Intelligent electric nursing bed control method and device based on voice interaction

Abstract

The present invention relates to an intelligent electric nursing bed control method and device based on voice interaction, including the following: obtaining user voice information; performing voice recognition on the voice information to obtain text information; performing semantic analysis on the text information to obtain user intention; determining an initial adjustment plan based on user intention, and determining a final plan through continuous iteration based on user voice feedback when the initial adjustment plan is applied. The present invention realizes interaction with the user by performing semantic recognition on the user's voice information, and can automatically control the electric nursing bed according to the user's intention analyzed by the user's voice information each time, which can facilitate the user to control the electric nursing bed, and is particularly suitable for people who find it difficult to effectively control themselves. And by optimizing the structure of the electric nursing bed, the implementation method of adjusting the back panel tilt angle and the angle adjustment method between the leg board and the calf board are made more convenient, which facilitates the rapid adjustment of relevant angles during voice interaction.

Description

Technical Field

[0001] The present invention relates to the technical field related to electric nursing beds, and in particular to an intelligent electric nursing bed control method and device based on voice interaction. Background Art

[0002] The electric nursing bed is a modern medical equipment that provides a more convenient and safe way of nursing, especially for patients with limited mobility or who need special care.

[0003] At present, electric nursing beds on the market are usually raised and lowered by electric buttons set at the head or foot of the bed, which is convenient for patients to get up, lie down, eat and read, and also convenient for nursing staff to take care of them. This method often requires users to have certain common sense and be able to control the electric buttons by themselves. For people with mobility difficulties such as those after surgery or those who are old and have low education, it is difficult for them to effectively control it without the care of others.

[0004] The current market needs a method and hardware that can realize the control of electric nursing beds through voice interaction to effectively solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to solve at least one of the deficiencies of the prior art and to provide a method and device for controlling an intelligent electric nursing bed based on voice interaction.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] Specifically, a control method of an intelligent electric nursing bed based on voice interaction is proposed, including the following:

[0008] Step 110: Obtain the user's voice information;

[0009] Step 120: Perform voice recognition on the voice information to obtain text information;

[0010] Step 130: Perform semantic analysis on the text information to obtain user intention;

[0011] Step 140: Determine a nursing bed control scheme based on the user's intention, wherein the control scheme includes a backboard fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between the leg board and the calf board;

[0012] Step 150, randomly select a first angle α from [A, B] and a second angle β from [C, D] to control the electric nursing bed to operate accordingly;

[0013] Step 160: After the operation is completed, the user's feedback voice is obtained, and the recognition content is obtained by performing semantic recognition on the feedback voice;

[0014] Step 170: based on the identified content, re-update the angles in [A, B] and [C, D] to control the electric nursing bed to operate accordingly;

[0015] Step 180, repeating steps 160 to 170 until the final first angle α' and the final second angle β' are found, and the control of the electric nursing bed is completed.

[0016] Further, specifically, performing speech recognition on the speech information to obtain text information includes:

[0017] Preprocessing the voice information to perform noise reduction, reverberation removal, and voice power-off detection to obtain preprocessed voice information;

[0018] Performing feature extraction on the preprocessed speech information to obtain extracted features;

[0019] The extracted features are converted through a pre-trained acoustic model to obtain the text information.

[0020] Further, specifically, semantic analysis is performed on the text information to obtain the user intention, including:

[0021] Decomposing the user intention into individual words and determining the part of speech of each word, wherein the part of speech includes noun, verb and adjective;

[0022] Constructing a grammatical structure of text information, wherein the grammatical structure includes a phrase structure grammar tree and a dependency grammar tree;

[0023] The semantic relationship of text information is represented as dependency relationship, and the text information of dependency relationship is detected through syntactic semantic model to obtain word vector and sentence vector;

[0024] Based on AllenNLP, semantic analysis is performed on the word vector and sentence vector to obtain the user intention.

[0025] Further, specifically, determining a nursing bed control scheme based on user intention includes:

[0026] Multiple adjustment gears are preset, each adjustment corresponds to a back panel fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between the leg panel and the calf panel;

[0027] A mapping relationship between user intention and adjustment gear position is established by building an SVM classifier;

[0028] When the user's intention is obtained, the back panel fuzzy tilt angle range [A, B] and the second angle fuzzy range [C, D] between the leg panel and the calf panel are determined by calling the mapping relationship.

