Medical indoor inspection robot safety evaluation method and system and storage medium

Abstract

The invention discloses a medical indoor inspection robot safety evaluation method and system and a storage medium. Comparing a plurality of preset safety evaluation indexes in pairs, and determining a relative weight between the to-be-compared index and the comparison reference index; establishing a judgment matrix according to the relative weight of each safety evaluation index; summing each row of the matrix obtained by regularizing the judgment matrix to obtain a first matrix; regularizing the first matrix to obtain an index weight vector; according to each safety evaluation index score, determining a membership degree of each safety evaluation index of the to-be-evaluated inspection robot and each preset safety grade, and obtaining a membership degree matrix; and according to the index weight vector and the membership matrix, obtaining safety evaluation data of the to-be-evaluated inspection robot in different dimensions. Standard safety evaluation can be carried out on the patrol and ward-round intelligent robot, and the safety performance of the medical indoor patrol robot can be comprehensively and quantitatively evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent robots, and particularly relates to a safety evaluation method, system and storage medium for a medical indoor patrol robot. Background Art

[0002] With the improvement of AI technology and automation technology, more and more intelligent robots for rounds of inspection and ward rounds have entered practical applications. The significance of using a medical indoor patrol robot lies in improving efficiency, reducing the burden on medical staff, reducing the risk of infectious disease transmission, providing patient companionship and services, and real-time monitoring and data recording. However, there is a lack of a standardized safety evaluation method for intelligent robots for rounds of inspection and ward rounds, and their safety performance cannot be comprehensively and quantitatively evaluated. Summary of the Invention

[0003] In order to solve the above problems, the inventor made the present invention, and through specific embodiments, provides a safety evaluation method, system and storage medium for a medical indoor patrol robot.

[0004] In a first aspect, an embodiment of the present invention provides a safety evaluation method for a medical indoor patrol robot, including the following steps:

[0005] Compare multiple preset safety evaluation indicators pairwise to determine the relative weight between the indicator to be compared and the comparison reference indicator;

[0006] According to the relative weight of each safety evaluation indicator, establish a judgment matrix. Each row in the judgment matrix represents the relative weight of the corresponding safety evaluation indicator to each safety evaluation indicator when the corresponding safety evaluation indicator is the indicator to be compared, and each column represents the relative weight of the corresponding safety evaluation indicator to each safety evaluation indicator when the corresponding safety evaluation indicator is the comparison reference indicator;

[0007] Sum each row of the matrix obtained by normalizing the judgment matrix to obtain a first matrix;

[0008] Normalize the first matrix to obtain an index weight vector;

[0009] Collect the scores of each safety evaluation indicator of the patrol robot to be evaluated, and according to the scores of each safety evaluation indicator, determine the membership degree of each safety evaluation indicator of the patrol robot to be evaluated to each preset safety classification level to obtain a membership degree matrix;

[0010] According to the index weight vector and the membership degree matrix, obtain the safety evaluation data of different dimensions of the patrol robot to be evaluated.

[0011] Specifically, setting safety evaluation indicators includes the following steps:

[0012] Multiple security evaluation indicators are set respectively from the user, the inspection robot, and the user-robot environment, and the multiple security evaluation indicators are classified into different dimensions.

[0013] Specifically, the judgment matrix is

[0014]

[0015] where V = (b ij ) n×n represents the judgment matrix, and b ij represents the relative weight when the i-th security evaluation indicator is the index to be compared and the j-th security evaluation indicator is the comparison reference indicator, and n represents the total number of security evaluation indicators.

[0016] Specifically, the sum of each row of the matrix obtained by normalizing the judgment matrix is calculated to obtain the first matrix, including the following steps:

[0017] According to normalize the judgment matrix, where b ij represents the relative weight when the i-th security evaluation indicator is the index to be compared and the j-th security evaluation indicator is the comparison reference indicator in the judgment matrix, n represents the total number of security evaluation indicators, k represents the row number of the judgment matrix, and aij represents the corresponding value of b ij after the judgment matrix is normalized;

[0018] According to sum each row of the matrix obtained by normalizing the judgment matrix to obtain the first matrix U = [U1, U2,..., U n T , where U i represents the sum of the elements in the i-th row of the matrix obtained by normalizing the judgment matrix, that is, the i-th element of the first matrix.

