appropriate.

The load value table 112 may store the load values detected by the load detection unit 130, values obtained by averaging/normalizing the load values, etc. The load value table 112 may store the raw data output by the load detection unit 130, or values converted into load values.

The drive unit 120 is a drive device for driving other devices, such as an actuator. That is, the drive unit 120 can operate the bottom, frame, etc. connected to it by transmitting a drive force to the bottom, frame, etc. by advancing or retracting the piston rod of the actuator. That is, the drive unit 120 can transmit a drive force to each mechanism in the bed body.

The driving unit 120 of this embodiment can detect the load on the piston rod as a load signal. The load detection unit 130 outputs information about the load (e.g., a load value or a waveform of the load signal) based on the load signal detected by the driving unit 120. The driving unit 120 and the load detection unit 130 may be integrated into one unit. That is, the driving unit 120 may output information about the load to the control unit 100.

The operation unit 140 accepts operations from the user. For example, the operation unit 140 accepts input from an operation button or via a touch panel. The display unit 150 displays various information to the user. The display unit 150 is configured, for example, by a liquid crystal display or an organic EL panel.

The notification unit 160 performs a notification operation for the user. The notification unit 160 may output voice or an alarm sound from a speaker, or output light from an LED or the like, for example. The communication unit 170 communicates with other devices.

The operation unit 140, display unit 150, notification unit 160, and communication unit 170 may be provided in the bed system 1 as necessary.

For example, the bed system 1 may be connected to another terminal device (smartphone) via the communication unit 170. In this case, the operation unit 140, the display unit 150, and the notification unit 160 may be realized by the terminal device (smartphone).

Next, the relationship between the control unit 100, the drive unit 120, and the load detection unit 130 will be explained with reference to FIG. 3.

The bottom control unit 102 can control the back drive unit 32, the knee drive unit 34, and the head drive unit 36 of the drive unit 120. For example, when a user performs an operation to raise the back, the bottom control unit 102 controls the back drive unit 32 to perform the back-raising operation.

Specifically, for example, one end of the back drive unit 32 is rotatably fixed to the frame of the bed device 10, and the other end (the tip of the piston rod) is rotatably fixed to the back bottom 22 via a link mechanism or the like. When the piston rod of the back drive unit 32 advances, it rotates around the waist bottom 24 side of the back bottom 22, raising the head side. This realizes the back-raising operation.

Similarly, when a knee-raising operation is performed by the user, the bottom control unit 102 controls the knee drive unit 34 to perform the knee-raising operation. When the knee drive unit 34 operates, the knee bottom 26 operates, but the foot bottom 28 may also operate in conjunction with it. That is, when the knee bottom 26 rotates around the waist side and the foot side rises, the knee bottom 26 side of the foot bottom 28 rises and the foot side descends. This allows the knee bottom 26 and foot bottom 28 to form a mountain shape, making it possible to properly support the user's knees and feet.

In addition, if the back bottom 22 is divided at the head portion, the bottom on the head side may be operated by the head drive unit 36.

The height control unit 104 controls the height drive unit 40 when the user operates to raise or lower the height of the bed device 10. The height control unit 104 adjusts the height of the bed device 10 by controlling the height drive unit 40. That is, the height control unit 104 can raise or lower the bed height. In addition, raising the bed height can be called bed raising, and lowering the bed height can be called bed lowering.

Here, the height drive unit 40 may be a different drive device (actuator) for the head side and the foot side. By providing different drive devices for the head side and the foot side, the height drive unit 40 can not only change the height of the bed device 10 but also perform tilt control.

In addition, the control unit 100 has been described as a method of controlling the bed device 10 using the bottom control unit 102 and the height control unit 104 as examples, but other controls may also be performed. For example, the control unit 100 may perform control to expand and contract the bed device 10 in the width direction (length direction). In addition, the control unit 100 may perform control to change the position of the board or change the shape of the side rails 52, for example.

The load detection unit 130 receives a load signal from the drive unit 120. The load detection unit 130 receives load signals from the back drive unit 32, the knee drive unit 34, the head drive unit 36, and the height drive unit 40. The load detection unit 130 then performs a predetermined process on the received load signal, for example, and outputs it to the control unit 100.

The function of the load detection unit 130 may be provided in each drive unit. In this case, the back drive unit 32, knee drive unit 34, head drive unit 36, and height drive unit 40 output a load signal to the control unit 100.