[0029] Further, specifically, the process of finding the final angles α′ and β′ includes:

[0030] The feedback speech is semantically recognized to obtain the recognized content. If the user intention of the recognized content is suitable for control, the initial first angle α and the initial second angle β are directly used as the final angle α′ and the final angle β′. If the user intention of the recognized content is not suitable for control,

[0031] Then continue to obtain the adjustment trend in the user's intention. If it is a positive trend, randomly update the angle from (α, B] and (β, D). If it is a negative trend, randomly update the angle from [A, α) and [C, β). If it is a fuzzy trend, randomly update the angle from [α-P, α)∩(α, α+P] and [β-Q, β)∩(β, β+Q]. P and Q are adjustment parameters. It is necessary to ensure that the adjusted angle does not exceed the original range, that is, to ensure that α-P>A, α+P<B; β-Q>C, β+Q<D.

[0032] The above steps are repeated until the final first angle α′ and the final second angle β′ are found, thereby completing the control of the electric nursing bed.

[0033] Further, specifically, the method for adjusting the inclination angle of the back panel and the method for adjusting the angle between the leg panel and the calf panel include:

[0034] The related components of the electric nursing bed are regarded as multiple segments, which are, from left to right, a back plate 1, a first soft connection component 5, a seat plate 2, a leg plate 3, a second soft connection component 6, and a calf plate 4. A first support rod 7 is provided at the back plate 1, and a first support component 8 is provided at the point f in the middle of the first support rod 7. The k end of the first support rod 7 is fixedly arranged, the g end of the first support component 8 is connected to the second support component 9, and the h end of the second support component 9 is fixedly arranged. The h end of the second support component 9 is also connected to the rotating shaft 10, and the j end of the rotating shaft 10 is connected to the first electric push rod 11, and the m end of the first electric push rod 11 is fixedly arranged;

[0035] The leg plate 3 is connected to a second support rod 12, the n-end of the second support rod 12 is connected to a second electric push rod 13, and the p-end of the second electric push rod 13 is fixedly arranged;

[0036] The stroke of the first electric push rod 11 increases, pushing the rotating shaft 10 and the second support assembly 9 to rotate around h. The second support assembly 9 and the first support assembly 8 are connected together at g, forcing the first support assembly 8 to push the back panel 1 to rotate around k, thereby adjusting the inclination angle of the back panel; and the stroke of the second electric push rod 13 decreases, and under the connection at n, it pulls the second support rod 12 to move toward p, lifting the leg board 3, thereby adjusting the angle between the leg board and the calf board.

[0037] The present invention also proposes a device for controlling an intelligent electric nursing bed based on voice interaction, comprising the following:

[0038] A voice information acquisition module is used to acquire the user's voice information;

[0039] A speech recognition module, used for performing speech recognition on the speech information to obtain text information;

[0040] A first semantic recognition module, used to perform semantic analysis on the text information to obtain user intention;

[0041] A control scheme determination module, used to determine a nursing bed control scheme based on a user's intention, wherein the control scheme includes a backboard fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between a leg board and a calf board;

[0042] An angle determination module, used to randomly select a first angle α from [A, B] and a second angle β from [C, D] to control the electric nursing bed to operate accordingly;

[0043] The second semantic recognition module is used to obtain the user's feedback voice after the operation is completed, and obtain the recognition content by performing semantic recognition on the feedback voice;

[0044] An angle updating module, used to re-update the angles in [A, B] and [C, D] based on the identified content to control the electric nursing bed to perform corresponding operation;

[0045] The final angle determination module is used to repeatedly run the second semantic recognition module and the angle update module until the final first angle α′ and the final second angle β′ are found to complete the control of the electric nursing bed.

