Service robot operation quality evaluation method based on knowledge graph and related equipment

Abstract

The invention provides a service robot operation quality evaluation method based on a knowledge graph and related equipment, and relates to the technical field of robots. According to the invention, evaluation index quantification processing is carried out on the related operation information of the target service robot in the execution process of the target service task, so that the actual index values of the plurality of operation evaluation indexes are obtained; then, a robot operation knowledge graph matched with the target service robot is called to perform multi-level index analysis based on the obtained actual index value, so that an overall comprehensive index score of the target service robot for executing the target service task is determined, and when the overall comprehensive index score is lower than a preset index score threshold value, the target service robot is determined to execute the target service task; and calling a robot operation knowledge graph to perform reasoning to generate respective quality reduction reasons and index optimization suggestions of all the to-be-optimized evaluation indexes, thereby effectively improving the reliability of the service robot operation quality evaluation result by using the knowledge graph, and facilitating the auxiliary improvement of the related performance of the service robot in executing the complex long-term service task.

Description

Technical Field

[0001] The present application relates to the technical field of robots, and in particular, to a method for evaluating the operation quality of a service robot based on a knowledge graph and related devices. Background Art

[0002] With the continuous development of science and technology, robot technology has been increasingly widely applied in various industries. Among them, service robot technology applicable to the service field is an important research direction of current robot technology. This technology usually requires service robots to long-term execute complex service tasks in non-enclosed complex public scenarios with multiple dynamic moving targets or obstacles, which puts higher performance requirements on the robot performance such as environmental perception, robot actions, multi-modal interactions (including human-robot interaction and multi-robot interaction), and mobile navigation during the execution of service tasks. Therefore, as a necessary link in the process of optimizing and improving the performance of service robots, the reliability of the operation quality evaluation operation of robots is particularly important.

[0003] However, it should be noted that traditional robot operation quality evaluation schemes are mostly applicable to evaluating the execution status of robots for short-term single operation tasks in closed industrial scenarios, and are difficult to be compatible with the evaluation of the operation quality of robots for complex long-term service tasks. They cannot trace the relevant reasons when the execution effect of service tasks is not good, which is not conducive to optimizing the performance of robots. Summary of the Invention

[0004] In view of this, the purpose of the present application is to provide a method for evaluating the operation quality of a service robot based on a knowledge graph, a computer device, and a readable storage medium, which can use the knowledge graph to perform multi-level index reasoning and evaluation on the execution quality of complex long-term service tasks of service robots in non-enclosed complex public scenarios, and simultaneously achieve the tracing effect of the reasons for poor task effects and the recommendation effect of reference optimization suggestions, so as to effectively improve the reliability of the operation quality evaluation results of service robots and facilitate assisting in improving the relevant performance of service robots in executing complex long-term service tasks.

[0005] To achieve the above purpose, the technical solutions adopted in the embodiments of the present application are as follows:

[0006] In the first aspect, the present application provides a method for evaluating the operation quality of a service robot based on a knowledge graph, and the method includes:

[0007] Obtain relevant operation information during the single task execution of a target service robot for a target service task;

[0008] Perform quantization processing on the evaluation indexes of the relevant operation information to obtain the actual index values of various operation evaluation indexes;

[0009] Invoke the robot operation knowledge graph adapted to the target service robot, perform multi-level index analysis on the actual index values of the multiple operation evaluation indexes, and obtain the overall comprehensive index score;

[0010] In the case that the overall comprehensive index score is lower than the preset index score threshold, determine at least one evaluation index to be optimized according to the actual index values of the multiple operation evaluation indexes, and invoke the robot operation knowledge graph to infer and generate the quality degradation reasons and index optimization suggestions for all the evaluation indexes to be optimized;

[0011] Generate an operation quality report of the target service robot for the target service task based on the overall comprehensive index score and the quality degradation reasons and index optimization suggestions for all the evaluation indexes to be optimized.