The configuration of the bottom 20 and the drive unit 120 shown in FIG. 3 is an example, and can be selected as necessary. For example, the drive unit 120 may include only the back drive unit 32 and the knee drive unit 34. The drive unit 120 may also include a drive unit not disclosed in FIG. 3 (for example, a drive unit for causing the user P to perform a rotational motion (rolling motion)).

Although the load detection unit 130 detects the load signal from the drive unit 120, it may also detect the load from another device. For example, if an air mattress composed of air cells is placed on the bottom 20, the load related to the air cells may be detected. The load signal may also be detected from a load detector such as a load cell provided on the bed frame of the bed device 10.

3. Processing Flow
Next, the flow of processing in this embodiment will be described with reference to Fig. 4. The main processing in this embodiment will be described with reference to Fig. 4.

[3.1 Main Processing]
First, the control unit 100 judges whether the bed apparatus 10 is in operation (step S102). The bed apparatus 10 is in operation when the bed apparatus 10 is performing some kind of movement. For example, the state in which the bed apparatus 10 is in operation refers to a state in which the control unit 100 (bottom control unit 102) is controlling the drive unit 120 and a back-raising operation (back-lowering operation) or a knee-raising operation (knee-lowering operation) is being performed. The state in which the bed apparatus 10 is in operation also includes a state in which the control unit 100 (height control unit 104) is controlling the drive unit 120 and an operation of changing the height of the bed apparatus 10 is being performed.

In addition, the operation of the bed apparatus 10 may include not only a state in which an operation is actually being performed in response to a user operation, but also a state in which a user operation has been performed and an operation is about to be performed. In addition, the operation of the bed apparatus 10 may include a predetermined period after the user operation has ended.

Next, when the bed apparatus 10 is in operation (step S102; Yes), the control unit 100 executes an abnormality determination process (step S104). By executing the abnormality determination process, the control unit 100 recognizes the user's actions and determines whether there is an action that is determined to be abnormal (an abnormal action) among the recognized user actions.

Here, abnormality refers to an action that is different from normal, for example, an action that may cause a dangerous event for the user compared to normal actions. For example, abnormal actions are actions that are different from actions normally performed by the user, actions that may lead to the user tripping or falling, or actions that may lead to an accident involving the user. Furthermore, when determining whether an abnormality exists, it may be determined that an abnormality exists based not only on a simple action, but also on the state, environment, or actions of staff other than the user, etc.

For example, if the bed apparatus 10 is controlled to lower its height even though an object is placed under it (under the lower frame), the object will end up being caught between the floor and the frame. Such an operation of the bed apparatus 10 may also cause a dangerous event, so the control unit 100 judges it as an abnormality.

In this way, the control unit 100 determines that there is an abnormality not only in the user's movements but also in the movements of the bed device 10 if there is any movement that could lead to an accident.

When the control unit 100 determines that an abnormality has occurred in the abnormality determination process (step S106; Yes), it outputs an alert (step S108). Here, the control unit 100 may output the alert, for example, by issuing an alarm sound or voice from the notification unit 160, by displaying a message on the display unit 150, or by notifying another device or sending an email via the communication unit 170.

Then, the control unit 100 executes a response process (step S110). When the control unit 100 executes the response process, if it determines that an abnormality has occurred, it executes a process corresponding to the abnormality. Then, the control unit 100 repeatedly executes the process until the entire operation is completed (step S112; No → step S102).

The control unit 100 can select and execute steps of this process as necessary. For example, the control unit 100 may not output an alert in step S108, or may not execute the response process in step S110.

For convenience of explanation, steps S104 and S106 have been explained separately, but the control unit 100 may execute them as a single process. For example, the abnormality determination process may be executed appropriately by the control unit 100, and an interrupt process may be generated when an abnormality is determined. The abnormality determination process and response process in this embodiment will be explained below.

[3.2 Abnormality determination process]
[3.2.1 First abnormality determination process]
FIG. 5 is a flow diagram for explaining the first abnormality determination process. First, the control unit 100 acquires a load signal from the load detection unit 130 (step S1102). The control unit 100 detects the peaks of the load signal (step S1104). The control unit 100 determines that an abnormality has occurred when the difference in numerical value between the peaks of the detected load signal is equal to or greater than a certain value (step S1108). For example, the control unit 100 detects the load signal preferably every 0.1 to 1 second and detects the peak value. In addition, in step S1108, the control unit 100 may determine the difference in absolute value or in relative ratio. For example, the control unit 100 may determine that an abnormality has occurred when the load fluctuates preferably by 5% to 20%. In addition, for example, the control unit 100 may determine that an abnormality has occurred when the load fluctuates preferably by 5 kg to 10 kg.