[0046] The beneficial effects of the present invention are:

[0047] The present invention proposes an intelligent electric nursing bed control method and device based on voice interaction, which can realize interaction with the user by semantically recognizing the user's voice information, and can automatically control the electric nursing bed according to the user's intention analyzed by the user's voice information each time, which can facilitate the user to control the electric nursing bed, and is particularly suitable for people who find it difficult to effectively control themselves. And by optimizing the structure of the electric nursing bed, the implementation method of adjusting the backboard tilt angle and the angle adjustment method between the leg board and the calf board are made more convenient, which facilitates the rapid and accurate adjustment of the relevant angles during voice interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The above and other features of the present disclosure will become more apparent by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present disclosure represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:

[0049] Figure 1 Shown is a flow chart of a method for controlling an intelligent electric nursing bed based on voice interaction according to the present invention;

[0050] Figure 2 The figure shows an example of the hardware structure of the electric nursing bed used in the present invention;

[0051] Figure 3 The figure shows a schematic diagram of the hardware structure of the electric nursing bed used in the present invention, in which the bed panel is in a flat state;

[0052] Figure 4 The figure shows a schematic diagram of the hardware structure of the electric nursing bed used in the present invention with the back plate folded up;

[0053] Figure 5 The figure shows a schematic diagram of the hardware structure of the electric nursing bed used in the present invention with the back plate folded up;

[0054] Figure 6 It is a schematic diagram showing a preferred embodiment of the hardware structure of the electric nursing bed used in the present invention, in which two soft connection components are used;

[0055] Figure 7 The figure shows the structural principle diagram of the sitting up and knee bending principle involved in the present invention in a lying state;

[0056] Figure 8 The figure shows the structural principle diagram of the sitting-up and knee-bending principle involved in the present invention in the sitting-up state;

[0057] Fig. 9 It is a schematic diagram showing a bed board in a lying state and a sitting state in an example of determining the electric push rod parameters of the back board of the present invention;

[0058] Fig.10 The figure shows a schematic diagram of the leg board and the calf board in a lying state and a sitting state in an example of determining the electric push rod parameters of the back board of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other without conflict. The same reference numerals used throughout the drawings indicate the same or similar parts.

[0060] Example 1, reference Figure 1 The present invention proposes a control method for an intelligent electric nursing bed based on voice interaction, comprising the following steps:

[0061] Step 110: Obtain the user's voice information;

[0062] Step 120: Perform voice recognition on the voice information to obtain text information;

[0063] Step 130: Perform semantic analysis on the text information to obtain user intention;

[0064] Step 140: Determine a nursing bed control scheme based on the user's intention, wherein the control scheme includes a backboard fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between the leg board and the calf board;

[0065] Step 150, randomly select a first angle α from [A, B] and a second angle β from [C, D] to control the electric nursing bed to operate accordingly;

[0066] Step 160: After the operation is completed, the user's feedback voice is obtained, and the recognition content is obtained by performing semantic recognition on the feedback voice;

[0067] Step 170: based on the identified content, re-update the angles in [A, B] and [C, D] to control the electric nursing bed to operate accordingly;

[0068] Step 180, repeating steps 160 to 170 until the final first angle α' and the final second angle β' are found, and the control of the electric nursing bed is completed.

[0069] In this preferred embodiment, by semantically recognizing the user's voice information and thus interacting with the user, the electric nursing bed can be automatically controlled according to the user's intention analyzed by the user's voice information each time, which can facilitate the user to control the electric nursing bed, especially for people who find it difficult to effectively control themselves. And by optimizing the structure of the electric nursing bed, the implementation method of adjusting the backboard tilt angle and the angle adjustment method between the leg board and the calf board are made more convenient, which facilitates the rapid adjustment of relevant angles during voice interaction.

[0070] As a preferred real-time method of the present invention, specifically, performing speech recognition on the speech information to obtain text information includes:

[0071] Preprocessing the voice information to perform noise reduction, reverberation removal, and voice power-off detection to obtain preprocessed voice information;

[0072] Performing feature extraction on the preprocessed speech information to obtain extracted features;

[0073] The extracted features are converted through a pre-trained acoustic model to obtain the text information.

[0074] In this preferred embodiment, the acquired user's voice information is preprocessed to remove noise, and then a mature acoustic model is used to recognize the extracted features to complete the recognition of the user's voice information to obtain text information.

[0075] As a preferred real-time method of the present invention, specifically, semantic analysis is performed on the text information to obtain the user intention, including:

[0076] Decomposing the user intention into individual words and determining the part of speech of each word, wherein the part of speech includes noun, verb and adjective;

[0077] Constructing a grammatical structure of text information, wherein the grammatical structure includes a phrase structure grammar tree and a dependency grammar tree;

[0078] The semantic relationship of text information is represented as dependency relationship, and the text information of dependency relationship is detected through syntactic semantic model to obtain word vector and sentence vector;

[0079] Based on AllenNLP, semantic analysis is performed on the word vector and sentence vector to obtain the user intention.