[0012] In an alternative embodiment, the robot operation knowledge graph records the hierarchical relationship between the overall comprehensive index, multiple measure item comprehensive indexes, and multiple operation evaluation indexes. Then, the step of invoking the robot operation knowledge graph adapted to the target service robot and performing multi-level index analysis on the actual index values of the multiple operation evaluation indexes to obtain the overall comprehensive index score includes:

[0013] For each measure item comprehensive index, determine all the operation evaluation indexes associated with the measure item comprehensive index according to the hierarchical relationship between the measure item comprehensive index and the multiple operation evaluation indexes;

[0014] Perform grey relational degree calculation on the actual index values of all the operation evaluation indexes associated with the measure item comprehensive index to obtain the actual grey relational degree value of the measure item comprehensive index;

[0015] According to the hierarchical relationship between the overall comprehensive index and the multiple measure item comprehensive indexes, determine the influence weights of the multiple measure item comprehensive indexes for the overall comprehensive index;

[0016] Perform weighted summation operation on the actual grey relational degree values of the multiple measure item comprehensive indexes according to the corresponding influence weights of the multiple measure item comprehensive indexes to obtain the overall comprehensive index score.

[0017] In an alternative embodiment, for each measure item comprehensive index, the step of performing grey relational degree calculation on the actual index values of all the operation evaluation indexes associated with the measure item comprehensive index to obtain the actual grey relational degree value of the measure item comprehensive index includes:

[0018] Normalize the ideal index values of all job evaluation indicators associated with the comprehensive index of this type of measure item at the robot operation knowledge graph to obtain a reference value sequence for the comprehensive index of this type of measure item, and normalize the actual index values of all job evaluation indicators to obtain an actual value sequence for the comprehensive index of this type of measure item; wherein, the reference value sequence is composed of the normalized ideal index values of all job evaluation indicators, and the actual value sequence is composed of the normalized actual index values of all job evaluation indicators;

[0019] Based on the reference value sequence and the actual value sequence, calculate the grey correlation coefficients of all job evaluation indicators;

[0020] Calculate the arithmetic mean of the grey correlation coefficients of all job evaluation indicators to obtain the actual grey correlation degree value of the comprehensive index of this type of measure item.

[0021] In an alternative embodiment, the step of determining at least one evaluation indicator to be optimized according to the actual index values of the multiple job evaluation indicators includes:

[0022] For each type of comprehensive index of measure item, calculate the mathematical expectation of the actual value sequence of the comprehensive index of this type of measure item to obtain the correlation expected value of the comprehensive index of this type of measure item; wherein, the actual value sequence of each type of comprehensive index of measure item is composed of the normalized actual index values of all job evaluation indicators associated with the comprehensive index of this type of measure item;

[0023] According to the influence weights, actual grey correlation degree values and correlation expected values of the multiple types of comprehensive index of measure item, calculate the contribution rates of job quality decline of the multiple types of comprehensive index of measure item;

[0024] Select all the comprehensive index of measure item to be optimized whose corresponding contribution rates of job quality decline meet the first optimization constraint condition from the multiple types of comprehensive index of measure item;

[0025] For each type of comprehensive index of measure item to be optimized, select the target job evaluation indicator whose corresponding contribution rate of correlation meets the second optimization constraint condition from all job evaluation indicators associated with the comprehensive index of this type of measure item to be optimized as the evaluation indicator to be optimized.

[0026] In an alternative embodiment, the contribution rate of job quality decline of the i-th type of comprehensive index of measure item among the multiple types of comprehensive index of measure item is calculated by the following formula:

[0027]

[0028] wherein, C i is used to represent the contribution rate of job quality decline of the i-th type of comprehensive index of measure item, Wi used to represent the influence weight of the comprehensive index of the i-th measure item, γ i used to represent the actual grey correlation degree value of the comprehensive index of the i-th measure item, E i used to represent the correlation expected value of the comprehensive index of the i-th measure item, W k used to represent the influence weight of the k-th measure item comprehensive index among the multiple measure item comprehensive indexes, γ k used to represent the actual grey correlation degree value of the comprehensive index of the k-th measure item, E k used to represent the correlation expected value of the comprehensive index of the k-th measure item, m is used to represent the number of types of indexes of the multiple measure item comprehensive indexes.

[0029] In an optional implementation manner, for each comprehensive index of the measure item to be optimized, the correlation contribution rate of the j-th operation evaluation index associated with this comprehensive index of the measure item to be optimized is calculated by the following formula:

[0030]

[0031] wherein, G j used to represent the correlation contribution rate of the j-th operation evaluation index, ξ j used to represent the grey correlation coefficient of the j-th operation evaluation index, ξ l used to represent the grey correlation coefficient of the l-th operation evaluation index associated with this comprehensive index of the measure item to be optimized, n is used to represent the number of types of indexes of all operation evaluation indexes associated with this comprehensive index of the measure item to be optimized.