As an example of this process, the control unit 100 determines that the abnormality detected at this time is "trapped" depending on the type of the drive unit 120 that detected the load signal. Trapped refers to a foreign object (object or human body) being trapped between the frame of the bed device 10 and the floor, or a foreign object being trapped between the bottom and the frame.

[3.2.2 Second abnormality determination process]
6 is a flow chart for explaining the second abnormality determination process, which is a process for determining an abnormality by comparing the load signal with the average.

The control unit 100 acquires a load signal from the load detection unit 130 (step S1202). The control unit 100 also calculates the average of the acquired load signals (step S1204).

Here, the control unit 100 determines whether the difference between the load signal acquired from the load detection unit 130 and the average load signal is equal to or greater than a threshold value (step S1206). If the difference between the load signal and the average load signal is equal to or greater than the threshold value, the control unit 100 determines that an abnormality has occurred (step S1206; Yes -> step S1208).

The mechanism by which the control unit 100 determines an abnormality will be described with reference to FIG. 7. FIG. 7 is a graph that shows a schematic representation of a load signal. The solid line shows a graph based on the load signal. The dashed line shows a graph based on the average of the load signal.

Here, the control unit 100 compares the load signal with the average. Then, if the difference between the load signal and the average (e.g., hc in FIG. 7) is equal to or greater than a predetermined difference, the control unit 100 determines that an abnormality has occurred.

Here, the difference from the average determined by the control unit 100 may be, for example, an absolute value, or a percentage of the difference between the average and the load signal (load value). For example, the control unit 100 may determine that an abnormality has occurred if the deviation from the average is 20% or more. The value determined by the control unit 100 may also be based on the integral value of the difference between the average value and the load signal (load value) or the moving average value. For example, the control unit 100 may determine that an abnormality has occurred (such as pinching) if the calculated integral value exceeds a predetermined threshold value.

In this way, the second abnormality determination process can determine an abnormality by using the average of the load values detected by the drive unit 120 (e.g., actuator) and the load detection unit 130. For example, when a load sensor built into the actuator is used to check the load change during operation, the following error factors can be considered:

First, the load applied to the load sensor changes depending on the rotation angle of the screw shaft. Second, the load applied to the load sensor changes depending on the bed height and angle of the bed device 10. Therefore, when determining an abnormality (for example, detecting entrapment), an averaging process is performed, and if the load changes significantly from the average value, it is determined to be entrapment, making it possible to make a more accurate determination. In addition, averaging the load values is expected to have the effect of improving resistance to external noise.

The control unit 100 can also select the load signal to be acquired depending on the type of abnormality to be determined. For example, the control unit 100 can determine the type of abnormality (type and position of pinching) by selecting the following drive units 120:

In case of being pinched under the bed: Height driving unit 40
In case of pinching with the back lowered, the back drive unit 32
In case of pinching with knees down, knee drive unit 34
・ Prevents the user or objects such as arms or legs from getting caught in the bed rails Back drive unit 32, knee drive unit 34

The control unit 100 may also identify the position of entrapment based on the load signal output from the drive unit 120 described above. For example, when the control unit 100 acquires a load signal acquired from the height drive unit 40 that has a numerical difference between peaks of the load signal that is greater than or equal to a certain value, it is possible to identify that entrapment has occurred under the bed.

The control unit 100 may also identify the cause of the abnormality based on the way in which the load signal output from the drive unit 120 described above changes. For example, suppose that pinching is determined as an abnormality when controlling the lowering of the height of the bed device 10. In this case, if the change in the load value is steep, the control unit 100 determines that a hard object has been pinched. Also, if the change in the load value is gradual, the control unit 100 determines that a soft object has been pinched.

In addition, if there are multiple height drive units 40, an abnormality may be determined based on the output from each height drive unit 40. In other words, the height drive unit 40 is equipped with two drive units (actuators), a head drive unit and a foot drive unit, and the control unit 100 controls the height using these two drive units.