[0019] Specifically, normalizing the first matrix to obtain the index weight vector includes the following steps:

[0020] According to normalize the first matrix to obtain the index weight vector w = [w1, w2,..., w n T , where U i represents the i-th element of the first matrix, U j represents the j-th element of the first matrix, n represents the total number of security evaluation indicators, and W i represents the i-th element of the index weight vector.

[0021] Specifically, establishing a judgment matrix according to the relative weight of each security evaluation indicator also includes the following steps:

[0022] By​​ Perform a consistency test on the judgment matrix, where CR represents the consistency ratio; RI represents the average random consistency index; CI represents the consistency index; when CR is less than 0.1, it indicates that the consistency test is passed; when CR is greater than or equal to 0.1, readjust and modify the judgment matrix until the judgment matrix passes the consistency test;

[0023] Among them n represents the total number of safety evaluation indicators, (Vw) i represents the sum of the elements in the i-th row after the normalization process of the judgment matrix, and W i represents the i-th element of the index weight vector.

[0024] Specifically, according to the index weight vector and the membership degree matrix, obtain the safety evaluation data of the to-be-evaluated inspection robot in different dimensions, which also includes the following steps:

[0025] Map each element in the membership degree matrix to between 0 and 1;

[0026] According to the dimensions to which the safety evaluation indicators belong, divide the index weight vector and the membership degree matrix to obtain the index weight vector components and membership degree matrix components corresponding to each dimension, and multiply the index weight vector components and membership degree matrix components of the same dimension to obtain the safety evaluation data of the corresponding dimension.

[0027] Specifically, mapping each element in the membership degree matrix to between 0 and 1 also includes the following steps:

[0028] When the safety evaluation indicator corresponding to the element in the membership degree matrix is positively correlated with safety, update the corresponding element according to r ij ′ = a′ ij / maxa′ ij ,

[0029] When the safety evaluation indicator corresponding to the element in the membership degree matrix is negatively correlated with safety, update the corresponding element according to r ij ′ = mina′ ij / a ij ′,

[0030] Among them, r ij ′ represents the updated value of the element in the i-th row and j-th column a′ ij in the membership degree matrix, and maxa′ ij represents the maximum value among all membership degrees of the same safety evaluation indicator of the same to-be-evaluated inspection robot for the same preset safety classification corresponding to a′ ij , and mina′ ij represents a′ ijThe minimum value among all the membership degrees of the same safety evaluation indicators corresponding to the same inspection robot to be evaluated with respect to the same preset safety grading levels.

[0031] In a second aspect, an embodiment of the present invention provides a safety evaluation system for a medical indoor inspection robot, including:

[0032] An index weight vector determination module, configured to compare multiple preset safety evaluation indicators pairwise to determine the relative weights between the indicators to be compared and the comparison reference indicators; establish a judgment matrix according to the relative weights of each safety evaluation indicator, where each row in the judgment matrix respectively represents the relative weights of the corresponding safety evaluation indicator and each safety evaluation indicator when the corresponding safety evaluation indicator is the indicator to be compared, and each column respectively represents the relative weights of the corresponding safety evaluation indicator and each safety evaluation indicator when the corresponding safety evaluation indicator is the comparison reference indicator; sum the elements of each row of the matrix obtained by normalizing the judgment matrix to obtain a first matrix; normalize the first matrix to obtain the index weight vector;

[0033] A membership degree matrix determination module, configured to collect the scores of each safety evaluation indicator of the inspection robot to be evaluated, and determine the membership degrees of each safety evaluation indicator of the inspection robot to be evaluated with respect to each preset safety grading level according to the scores of each safety evaluation indicator, to obtain a membership degree matrix;

[0034] A safety evaluation data determination module, configured to obtain the safety evaluation data of different dimensions of the inspection robot to be evaluated according to the index weight vector and the membership degree matrix.