[0080] In this preferred embodiment, semantic recognition of the user's speech text is performed in the above manner, and accurate user intention can be obtained based on the mature AllenNLP model.

[0081] As a preferred real-time method of the present invention, specifically, determining a nursing bed control scheme based on user intention includes:

[0082] Multiple adjustment gears are preset, each adjustment corresponds to a back panel fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between the leg panel and the calf panel;

[0083] A mapping relationship between user intention and adjustment gear position is established by building an SVM classifier;

[0084] When the user's intention is obtained, the back panel fuzzy tilt angle range [A, B] and the second angle fuzzy range [C, D] between the leg panel and the calf panel are determined by calling the mapping relationship.

[0085] In the preferred embodiment, an SVM classifier is established by associating multiple back panel fuzzy tilt angle ranges [A, B] and the second angle fuzzy range [C, D] between the leg panel and the calf panel with different SVM classifiers, each corresponding to a vocabulary group of user intention, that is, when the identified user intention falls into any vocabulary group of user intention, the back panel fuzzy tilt angle range [A, B] and the second angle fuzzy range [C, D] between the leg panel and the calf panel are directly determined quickly based on the mapping relationship.

[0086] As a preferred real-time method of the present invention, specifically, the process of finding the final angles α′ and β′ includes:

[0087] The feedback speech is semantically recognized to obtain the recognized content. If the user intention of the recognized content is suitable for control, the initial first angle α and the initial second angle β are directly used as the final angle α′ and the final angle β′. If the user intention of the recognized content is not suitable for control,

[0088] Then continue to obtain the adjustment trend in the user's intention. If it is a positive trend, randomly update the angle from (α, B] and (β, D). If it is a negative trend, randomly update the angle from [A, α) and [C, β). If it is a fuzzy trend, randomly update the angle from [α-P, α)∩(α, α+P] and [β-Q, β)∩(β, β+Q]. P and Q are adjustment parameters. It is necessary to ensure that the adjusted angle does not exceed the original range, that is, to ensure that α-P>A, α+P<B; β-Q>C, β+Q<D.

[0089] The above steps are repeated until the final first angle α′ and the final second angle β′ are found, thereby completing the control of the electric nursing bed.

[0090] In this preferred embodiment, the user's intention is obtained by semantically recognizing the user's feedback voice after each updated adjustment of the first angle and the second angle, and the first angle and the second angle are continuously iterated according to the user's intention, that is, fine-tuned, to finally obtain the relevant angles that satisfy the user, thereby completing the regulation of the electric nursing bed.

[0091] Among them, judging whether it is a control appropriateness situation based on user intention can be done through pre-training, that is, if the user intention is a positive word such as "appropriate" or "in place", it means a control appropriateness situation; if the user intention is a positive trend, it is a word of the nature of "too low" or "not high enough"; if the user intention is a negative trend, it is a word of the nature of "too high" or "too high"; if the user intention is a fuzzy trend, it is "not sure" or "not suitable" and other related content without a clear concept but more hesitant. The specific identification method can be determined by the similarity of word vectors to determine the trend of the user intention identified by the AllenNLP model.

[0092] Reference Figure 6 , Figure 7 as well as Figure 8 As a preferred real-time method of the present invention, specifically, the method for adjusting the backboard inclination angle and the method for adjusting the angle between the leg board and the calf board include:

[0093] The related components of the electric nursing bed are regarded as multiple segments, which are, from left to right, a back plate 1, a first soft connection component 5, a seat plate 2, a leg plate 3, a second soft connection component 6, and a calf plate 4. A first support rod 7 is provided at the back plate 1, and a first support component 8 is provided at the point f in the middle of the first support rod 7. The k end of the first support rod 7 is fixedly arranged, the g end of the first support component 8 is connected to the second support component 9, and the h end of the second support component 9 is fixedly arranged. The h end of the second support component 9 is also connected to the rotating shaft 10, and the j end of the rotating shaft 10 is connected to the first electric push rod 11, and the m end of the first electric push rod 11 is fixedly arranged;