At this time, the control unit 100 determines that an abnormality has occurred (for example, pinching has occurred) based on the following conditions.
(1) Determine whether there is an abnormality based on the load value detected based on the head-side drive unit. (2) Determine whether there is an abnormality based on the load value detected based on the foot-side drive unit. (3) Determine whether there is an abnormality based on the sum of the load value detected based on the head-side drive unit and the load value detected based on the foot-side drive unit.

[3.2.3 Third abnormality determination process]
8 is a flow diagram illustrating the third abnormality determination process. When the third abnormality determination process is executed, the bed system 1 stores the trained model 114 that has already been trained.

The trained model 114 is information about a model in which the load detection unit 130 uses information about the load based on the load signal output from the drive unit 120 as an input element and an abnormality (e.g., the type of abnormality or the details of the abnormality) as an output element. In other words, by using the trained model 114, the control unit 100 can determine that an abnormality has occurred and identify the type of abnormality and the location where the abnormality has occurred.

To generate the trained model 114, information about the load (load signal, load value, load waveform) for each type of drive unit 120 (back drive unit 32, knee drive unit 34, head drive unit 36, height drive unit 40) is prepared as input data for learning, and the actions at that time (including abnormal actions and normal actions) are prepared as teacher data. Note that the input data may also include other information (for example, the user's height, weight, age, information about illness, biometric information, gender, and treatment details).

When input data is given to the learning model, a learning process is performed so that the data is classified as either an abnormal motion or a normal motion and output. When the learning process is performed, the output data is compared with the teacher data, and the weights in the learning model are adjusted. This generates the trained model 114.

The trained model 114 is constructed through analysis using machine learning techniques such as neural networks, decision trees, and support vector machines. In addition, the control unit 100 may train (update) the trained model 114 using machine learning at a predetermined timing, using information related to the load and abnormalities (e.g., abnormal movements) input by staff or the like as training data.

The trained model 114 may also learn by associating the type of drive unit 120 with an abnormality. For example, FIG. 9 is a table showing combinations of drive units 120 and types of abnormalities that can be determined (classification of abnormalities). For example, the control unit 100 can identify the type of abnormality as standing up on the bed based on the load signals from the back drive unit 32 and the height drive unit 40.

Similarly, the control unit 100 can identify the type of abnormality as the user climbing over a fence based on the load signals from the back drive unit 32 and the height drive unit 40.

In addition, the control unit 100 can determine whether the user is standing up based on the load signals from the back drive unit 32 and the head drive unit 36.

The control unit 100 can also identify changes in the activities of daily living (ADL) based on the load signals from the back drive unit 32 and the height drive unit 40. This allows the control unit 100 to identify the type of abnormality by measuring the time it takes to transition between the user's states (lying, sitting on the edge, standing) and whether or not tools are being used (side rails, assistance bars, etc.) based on the load signals.

The control unit 100 can also identify abnormal sleeping directions based on the load signals from the knee drive unit 34 and the height drive unit 40. Here, an abnormal sleeping direction means that the user's sleeping posture is abnormal, and in particular, the sleeping direction (sleeping direction) is not the correct direction.

When the control unit 100 identifies an abnormal sleeping position as an abnormality, it can further identify that the user is sleeping with the head facing backwards. This enables the control unit 100 to prevent the user from performing back or knee lifting movements when the user is not in the correct sleeping position.

As mentioned above, abnormality here refers to movements or states that are different from normal. Therefore, for the sake of convenience, movements that are not generally considered "abnormal," such as the user simply getting up or changes in daily living activities, are also included in the "abnormal" category.

The control unit 100 can also estimate an abnormality by using the trained model 114. That is, the control unit 100 can not only determine that there is an abnormality in the bed apparatus 10 or the user at the present time, but can also estimate that an abnormality may occur in the future.

The third abnormality determination process executed by the control unit 100 will be described with reference to FIG. 8. First, the control unit 100 acquires a load signal from the load detection unit 130 (step S1302). Next, the control unit 100 selects the drive unit 120 from which the load signal is to be acquired (step S1304).

The control unit 100 may select the drive unit 120 from which the load signal is obtained depending on the abnormality to be determined. As described above, for example, as shown in the table of FIG. 9, the control unit 100 selects the load signal output from the drive unit 120 when determining whether or not there is an abnormality or identifying the type of abnormality. At this time, the control unit 100 may determine whether or not there is an abnormality or identify the type of abnormality by selecting information related to the load output from the load detection unit 130.