[0035] Based on the same inventive concept, an embodiment of the present invention provides a computer storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed, the foregoing safety evaluation method for a medical indoor inspection robot is implemented.

[0036] The beneficial effects of the above technical solutions provided by the embodiments of the present invention at least include:

[0037] It can perform a standardized safety evaluation on the intelligent robot for rounds of inspections and ward rounds, and can comprehensively and quantitatively evaluate the safety performance of the medical indoor inspection robot.

[0038] Other features and advantages of the present invention will be described in subsequent specifications, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in the written specifications, claims, and drawings.

[0039] Next, through the drawings and embodiments, the technical solutions of the present invention will be further described in detail. Description of the Drawings

[0040] The accompanying drawings are used to provide a further understanding of the present invention and form a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the accompanying drawings:

[0041] Figure 1 It is a flowchart of a safety evaluation method for a medical indoor patrol robot in an embodiment of the present invention. Specific embodiments

[0042] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0043] To solve the problems existing in the prior art, embodiments of the present invention provide a safety evaluation method, system and storage medium for a medical indoor patrol robot.

[0044] Embodiments of the present invention provide a safety evaluation method for a medical indoor patrol robot, and its process is as Figure 1 shown, including the following steps:

[0045] Step S1: Compare multiple preset safety evaluation indicators pairwise to determine the relative weights between the indicators to be compared and the comparison reference indicators; establish a judgment matrix according to the relative weights of each safety evaluation indicator. Each row in the judgment matrix represents the relative weights of the corresponding safety evaluation indicator and each safety evaluation indicator when the corresponding safety evaluation indicator is the indicator to be compared, and each column represents the relative weights of the corresponding safety evaluation indicator and each safety evaluation indicator when the corresponding safety evaluation indicator is the comparison reference indicator; sum each row of the matrix obtained by normalizing the judgment matrix to obtain a first matrix; normalize the first matrix to obtain an index weight vector.

[0046] Set safety evaluation indicators, including the following steps: Set multiple safety evaluation indicators respectively from the aspects of users, patrol robots, and the association between users and the robot environment, and classify the multiple safety evaluation indicators into different dimensions. For example, according to the three perspectives of humans, machines, and human-machine interaction, preliminarily formulate a human-machine collaboration safety evaluation index system for medical indoor patrol robots.

[0047]

[0048]

[0049] From the perspectives of human, machine, and human-machine interaction, the human-machine collaboration safety evaluation indicators of medical indoor patrol robots are classified into four dimensions: namely, operation convenience, the inherent safety of medical indoor patrol robots, appearance safety, motion safety, and the safety evaluation indicators under each dimension. Thus, the human-machine collaboration safety evaluation indicators of medical indoor patrol robots are preliminarily formulated.

[0050]

[0051] Use the analytic hierarchy process to construct the hierarchical model of its human-machine collaboration safety evaluation index system.

[0052] The judgment matrix is

[0053]

[0054] Among them, V=(b ij ) n×n represents the judgment matrix, b ij represents the relative weight when the i-th safety evaluation index is the index to be compared and the j-th safety evaluation index is the comparison reference index, and n represents the total number of safety evaluation indicators. For example, according to the Saaty 1-9 scale method for calibrating the relative weight, the relative weight is a quantitative data of the importance degree of one index to another. When comparing index A and index B, if the relative weight is 3, then when comparing index B and index A, the relative weight is 1 / 3.

[0055] Sum the elements of each row of the matrix obtained by normalizing the judgment matrix to obtain the first matrix, including the following steps:

[0056] According to normalize the judgment matrix, b ij represents the relative weight when the i-th safety evaluation index is the index to be compared and the j-th safety evaluation index is the comparison reference index in the judgment matrix, n represents the total number of safety evaluation indicators, k represents the row number of the judgment matrix, and a ij represents the corresponding value of b ij after normalizing the judgment matrix;

[0057] According to sum the elements of each row of the matrix obtained by normalizing the judgment matrix to obtain the first matrix U = [U1, U2,..., U n T , U i represents the sum of the elements of the i-th row of the matrix obtained by normalizing the judgment matrix, that is, the i-th element of the first matrix.