[0094] The leg plate 3 is connected to a second support rod 12, the n-end of the second support rod 12 is connected to a second electric push rod 13, and the p-end of the second electric push rod 13 is fixedly arranged;

[0095] The stroke of the first electric push rod 11 increases, pushing the rotating shaft 10 and the second support assembly 9 to rotate around h. The second support assembly 9 and the first support assembly 8 are connected together at g, forcing the first support assembly 8 to push the back panel 1 to rotate around k, thereby adjusting the inclination angle of the back panel; and the stroke of the second electric push rod 13 decreases, and under the connection at n, it pulls the second support rod 12 to move toward p, lifting the leg board 3, thereby adjusting the angle between the leg board and the calf board.

[0096] In this preferred embodiment, two electric push rods are allocated to control the backboard and the leg board and calf board respectively, and the change of the stroke of the connecting rod and the electric push rod is used to achieve sitting up and bending the knees. The advantages are relatively simple structure, easy processing, convenient assembly, precise transmission, and high cost performance. The bed surface is composed of a bed board and a soft connection group, wherein the bed board is 4 pieces, namely 1, 2, 3, and 4, and the soft connection group is 2 groups, namely 5 and 6; the 1st and 2nd boards are connected by the 5th soft connection group, and the 3rd and 4th boards are connected by the 6th soft connection group; when sitting up, the 1st board flips around the Y1 axis, thereby achieving the purpose of sitting up; when bending the knees, the 3rd board flips around the Y2 axis. Since the 3rd board and the 4th board are connected by Y3, when the 3rd board flips, the 4th board is dragged up and rotates relative to the 3rd board around Y3, thereby achieving the purpose of bending the knees.

[0097] For specific application, please refer to Figure 2 , Electric nursing bed: (1) Whole bed lifting mechanism: mainly composed of bottom frame, whole bed lifting motor, motor mounting seat, seven-shaped lever arm, guide rail, short bending frame and short deforming frame.

[0098] Assembly: The bottom frame is connected to the short curved frame, the seven-shaped force arm is connected to the guide rail with bolts, the head of the whole bed lifting motor is connected to the motor mounting seat with bolts, and the motor push rod is connected to the guide rail mounting hole with bolts.

[0099] Function realization: Operate the wired control handle to press the ▲up / ▼down direction buttons of the whole bed lifting, and control the motor push rod of the whole bed lifting motor to make linear push and pull movements through the electric drive control box. The motor push rod extends to drive the guide rail, short curved frame, and bed frame to make linked movements. The whole bed lifting and height adjustment function is realized through the motor push rod travel distance extension or retraction action, and the adjustment range is >250mm.

[0100] Bed lifting motor stroke: 200mm

[0101] (2)Reference Figure 3 as well as Figure 4 , Backboard lifting mechanism: mainly includes a backboard, a backboard fixing block, a support shaft, a transmission block, a long curved frame, a swing arm and a back lifting motor.

[0102] Assembly: The back plate, transmission block and support shaft are connected and installed in the mounting hole of the back plate fixing block. The back support block of the long curved frame is connected to the transmission block with pins. The back lifting motor and motor mounting seat, swing arm and electric push rod are installed with bolts.

[0103] The back panel lifting function is realized by operating the wired control handle to press the ▲up / ▼down direction buttons of the back panel lifting, and through the electric drive control box, the motor push rod of the back panel lifting motor is controlled to make a linear push / pull movement. The motor push rod extends to drive the long bending frame, swing arm, back support block, transmission block, support shaft back panel fixing block, and connects the support back panel to make the support axial rotation movement. The back panel folding angle is realized by extending or retracting the motor push rod travel distance, and the angle adjustment range is 0-70°±5°.

[0104] Back panel lifting motor stroke: 150mm.

[0105] (3)Reference Figure 5 , leg board lifting mechanism: mainly composed of leg lifting motor, swing arm, leg board, leg board support plate, support shaft

[0106] Assembly: The leg lifting motor mounting head is connected to the bed frame motor mounting seat with bolts, and the leg lifting motor push rod mounting hole is connected to the leg board swing arm mounting hole with pins. The leg board and the calf board connection ears are connected with bolts.