The control unit 100 may also acquire all of the load signals in step S1304. In this case, the control unit 100 may input all of the load signals as input values to the trained model 114.

The control unit 100 determines (estimates) that an abnormality has occurred (step S120). The control unit 100 inputs the load-related information acquired in step S1304 into the trained model 114. When the control unit 100 uses the trained model 114 to output that an abnormality has occurred (step S1306; Yes), it determines that an abnormality has occurred (step S1308).

That is, in steps S1306 and S1308, the control unit 100 can determine whether an abnormality is currently occurring or will occur in the future. In addition, by using the trained model 114, the control unit 100 can also identify (estimate) the type of abnormality that will occur and the nature of the abnormality.

3.3 Response Processing
10 is a flow diagram showing an example of a response process executed by the control unit 100. This process is a process for explaining a response when, for example, in adjusting the height of the bed apparatus 10, the control unit 100 lowers the height of the bed apparatus 10.

When the control unit 100 determines that an abnormality has occurred during operation of the drive unit 120 (step S2102; Yes → S2104; Yes), it stops the operation of the drive unit 120 (step S2106). For example, when the height drive unit 40 is executing control to lower the height of the bed apparatus 10, the control unit 100 determines that a pinching motion under the bed is an abnormal motion. In this case, the control unit 100 (height control unit 104) outputs a stop signal to the height drive unit 40 and stops the operation of lowering the height of the bed apparatus 10.

If a return operation is required (step S2108; Yes), the control unit 100 executes an operation to return the drive unit 120 by a predetermined amount (step S2110). For example, the control unit 100 (height control unit 104) outputs a control signal to the height drive unit 40 to raise the height. The control unit 100 may, for example, return the amount of return by a time (e.g., about 1 to 2 seconds), or may perform control such as returning by a predetermined height (e.g., about 3 cm).

The control unit 100 may also change the amount of return during the return operation depending on the change in load. For example, the control unit 100 may reduce the amount of return when the pinched object is a "hard object." The control unit 100 may also increase the amount of return when the pinched object is a "soft object."

Figure 11 is a diagram explaining what operations the control unit 100 performs on the bed device 10. In Figure 11, operations are specified based on the user's posture and body movements depending on whether the user is present/absent (in bed/out of bed).

Here, the user's postures/body movements are specified as standing, lying down, short and long sitting, edge sitting, rolling over, getting up, shifting from edge sitting to standing. In addition, for each, the actions when the object sandwiched is "hard" and "soft" are specified.

Here, "hard" and "soft" are expressed in a relative sense when referring to "hardness." For example, an object that exceeds a certain threshold is called "hard," and an object that is below that threshold is called "soft." For example, objects such as metal and plastic are "hard," while objects such as cushions and mattresses are "soft."

The threshold value may be preset or may be changed by settings. For example, the sensitivity may be changed to switch between "hard" and "soft."

As a result, "go back a little" and "go back a lot" are expressed relatively. For example, "go back a little" and "go back a lot" indicate the amount of go back (the amount of movement when performing the opposite movement of the current drive unit).

For example, if the amount of return for "returning a little" is A and the amount of return for "returning a lot" is B, then A<B. Here, A and B may be set by time or by the amount of change in the load value. Furthermore, it is sufficient that A and B result in a "returning a little" or "returning a lot" state. Therefore, the control unit 100 may simply calculate the amount of return.

As an example, when the user is in bed and in a recumbent position, the control unit 100 determines that an entrapment action has occurred. At this time, if the control unit 100 determines that the object trapped is a hard object, it performs control to return the height of the bed apparatus 10 slightly. On the other hand, if the control unit 100 determines that the object trapped is a soft object, it performs control to return the height of the bed apparatus 10 a greater amount.

The control unit 100 may determine whether the object is "hard" or "soft" based on the stop time. Figure 12 is a graph that shows a schematic diagram of the load change when the object is pinched. The vertical axis shows the load value, and the horizontal axis shows the time.

Figure 12(a) is a graph showing the case where a hard object is sandwiched, and Figure 12(b) is a graph showing the case where a soft object is sandwiched. At this time, the control unit 100 acquires the time W [N] during which a certain load is applied for each case.

In FIG. 12(a), the time during which a constant load W is applied is T1 (seconds), and in FIG. 12(b), the time during which a constant load W is applied is T3 (seconds). Here, T3>T1, and it can be seen that a hard object and a soft object are sandwiched between each other.