[0058] Normalize the first matrix to obtain the index weight vector, including the following steps:

[0059] According to​ Regularize the first matrix to obtain the index weight vector w = [w1, w2,..., w n T , where U i represents the i-th element of the first matrix, and U j represents the j-th element of the first matrix. n represents the total number of safety evaluation indicators, and W i represents the i-th element of the index weight vector.

[0060] According to the relative weights of each safety evaluation indicator, establish a judgment matrix, which also includes the following steps:

[0061] Through Conduct a consistency test on the judgment matrix, where CR represents the consistency ratio; RI represents the average random consistency index; CI represents the consistency index; when CR is less than 0.1, it means passing the consistency test; when CR is greater than or equal to 0.1, readjust and modify the judgment matrix until the judgment matrix passes the consistency test; the value of RI changes with the total number of indicators n, as shown in the following table:

[0062]

[0063]

[0064] where n represents the total number of safety evaluation indicators, (Vw) i represents the sum of the elements in the i-th row after normalizing the judgment matrix, and w i represents the i-th element of the index weight vector.

[0065]

[0066]

[0067] Step S2: Collect the scores of each safety evaluation indicator of the inspection robot to be evaluated. According to the scores of each safety evaluation indicator, determine the membership degrees of each safety evaluation indicator of the inspection robot to be evaluated to each preset safety classification, and obtain the membership matrix. For example, adopt the method of scoring the indicators from 0 to 100 to establish a relative membership matrix, and use this to compare the inspection robots, so as to select a better inspection robot. It is divided into five grades: unqualified, poor, medium, good, and excellent, and the interval is divided from 0 to 100 to strengthen the sense of scoring boundaries in use, as shown in the following table.

[0068] Evaluation description Unqualified Poor Medium Good Excellent Score range [0，60) [60，70) [70，80) [80，90) [90，100]

[0069] ​Three types of medical indoor patrol robots are selected, namely humanoid robots, small robot vehicles, and interactive robots. In actual evaluation applications, corresponding adjustments were made according to the actual situations of these three types of medical indoor patrol robots, that is, indicators not involving the design of monitors, controllers, and protective devices were not evaluated. Finally, the proposed scheme of the present invention was used to construct a human-machine collaboration safety evaluation index system for medical indoor patrol robots. Calculate the index weights and round them according to the Saaty 1-9 scale method. The lowest average score of the index importance is 3.80 points, with a full score of 5 points. Therefore, based on 3.74 points, the range of 3.74 - 5 points is evenly divided into 9 intervals from 0 - 9, and the importance score of each interval is the value of that interval, as shown in the following table.

[0070] Division range (3.74，3.88] (4.02，4.16] (4.30，4.44] (4.58，4.72] (4.86，5.00] Rounding in scale method 1 3 5 7 9

[0071] According to the above method, the index weights of each level of indicators are obtained, as shown in the following table. And according to the formula, each indicator passes the consistency test.

[0072] Index weights of dimensions B1 - B4

[0073] <![CDATA[Scale b ij > <![CDATA[B1]]> <![CDATA[B2]]> <![CDATA[B3]]> <![CDATA[B4]]> Weight w <![CDATA[B1]]> 1 3 / 5 3 / 4 3 / 5 0.1764 <![CDATA[B2]]> 5 / 3 1 5 / 4 1 0.2940 <![CDATA[B3]]> 4 / 3 4 / 5 1 4 / 5 0.2353 <![CDATA[B4]]> 5 / 3 1 5 / 4 1 0.2943

[0074] Index weights of operation convenience

[0075] <![CDATA[Scale b ij > <![CDATA[C 11 > <![CDATA[C 12 > <![CDATA[C 13 > Weight w <![CDATA[C 11 > 1 1 4 0.4444 <![CDATA[C 12 > 1 1 4 0.4444 <![CDATA[C 13 > 1 / 4 1 / 4 1 0.1111