[0107] The leg board lifting function is realized by operating the wired control handle to press the ▲up / ▼down direction buttons of the leg board lifting, and through the electric drive control box, the motor push rod of the leg board lifting motor is controlled to make a linear push / pull motion. The motor push rod retracts to drive the swing arm of the leg board support shaft to make an axial movement, and the calf board makes a linked movement. The leg board folding angle is realized by extending or retracting the motor push rod travel distance, and the angle adjustment range is 0-30°±5°.

[0108] Leg lifting motor stroke: 100mm.

[0109] (4) The synchronous function of the backboard and leg board is realized: operate the wired control handle to synchronously lift the backboard and leg board ▲up / ▼down direction buttons, and through the electric drive control box, synchronously control the motor push rods of the backboard and leg board lifting motors to make linear push / pull movements. The motor push rods extend to drive the backboard support shaft, and the motor retracts to drive the swing arm of the leg board support shaft to make axial movements. The synchronous folding angle of the backboard and leg board is realized through the extension or retraction of the motor push rod travel distance.

[0110] In addition, in this preferred embodiment, the above product design description, calculation formula and related calculation method description are

[0111] Product design description:

[0112] According to the medical device industry standard YY 0003-90, combined with our company's actual production situation and actual medical needs, the basic parameters of our company's multifunctional electric nursing bed are determined as follows: the total length of the bed panel is 2000mm, the width is 900mm, the minimum height from the ground is 450mm, the back frame angle range is ≥60°, the leg frame angle range is ≥30°, the foot frame angle range is ≥10°, the head down supine inclination angle is 0~12°, and the bed lifting travel distance is 300mm.

[0113] Calculation formula and explanation of relevant calculation method:

[0114] Selection of soft connection series back frame electric push rod:

[0115] According to the test requirements, the backboard load is 45% of the total weight, the maximum load is 1700N, the lifting angle is 65°, the lifting time is set at 28 (equivalent to t) seconds, and the weight of the bed board assembly itself is about 180N.

[0116] Load T1 = 1700 × 0.45 + 180 = 945 N

[0117] Angular velocity of arm

[0118] Arm speed υ1=ω1×t1=0.0404×0.2845=0.0115m / s

[0119] The power transmitted by the arm is P1 = T1 × υ1 = 945 × 0.0115 = 11.5 m / s

[0120] Permitted power of electric actuator

[0121] Actual power of electric push rod P 实1 =P 许1 ×1.2=31.2×1.2=38W

[0122] Electric linear actuator thrust

[0123] Determination of electric push rod stroke:

[0124] Depend on Fig. 9 It can be seen that the length of the electric push rod is 482.96mm when sitting up and 284.2mm when lying flat.

[0125] Electric push rod stroke L = 482.92-284.2 = 198.72 ≈ 200mm

[0126] Selection of soft connection series leg frame electric push rod:

[0127] According to the test requirements, the leg frame load is 30% of the total weight, the maximum load is 1700N, the lifting angle is 32°, the lifting time is set at 18 seconds, and the weight of the bed board assembly itself is approximately 208N.

[0128] Load T2 = 1700 × 0.3 + 208 = 718N

[0129] Angular velocity of arm

[0130] Arm speed υ2 = ω2 × t2 = 0.032 × 0.205 = 0.0066 m / s

[0131] The power transmitted by the arm is P2 = T2 × υ2 = 718 × 0.0066 = 4.8 m / s

[0132] Permitted power of electric actuator

[0133] Actual power of electric push rod P 实2 =P 许2 ×1.2=13.1×1.2=15.8W

[0134] Electric linear actuator thrust

[0135] Determination of electric push rod stroke:

[0136] Depend on Fig.10 It can be seen that the length of the electric push rod is 381.72mm when lying flat and 282.18mm when raising the legs.

[0137] Electric push rod stroke L = 381.72-282.18 = 99.54 ≈ 100 mm

[0138] Determination of electric push rod stroke:

[0139] The lifting of the medical electric nursing bed is directly changed by the rotating bracket and the height change of the short curved frame, that is, the stroke of the electric push rod is the lifting range of the bed panel. From the operating parameters in Table 1, it can be seen that the stroke of the electric push rod is 150mm.