In this case, it is preferable for the control unit 100 to reverse the operation for the time it took to determine the entrapment. For example, in FIG. 12(a), the control unit 100 performs the operation for time T1, which is the opposite of the current operation. For example, if the control unit 100 is performing the operation to lower the bed apparatus 10, it will perform the operation to raise the bed apparatus 10. Also, in FIG. 12(b), since the bed apparatus 10 has been lowered for time T3, it is preferable for the control unit 100 to perform the operation to raise the bed apparatus 10 for the same time T3. In this way, it is preferable for the control unit 100 to perform the opposite operation to the current operation on the drive unit 120 for the time it was determined that entrapment occurred.

The control unit 100 may also determine whether the object is hard or soft by comparing the time it takes to determine whether the object is pinched with a threshold value. For example, if it takes one second from hitting the object to stopping, the control unit 100 may determine that the object is "hard." If it takes three seconds from hitting the object to stopping, the control unit 100 may determine that the object is "soft."

The control unit 100 may also perform control to stop the return movement depending on the degree of danger when pinching is detected. For example, if a patient on the bed is about to stand up when pinching occurs under the bed, the bed movement may be stopped because a return movement may lead to a risk of falling.

As such, in FIG. 12, the control unit 100 causes the drive unit 120 to perform the opposite operation for the time T1 or T3 when it detects that something is trapped. However, if the control unit 100 determines that something hard is trapped, it may set it to "return less" and perform the opposite operation of the amount of operation for a lesser return (e.g., 1 second), and if it determines that something soft is trapped, it may set it to "return more" and perform the opposite operation of the amount of operation for a moreer return (e.g., 3 seconds).

[4. Operational Example]
Next, a specific example of load change will be described with reference to the drawings. Fig. 13 is a graph showing each load change and the average. For example, the graph in Fig. 13 shows load on the vertical axis and time on the horizontal axis.

And it is a graph showing the load applied to the actuator (drive unit 120) when the height of the bed apparatus 10 is lowered from the maximum height to the minimum height. This graph shows that two actuators are used, one on the head side and one on the foot side, to control the height of the bed apparatus 10.

Graph HR is a graph showing the load value (load signal) on the head side. Graph HA is a graph showing the average load value on the head side. Graph FR is a graph showing the load value (load signal) on the foot side. Graph FA is a graph showing the average load value on the foot side (1 second average).

As shown in the graph in Figure 13(a), the load on the head side and the foot side are not the same. In addition, as the drive unit 120 (actuator) operates, the value measured (output) by the load sensor built into the actuator fluctuates. This is because the value fluctuates up and down and fluctuates in accordance with, for example, the rotation period of the screw shaft of the actuator. For this reason, in this figure, the value averaged over a specified time interval is output.

Note that the respective load values change depending on the height of the bed device 10. Here, when the bed device 10 starts to operate (when the actuator starts to operate), the load value on the head side decreases, but the load value on the foot side increases. Here, whether the load value increases or decreases at the start of operation varies depending on the position at which the screw shaft stops. For example, depending on the position of the screw shaft, the load value may increase on both the head side and the foot side, or decrease on both sides. In such a case, the control unit 100 may set a mask time for the load value detected at the start of operation to avoid erroneous detection of an abnormality.

In this way, when the drive unit 120 (e.g., actuator) starts to operate, the load increases or decreases significantly. When determining whether there is an abnormality, the control unit 100 must not detect entrapment when the drive unit 120 starts to operate, otherwise there may be an increase in false detections. Therefore, when determining whether there is an abnormality, the control unit 100 may set a mask time (preferably about 0.2 seconds to 1 second) and not determine whether there is an abnormality during the mask time. In other words, when the drive unit 120 starts to operate, it may not be possible to determine whether a foreign object has actually been entrapped, and in such cases, setting a mask time has the effect of reducing false detections by the control unit 100.

The mask time may be set by the user or staff at will, or the user may be able to set whether or not to use the mask time. Also, if other load detection methods are used (for example, if a four-point load cell method is used), the mask time may not be used.

Figure 13(b) is a graph showing the load change when the bed device 10 pinches an object while descending, determines that an object has been pinched, and then performs an inversion operation. As with Figure 13(a), the vertical axis shows the load value and the horizontal axis shows the time.