[0076] Essential reliability C of medical indoor patrol robots 21 ~C 25 Index weights

[0077] <![CDATA[Scale b ij > <![CDATA[C 21 > <![CDATA[C 22 > <![CDATA[C 23 > <![CDATA[C 24 > <![CDATA[C 25 > Weight w <![CDATA[C 21 > 1 1 / 6 1 / 4 1 1 / 7 0.0526 <![CDATA[C 22 > 6 1 3 / 2 6 6 / 7 0.3158 <![CDATA[C 23 > 4 2 / 3 1 4 4 / 7 0.2105 <![CDATA[C 24 > 3 1 / 6 1 / 4 1 1 / 7 0.0526 <![CDATA[C 25 > 7 7 / 6 7 / 4 7 1 0.3684

[0078] Appearance safety C 31 ~C 35 Index weights

[0079] <![CDATA[Scale b ij > <![CDATA[C 31 > <![CDATA[C 32 > <![CDATA[C 33 > <![CDATA[C 34 > <![CDATA[C 35 > Weight w <![CDATA[C 31 > 1 1 / 7 1 / 5 7 / 5 1 0.0666 <![CDATA[C 32 > 7 1 7 / 5 7 7 0.4666 <![CDATA[C 33 > 5 5 / 7 1 5 5 0.3333 <![CDATA[C 34 > 1 1 / 7 1 / 5 1 1 0.0667 <![CDATA[C 35 > 1 1 / 7 1 / 5 1 1 0.0667

[0080] Motion matching C 41 ~C 44 Index weights

[0081] <![CDATA[Scale b ij > <![CDATA[C 41 > <![CDATA[C 42 > <![CDATA[C 43 > <![CDATA[C 44 > Weight w <![CDATA[C 41 > 1 3 / 8 3 / 5 3 0.1430 <![CDATA[C 42 > 8 / 3 1 8 / 5 8 / 5 0.3810 <![CDATA[C 43 > 5 / 3 5 / 8 1 1 0.2381 <![CDATA[C 44 > 1 / 3 5 / 8 1 1 0.2381

[0082] In a specific embodiment, 8 hospitals scored the human-machine collaboration safety of the three types of medical indoor patrol robots from 0 - 100 according to actual use, obtained the average values respectively, and calculated the average values of these three types of medical indoor patrol robots as the standardized reference scheme. The results obtained after summarization are shown in Table 11.

[0083] Summary of Mean and Standardization Schemes for Human-Robot Collaboration Safety Evaluation

[0084] Index layer Humanoid robot Small robot vehicle Interactive robot (Mean) Operational convenience 76.565 92.482 85.635 84.894 Intrinsic safety 80.235 75.632 91.214 82.360 External safety 89.245 65.325 85.647 80.072 Motion safety 80.145 91.251 80.356 83.917 Easy to operate 71.265 84.324 88.657 81.415 Autonomous navigation ability 91.234 95.427 59.364 82.008 Task execution efficiency 85.624 90.215 93.254 89.697 Semantic understanding 86.546 60.254 75.489 74.096 Emotional communication 87.274 57.216 60.114 68.201 Service attitude 88.223 61.564 70.021 73.269 Data processing 60.245 80.214 89.998 76.819 Analysis ability 91.230 75.489 88.147 84.955 Humanized design 89.758 66.245 75.485 77.162 Failure rate 62.345 70.241 80.245 70.943

[0085] Based on the human-robot collaboration safety scores and standardized evaluation results of three medical indoor patrol robots, the membership matrix is obtained as follows:

[0086]

[0087] Step S3: According to the index weight vector and the membership matrix, obtain the safety evaluation data of the patrol robot to be evaluated in different dimensions.

[0088] According to the index weight vector and the membership matrix, obtaining the safety evaluation data of the patrol robot to be evaluated in different dimensions also includes the following steps:

[0089] Map each element in the membership matrix to between 0 and 1;

[0090] According to the dimensions to which the safety evaluation indicators belong, divide the index weight vector and the membership matrix to obtain the index weight vector components and membership matrix components corresponding to each dimension. Multiply the index weight vector components and membership matrix components of the same dimension to obtain the safety evaluation data of the corresponding dimension.