[0140] In summary:

[0141] Ginseng Back frame electric push rod Leg frame electric push rod Lifting electric push rod push 6000N 6000N 2000N OK 200mm 100mm 150mm

[0142] Table 1

[0143] Implementation of forward and backward tilt

[0144] When the multifunctional electric nursing bed tilts forward or backward, the bed frame will be displaced relative to the bracket on the electric push rod during the tilting process. During the design and development process, the friction is reduced by improving the smoothness of the inner wall of the bracket to ensure smooth displacement during the tilting process.

[0145] The load of the electric nursing bed is 1700N, and the weight of the whole bed is only 800N.

[0146]

[0147] The copper bearing can withstand 10-12MPa, 1-2m / s, so it was decided to use a round steel sliding bearing with an outer diameter of 22 and a width of 40. After the installation test, the effect is good.

[0148] The present invention also proposes a device for controlling an intelligent electric nursing bed based on voice interaction, comprising the following:

[0149] A voice information acquisition module is used to acquire the user's voice information;

[0150] A speech recognition module, used for performing speech recognition on the speech information to obtain text information;

[0151] A first semantic recognition module, used to perform semantic analysis on the text information to obtain user intention;

[0152] A control scheme determination module, used to determine a nursing bed control scheme based on a user's intention, wherein the control scheme includes a backboard fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between a leg board and a calf board;

[0153] An angle determination module, used to randomly select a first angle α from [A, B] and a second angle β from [C, D] to control the electric nursing bed to operate accordingly;

[0154] The second semantic recognition module is used to obtain the user's feedback voice after the operation is completed, and obtain the recognition content by performing semantic recognition on the feedback voice;

[0155] An angle updating module, used to re-update the angles in [A, B] and [C, D] based on the identified content to control the electric nursing bed to perform corresponding operation;

[0156] The final angle determination module is used to repeatedly run the second semantic recognition module and the angle update module until the final first angle α′ and the final second angle β′ are found to complete the control of the electric nursing bed.

[0157] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0158] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or system that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0159] Although the description of the present invention has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims in view of the prior art by reference to the appended claims, thereby effectively covering the intended scope of the present invention. In addition, the above description of the present invention is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that have not yet been foreseen may still represent equivalent changes to the present invention.

[0160] The above is only a preferred embodiment of the present invention. The present invention is not limited to the above implementation. As long as the technical effect of the present invention is achieved by the same means, it should belong to the protection scope of the present invention. Within the protection scope of the present invention, its technical scheme and / or implementation method can have various modifications and changes.

Claims

1. A method for controlling an intelligent electric nursing bed based on voice interaction, characterized in that: Includes the following: Step 110: Obtain the user's voice information; Step 120: Perform voice recognition on the voice information to obtain text information; Step 130: Perform semantic analysis on the text information to obtain user intention; Step 140: Determine a nursing bed control scheme based on the user's intention, wherein the control scheme includes a backboard fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between the leg board and the calf board; Step 150, randomly select a first angle α from [A, B] and a second angle β from [C, D] to control the electric nursing bed to operate accordingly; Step 160: After the operation is completed, the user's feedback voice is obtained, and the recognition content is obtained by performing semantic recognition on the feedback voice; Step 170: based on the identified content, re-update the angles in [A, B] and [C, D] to control the electric nursing bed to operate accordingly; Step 180: Repeat steps 160 to 170 until the final first angle is found. And finally the second angle , complete the control of the electric nursing bed.

2. The intelligent electric nursing bed control method based on voice interaction according to claim 1 is characterized in that: Specifically, performing speech recognition on the speech information to obtain text information includes: Preprocessing the voice information to perform noise reduction, reverberation removal, and voice power-off detection to obtain preprocessed voice information; Performing feature extraction on the preprocessed speech information to obtain extracted features; The extracted features are converted through a pre-trained acoustic model to obtain the text information.

3. The intelligent electric nursing bed control method based on voice interaction according to claim 1 is characterized in that: Specifically, semantic analysis is performed on the text information to obtain the user's intention, including: Decomposing the user intention into individual words and determining the part of speech of each word, wherein the part of speech includes noun, verb and adjective; Constructing a grammatical structure of text information, wherein the grammatical structure includes a phrase structure grammar tree and a dependency grammar tree; The semantic relationship of text information is represented as dependency relationship, and the text information of dependency relationship is detected through syntactic semantic model to obtain word vector and sentence vector; Based on AllenNLP, semantic analysis is performed on the word vector and sentence vector to obtain the user intention.