In FIG. 13(b), pinching starts at time t10. The control unit 100 determines that pinching has occurred at time t12. Then, the reversal operation starts at time t14.

The dashed line graph HA2 shows the load change (average) of the load value (load signal) on the head side when there is no entrapment. In FIG. 13(b), there is entrapment, so there is no load change as shown in graph HA2, and there is a transition as shown in graph HA.

In FIG. 13(b), an abnormality is determined by the second abnormality determination process. That is, the control unit 100 determines an abnormality by using the difference between the load value and the average. In addition, the control unit 100 determines an abnormality by using, for example, a change in load on the head side, but may also determine an abnormality by using a change in load on the foot side.

5. Other embodiments
In addition to the above-described embodiment, the following embodiment may be envisioned, for example. The following embodiment can be realized in combination with the operations of the above-described embodiment.

(1) If the user stands up, turns violently, or puts their feet on the floor while the bed device 10 is in operation (e.g., while the drive unit 120 is in operation), the control unit 100 may detect a decrease in the load. As a result, the control unit 100 may erroneously determine that entrapment has occurred.

In order to reduce the frequency of the control unit 100 misjudging whether something is pinched, one parameter of the state of the bed apparatus 10 is added to the conditions. For example, when the height of the bed apparatus 10 is to be lowered, a predetermined height is added as a condition. Then, when the height of the bed apparatus 10 is equal to or higher than the predetermined height, the control unit 100 does not judge that an abnormality has occurred even if it judges that something is pinched. This is because, for example, something gets pinched under the bed apparatus 10 when the bed apparatus 10 is low, so the purpose of this is to avoid misjudgments by not judging that an abnormality has occurred when the bed apparatus 10 is clearly high (for example, 40 cm or higher).

(2) Control by Operation Mode Control is performed according to the operation mode of the bed apparatus. For example, one of the operation modes of the bed apparatus is the CPR operation mode. The CPR operation mode is an operation with a high degree of urgency, and when the control unit 100 enters the CPR operation mode, it quickly lowers the back bottom and the knee bottom to make the bottom flat. In addition, by lowering the height of the bed apparatus 10, it becomes possible for medical personnel to take a position that makes it easier to perform cardiopulmonary resuscitation on the user. If the control unit 100 determines an abnormality (for example, detects an object being pinched) during the execution of this operation mode, if it performs a reversal operation, the medical personnel will not be able to take the position that they originally wanted. Therefore, depending on the operation mode, the control unit 100 does not perform a reversal operation even if it determines that an abnormality is pinched.

(3) Changing parameters according to the user The control unit 100 may change parameters according to the user. For example, if the user is heavy, even a minor movement may cause a large change in load on the control unit 100, leading to a false determination of an abnormality (false detection of entrapment).

In this case, for example, the control unit 100 can avoid this by lengthening the mask time. Therefore, the mask time may be changed depending on the value of the load sensor before the operation. In addition, the control unit 100 may switch the operation for determining an abnormality. For example, a specification may be provided that allows the pinch detection function itself to be switched between ON and OFF. Also, for example, a remote control may display to the user whether the pinch detection function is currently enabled or disabled (ON or OFF).

(4) Operation speed The operation speed of the bed apparatus 10 (the driving speed and change amount of the driving unit 120) may be set to a plurality of stages. For example, the driving speed of the driving unit 120 may be switchable between three stages, namely, a first speed (normal speed), a second speed (high speed), and a third speed (ultra-high speed). For example, the operation speed may be the first speed (e.g., 30 mm/sec) < the second speed (e.g., 40 mm/sec) < the third speed (e.g., 60 mm/sec). In this case, when determining an abnormality, for example, when pinching is detected as an abnormality to be determined, the change in load differs depending on the bed apparatus 10 (the driving speed of the driving unit 120) even if the same pinching occurs. Therefore, a different detection threshold value, mask time, and averaging time may be provided for each driving speed of the driving unit 120 (for example, each speed at which the height of the bed apparatus 10 changes, each back-raising speed, etc.).

(5) Prediction: In order to detect entrapment earlier, the control unit 100 may predict entrapment from the load change speed at the beginning of entrapment and stop the bed operation. The control unit 100 may use any prediction algorithm. That is, the control unit 100 may predict the change in load value from the slope or rate of the amount of change. Then, the control unit 100 may determine whether the predicted load value will lead to abnormal operation.