[0091] Mapping each element in the membership matrix to between 0 and 1 also includes the following steps:

[0092] When the safety evaluation indicator corresponding to the element in the membership matrix is positively correlated with safety, update the corresponding element according to r ij ′ = a′ ij / maxa′ ij ,

[0093] When the safety evaluation indicator corresponding to the element in the membership matrix is negatively correlated with safety, update the corresponding element according to r ij ′ = mina′ ij / a ij ′,

[0094] where r ij ′ represents the updated value of the element in the i-th row and j-th column a′ ij in the membership matrix, maxa′ ij represents the maximum value among all the membership degrees of the same safety evaluation indicator of the same patrol robot to be evaluated for the same preset safety classification corresponding to a′ ij , mina′ ij represents the minimum value among all the membership degrees of the same safety evaluation indicator of the same patrol robot to be evaluated for the same preset safety classification corresponding to a′ ij in the membership matrix.

[0095] For example, in the membership matrix, the larger the evaluation score of each index of the medical indoor patrol robot, the better. Therefore, the formula r ij = a ij / maxa ij is used to update the membership matrix as follows:

[0096]

[0097] The safety evaluation data is obtained by multiplying the index weight vector and the updated membership matrix, and finally the human-machine collaboration safety evaluation result of the medical indoor patrol robot is obtained.

[0098] Operation convenience G1:

[0099]

[0100] w1 is the component of the index weight vector corresponding to operation convenience G1, and R1 is the component of the membership matrix corresponding to operation convenience G1.

[0101] Intrinsic safety G2:

[0102]

[0103] w2 is the component of the index weight vector corresponding to intrinsic safety G2, and R2 is the component of the membership matrix corresponding to intrinsic safety G2.

[0104] Appearance safety G3:

[0105]

[0106] w3 is the component of the index weight vector corresponding to appearance safety G3, and R3 is the component of the membership matrix corresponding to appearance safety G3.

[0107] Motion safety G4:

[0108]

[0109] w4 is the component of the index weight vector corresponding to motion safety G4, and R4 is the component of the membership matrix corresponding to motion safety G4.

[0110] The comprehensive evaluation results of three medical indoor inspection robots are summarized as shown in the following table. From the comprehensive evaluation results, it can be found that the interactive robot has the highest evaluation of human-machine collaboration safety, and it is higher than the standard level. From the evaluation results of each index, the evaluation results of the operation convenience of the small robot car and the interactive robot are similar, both higher than the average level, and better than the humanoid robot. The motion safety of the humanoid robot is significantly higher than that of the other two medical indoor inspection robots. The interactive robot has a relatively high evaluation result in terms of inherent safety, but is lower than the other two inspection robots in terms of motion safety and appearance safety.

[0111]

[0112] Based on the above evaluation results of the human-machine collaboration safety of the three medical indoor inspection robots, selection suggestions are put forward for medical-related personnel to choose medical indoor inspection robots:

[0113] (1) Humanoid robot. The humanoid robot has the highest score in terms of motion safety, and it is greater than the other two robots. Therefore, for those who are concerned about hospital ward inspections and services, etc., the humanoid robot can be considered.

[0114] (2) Small robot car. The small robot car is higher than the average value in terms of operation convenience, inherent safety, and appearance safety. Therefore, for tasks such as inspection, cleaning, and transportation, the small robot car can be considered. They are flexible and easy to operate and control in medical places.

[0115] (3) Interactive robot. The interactive robot has the highest score in terms of inherent safety, and it is greater than the other two robots. They usually have functions such as voice recognition, face recognition, and voice synthesis. Therefore, for tasks that require simple conversations with patients, providing basic information and entertainment, etc., the interactive robot can be selected.

[0116] In the above method of this embodiment, a standardized safety evaluation can be carried out on the intelligent robot for rounds of diagnosis and ward visits, and the safety performance of the medical indoor inspection robot can be comprehensively quantified and evaluated.