4. The intelligent electric nursing bed control method based on voice interaction according to claim 1 is characterized in that: Specifically, the nursing bed control scheme is determined based on the user's intention, including: Multiple adjustment gears are preset, each adjustment corresponds to a back panel fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between the leg panel and the calf panel; A mapping relationship between user intention and adjustment gear position is established by building an SVM classifier; When the user's intention is obtained, the back panel fuzzy tilt angle range [A, B] and the second angle fuzzy range [C, D] between the leg panel and the calf panel are determined by calling the mapping relationship.

5. The intelligent electric nursing bed control method based on voice interaction according to claim 1 is characterized in that: Specifically, find the final angle as well as The process includes, The feedback voice is semantically recognized to obtain the recognized content. If the user's intention of the recognized content is to control suitability, the first angle α and the second angle β are directly used as the final angle. And the final angle If the user intent of the identified content is not appropriate for control, Then continue to obtain the adjustment trend in the user's intention. If it is a positive trend, randomly update the angle from (α, B] and (β, D). If it is a negative trend, randomly update the angle from [A, α) and [C, β). If it is a fuzzy trend, randomly update the angle from [α-P, α)∩(α, α+P] and [β-Q, β)∩(β, β+Q]. P and Q are adjustment parameters. It is necessary to ensure that the adjusted angle does not exceed the original range, that is, to ensure that α-P>A, α+P<B; β-Q>C, β+Q<D. Repeat the above until you find the final first angle And finally the second angle , complete the control of the electric nursing bed.

6. The intelligent electric nursing bed control method based on voice interaction according to claim 1 is characterized in that: Specifically, the method for adjusting the inclination angle of the back plate and the method for adjusting the angle between the leg plate and the calf plate include: The relevant components of the electric nursing bed are regarded as a plurality of sections, which are, from left to right, a back plate (1), a first soft connection component (5), a seat plate (2), a leg plate (3), a second soft connection component (6), and a calf plate (4). A first support rod (7) is arranged at the back plate (1), a first support component (8) is arranged at point f in the middle of the first support rod (7), the k end of the first support rod (7) is fixedly arranged, the g end of the first support component (8) is connected to the second support component (9), the h end of the second support component (9) is fixedly arranged, the h end of the second support component (9) is also connected to a rotating shaft (10), the j end of the rotating shaft (10) is connected to a first electric push rod (11), and the m end of the first electric push rod (11) is fixedly arranged; The leg plate (3) is connected to a second support rod (12), the n-end of the second support rod (12) is connected to a second electric push rod (13), and the p-end of the second electric push rod (13) is fixedly arranged; The stroke of the first electric push rod (11) increases, pushing the rotating shaft (10) and the second support assembly (9) to rotate around h. The second support assembly (9) and the first support assembly (8) are connected together at g, forcing the first support assembly (8) to push the back plate (1) to rotate around k, thereby achieving adjustment of the back plate tilt angle; while the stroke of the second electric push rod (13) decreases, and at n connection, pulls the second support rod (12) to move toward p, lifting the leg plate (3), thereby achieving adjustment of the angle between the leg plate and the calf plate.

7. A device for controlling an intelligent electric nursing bed based on voice interaction, characterized in that: Includes the following: A voice information acquisition module is used to acquire the user's voice information; A speech recognition module, used for performing speech recognition on the speech information to obtain text information; A first semantic recognition module, used to perform semantic analysis on the text information to obtain user intention; A control scheme determination module, used to determine a nursing bed control scheme based on a user's intention, wherein the control scheme includes a backboard fuzzy tilt angle range [A, B] and a second angle fuzzy range [C, D] between a leg board and a calf board; An angle determination module, used to randomly select a first angle α from [A, B] and a second angle β from [C, D] to control the electric nursing bed to operate accordingly; The second semantic recognition module is used to obtain the user's feedback voice after the operation is completed, and obtain the recognition content by performing semantic recognition on the feedback voice; An angle updating module, used to re-update the angles in [A, B] and [C, D] based on the identified content to control the electric nursing bed to perform corresponding operation; The final angle determination module is used to repeatedly run the second semantic recognition module and the angle update module until the final first angle is found. And finally the second angle , complete the control of the electric nursing bed.

Images