In addition, any conventional predictable calculation formula or algorithm may be used to make the prediction. For example, if the predicted value is Y, the load reduction amount at the start of the prediction is T, the added value is U, the elapsed time is dt, and the parameters are A and B, the prediction may be made from the approximation formulas Y = T + U, U = A x ln(dt) + B.

[6. Modifications]
Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included in the scope of the claims.

In addition, in the embodiments, the programs that run on each device are programs that control the CPU and the like (programs that make the computer function) so as to realize the functions of the above-described embodiments. Information handled by these devices is temporarily stored in a temporary storage device (e.g., RAM) during processing, and is then stored in various ROM, HDD, and SSD storage devices, and is read, modified, and written by the CPU as necessary.

When distributing the program on the market, the program can be stored on a portable recording medium and distributed, or transferred to a server computer connected via a network such as the Internet. In this case, the storage device of the server computer is of course included in the present invention.

Then, by installing an application capable of implementing the functions realized by the control unit 100 on the smartphone, it becomes possible to realize the system of the above-mentioned embodiment on the smartphone.

In addition, generally, the bed frame structure is such that no load is placed on the actuator when the bed device is at its minimum bed height or minimum angle. Since the bed device 10 can change the bed height and back angle electrically, it is common to define a maximum user weight and use it within that range. However, since no load is placed on the actuator at the minimum bed height or minimum angle, it can be used by heavier users. In the case of such a bed device, when pinching is detected near the minimum bed height, it is not possible to determine whether the pinching has actually occurred or whether the load has been released normally due to the bed frame structure. Therefore, a separate contact sensor or the like may be provided to detect whether the bed device 10 has reached the minimum bed height.

In addition, in the above-described embodiment, the pinching is described as being performed in the downward direction for the bed device 10, but it may also be performed in the upward direction. For example, by increasing the height of the bed device 10, it is possible to pinch an object between a table, beam, furniture, etc. In this case, the load increases, but this can be achieved in the same manner as in the above-described embodiment.

REFERENCE SIGNS LIST 1 Bed system 10 Bed device 100 Control unit 102 Bottom control unit 104 Height control unit 106 False detection determination unit 110 Memory unit 112 Load value table 114 Trained model 120 Drive unit 130 Load detection unit 140 Operation unit 150 Display unit 160 Notification unit 170 Communication unit 20 Bottom; 22 Back bottom; 24 Waist bottom; 26 Knee bottom; 28 Foot bottom 32 Back drive unit; 34 Knee drive unit; 36 Head drive unit; 40 Height drive unit 50 Headboard; 52 Side rail; 55 Footboard

Claims (5)

A drive unit that transmits a drive force to a mechanism in the bed body;
An acquisition unit that acquires a load applied to the drive unit;
A determination unit that determines an abnormality based on the acquired load;
A memory unit that stores a trained model generated by machine learning using a type of a drive unit, a displacement of a load, a user's posture, and biometric information of the user,
The determination unit determines an abnormal motion of the user from the load acquired by the acquisition unit and the position of the drive unit acquired by using the trained model.
A bed apparatus characterized by the above. The bed apparatus according to claim 1, characterized in that the trained model is machine-learned to learn abnormal movements of the user, such as sitting up on a bed, climbing over a railing, sitting up on a bed, and a user's pinching movement on a bed apparatus, as learning data for abnormal movements of the user, and is capable of identifying learned abnormal movements. A drive unit that transmits driving force to a mechanism for adjusting the height of the bed body;
An acquisition unit that continuously acquires a load applied to the drive unit;
A determination unit that determines an abnormality based on the acquired load;
A control unit for controlling the mechanism;
Equipped with
The determination unit determines an abnormality in the bed apparatus from the acquired displacement of the load,
The determination unit determines a state in which an object is caught under the frame when the frame of the bed body is lowered by the control unit as an abnormality,
The control unit is
When it is determined that an object is caught under the frame, the lowering of the frame is stopped and the frame is raised by a predetermined height,
Determine the amount to raise the frame based on the object caught
A bed apparatus comprising: The bed apparatus according to claim 3 , wherein the determination unit determines an abnormality in the bed apparatus based on the acquired load and an average of the load. The bed apparatus according to claim 4 , wherein the determination unit determines an abnormality in the bed apparatus from an integral value of a difference between the acquired load and an average of the load.

Images