[0117] Those skilled in the art can change the above order without departing from the protection scope of the present disclosure.

[0118] Another embodiment of the present invention provides a safety evaluation system for a medical indoor inspection robot, including:

[0119] The index weight vector determination module is used to compare multiple preset safety evaluation indexes pairwise to determine the relative weight between the index to be compared and the comparison reference index; according to the relative weight of each safety evaluation index, a judgment matrix is established. Each row in the judgment matrix represents the relative weight between the corresponding safety evaluation index and each safety evaluation index when the corresponding safety evaluation index is the index to be compared, and each column represents the relative weight between the corresponding safety evaluation index and each safety evaluation index when the corresponding safety evaluation index is the comparison reference index; the sum of each row of the matrix obtained by normalizing the judgment matrix is calculated to obtain the first matrix; the first matrix is normalized to obtain the index weight vector.

[0120] The membership degree matrix determination module is used to collect the scores of each safety evaluation index of the inspection robot to be evaluated, and determine the membership degree of each safety evaluation index of the inspection robot to be evaluated to each preset safety level according to the scores of each safety evaluation index, so as to obtain the membership degree matrix.

[0121] The safety evaluation data determination module is used to obtain the safety evaluation data of different dimensions of the inspection robot to be evaluated according to the index weight vector and the membership degree matrix.

[0122] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0123] In this embodiment, a standardized safety evaluation can be carried out on the intelligent robot for ward rounds and inspections, and the safety performance of the inspection robot in the medical room can be comprehensively quantified and evaluated.

[0124] Based on the same inventive concept, an embodiment of the present invention provides a computer storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed, the foregoing safety evaluation method for the inspection robot in the medical room is implemented.

[0125] Any modifications, supplements, equivalent replacements, etc. made within the scope of the principles of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A safety evaluation method for a medical indoor inspection robot, characterized in that It includes the following steps: Compare multiple preset safety evaluation indicators pairwise to determine the relative weight between the indicator to be compared and the comparison reference indicator; Establish a judgment matrix according to the relative weight of each safety evaluation indicator. Each row in the judgment matrix represents the relative weight of the corresponding safety evaluation indicator to each safety evaluation indicator when the corresponding safety evaluation indicator is the indicator to be compared, and each column represents the relative weight of the corresponding safety evaluation indicator to each safety evaluation indicator when the corresponding safety evaluation indicator is the comparison reference indicator; Sum the elements of each row of the matrix obtained by normalizing the judgment matrix to obtain the first matrix; Normalize the first matrix to obtain the index weight vector; Collect the scores of each safety evaluation indicator of the inspection robot to be evaluated. According to the scores of each safety evaluation indicator, determine the membership degree of each safety evaluation indicator of the inspection robot to be evaluated to each preset safety level to obtain the membership degree matrix; Obtain the safety evaluation data of different dimensions of the inspection robot to be evaluated according to the index weight vector and the membership degree matrix.

2. The method according to claim 1, characterized in that, Set safety evaluation indicators, including the following steps: Set multiple safety evaluation indicators respectively from the aspects of users, inspection robots, and the environmental association between users and robots, and classify the multiple safety evaluation indicators into different dimensions.

3. The method according to claim 1, characterized in that The judgment matrix is Among them, V = (b ij ) n×n represents the judgment matrix, and b ij represents the relative weight when the i-th safety evaluation index is the index to be compared and the j-th safety evaluation index is the comparison reference index. n represents the total number of safety evaluation indexes.

4. The method according to claim 1, characterized in that, Sum the elements of each row of the matrix obtained by normalizing the judgment matrix to obtain the first matrix, including the following steps: According to normalize the judgment matrix, b ij represents the relative weight when the i-th safety evaluation index is the index to be compared and the j-th safety evaluation index is the comparison reference index in the judgment matrix, n represents the total number of safety evaluation indexes, k represents the row number of the judgment matrix, a ij represents the corresponding value of b ij after the judgment matrix is normalized; According to Sum the elements of each row of the matrix obtained by normalizing the judgment matrix to obtain the first matrix U = [U1, U2,..., U n T , where U i represents the sum of the elements of the i-th row of the matrix obtained by normalizing the judgment matrix, that is, the i-th element of the first matrix.​ 5. The method according to claim 1, characterized in that, Normalize the first matrix to obtain the index weight vector, including the following steps: According to regularize the first matrix to obtain the index weight vector w = [w1, w2,..., w n T , where U i represents the i-th element of the first matrix, U j represents the j-th element of the first matrix, n represents the total number of safety evaluation indicators, and W i represents the i-th element of the index weight vector.​ 6. The method according to claim 1, characterized in that, Establish a judgment matrix according to the relative weight of each safety evaluation indicator, and further include the following steps: By conducting a consistency test on the judgment matrix, where CR represents the consistency ratio; RI represents the average random consistency index; CI represents the consistency index; when CR is less than 0.1, it means that the consistency test is passed; when CR is greater than or equal to 0.1, the judgment matrix is readjusted and modified until the judgment matrix passes the consistency test; Among them n represents the total number of safety evaluation indicators, (Vw) i represents the sum of the elements in the i-th row after the normalization process of the judgment matrix, and W i represents the i-th element of the index weight vector.

7. The method according to claim 1, characterized in that, Obtain the safety evaluation data of different dimensions of the inspection robot to be evaluated according to the index weight vector and the membership degree matrix, and further include the following steps: Map each element in the membership degree matrix to between 0 and 1; According to the dimensions to which the safety evaluation indicators belong, divide the index weight vector and the membership degree matrix to obtain the index weight vector component and the membership degree matrix component corresponding to each dimension. Multiply the index weight vector component and the membership degree matrix component of the same dimension to obtain the safety evaluation data of the corresponding dimension.

8. The method according to claim 7, characterized in that, Map each element in the membership degree matrix to between 0 and 1, and further include the following steps: When the safety evaluation index corresponding to the element in the membership matrix is positively correlated with safety, the corresponding element is updated according to r ij ′ = a′ ij / maxa′ ij , where When the safety evaluation index corresponding to the element in the membership matrix is negatively correlated with safety, the corresponding element is updated according to r ij ′=mina′ ij / a ij ′ Among them, r ij ' represents the element a' in the i-th row and j-th column of the membership matrix ij Updated value, maxa′ ij Represents a′ ij The maximum value of all the memberships of the same safety evaluation index of the same inspection robot to be evaluated for the same preset safety classification, mina′ ij Represents a′ ij The minimum value among all the memberships of the same safety evaluation index of the inspection robot to be evaluated to the same preset safety classification.

9. A safety evaluation system for a medical indoor inspection robot, characterized in that It includes: An index weight vector determination module, which is used to compare multiple preset safety evaluation indicators pairwise to determine the relative weight between the indicator to be compared and the comparison reference indicator; Establish a judgment matrix according to the relative weight of each safety evaluation indicator. Each row in the judgment matrix represents the relative weight of the corresponding safety evaluation indicator to each safety evaluation indicator when the corresponding safety evaluation indicator is the indicator to be compared, and each column represents the relative weight of the corresponding safety evaluation indicator to each safety evaluation indicator when the corresponding safety evaluation indicator is the comparison reference indicator; Sum the elements of each row of the matrix obtained by normalizing the judgment matrix to obtain the first matrix; Normalize the first matrix to obtain the index weight vector; The membership degree matrix determination module is used to collect the scores of each safety evaluation index of the inspection robot to be evaluated, and determine the membership degree of each safety evaluation index of the inspection robot to be evaluated to each preset safety classification according to the scores of each safety evaluation index, so as to obtain the membership degree matrix; The safety evaluation data determination module is used to obtain the safety evaluation data of different dimensions of the inspection robot to be evaluated according to the index weight vector and the membership degree matrix.

10. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions, and when the computer-executable instructions are executed, the safety evaluation method of the medical indoor inspection robot according to any one of claims 1 to 8 is implemented.

Images