[0032] In an optional implementation manner, the robot operation knowledge graph records the logical association relationship between multiple reasons for operation quality decline, the causal association relationship between multiple reasons for operation quality decline and multiple operation evaluation indexes, and the mapping association relationship between multiple reasons for operation quality decline and multiple operation optimization suggestions. Then, the steps of invoking the robot operation knowledge graph to infer and generate the reasons for quality decline and index optimization suggestions for each evaluation index to be optimized include:

[0033] For each evaluation index to be optimized, according to the logical association relationship and causal association relationship recorded in the robot operation knowledge graph, search for the reasons for quality decline that match this evaluation index to be optimized;

[0034] For each reason for quality decline found, according to the mapping association relationship recorded in the robot operation knowledge graph, search for the operation optimization suggestion that matches this reason for quality decline as the index optimization suggestion for this evaluation index to be optimized.

[0035] In an optional implementation manner, the method further includes:

[0036] Obtain the task execution optimization data of the target service robot for the target service task based on the job quality report;

[0037] Update the robot operation knowledge graph according to the task execution optimization data.

[0038] In a second aspect, the present application provides a computer device, including a processor and a memory. The memory stores a computer program executable by the processor, and the processor can execute the computer program to implement the method for evaluating the job quality of a service robot based on a knowledge graph according to any one of the foregoing embodiments.

[0039] In a third aspect, the present application provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a computer device, it implements the method for evaluating the job quality of a service robot based on a knowledge graph according to any one of the foregoing embodiments.

[0040] In this case, the beneficial effects of the embodiments of the present application may include the following:

[0041] In the present application, the relevant operation information in the single task execution process of the target service robot for the target service task is quantified by evaluation indexes to obtain the actual index values of various operation evaluation indexes. Then, the robot operation knowledge graph adapted to the target service robot is called to perform multi-level index analysis on the actual index values of various operation evaluation indexes to obtain the overall comprehensive index score of the target service robot for executing the target service task. When the overall comprehensive index score is lower than the preset index score threshold, at least one evaluation index to be optimized is determined according to the actual index values of these various operation evaluation indexes. Then, the robot operation knowledge graph is called to reason and generate the quality decline reasons and index optimization suggestions for all the evaluation indexes to be optimized, so as to generate a corresponding job quality report. Therefore, it is possible to use the knowledge graph to perform multi-level index reasoning and evaluation on the execution quality of the service robot for complex long-term service tasks in a non-closed complex public scene, and simultaneously achieve the traceability effect of the reasons for poor task effects and the recommendation effect of reference optimization suggestions, so as to improve the reliability of the service robot job quality evaluation results and facilitate assisting in improving the relevant performance of the service robot for executing complex long-term service tasks (for example, robot performances such as environmental perception, robot actions, multi-modal interaction, and mobile navigation).

[0042] To make the above objects, features, and advantages of the present application more obvious and understandable, the following specific preferred embodiments are given in conjunction with the accompanying drawings and are described in detail as follows. Description of the Drawings

[0043] To more clearly illustrate the technical solutions of the embodiments of the present application, the accompanying drawings required for the embodiments will be briefly introduced below. It should be understood that the following accompanying drawings only show some embodiments of the present application and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related accompanying drawings can also be obtained based on these drawings.

[0044] Figure 1 Schematic diagram of the device composition of the computer device provided for the embodiments of the present application;

[0045] Figure 2 Schematic diagram of the system of the multi-level robot operation quality evaluation index system provided for the embodiments of the present application;

[0046] Figure 3 One of the flow schematic diagrams of the method for evaluating the operation quality of a service robot based on a knowledge graph provided for the embodiments of the present application;

[0047] Figure 4 For Figure 3 The flow schematic diagram of the sub-steps included in step S230 in

[0048] Figure 5 For Figure 3 One of the flow schematic diagrams of the sub-steps included in step S240 in

[0049] Figure 6 For Figure 3 The second of the flow schematic diagrams of the sub-steps included in step S240 in

[0050] Figure 7 The second of the flow schematic diagrams of the method for evaluating the operation quality of a service robot based on a knowledge graph provided for the embodiments of the present application.

[0051] Icons: 10 - Computer device; 11 - Memory; 12 - Processor; 13 - Communication unit. Detailed implementation manners

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Usually, the components of the embodiments of the present application described and shown in the accompanying drawings here can be arranged and designed in various different configurations.

[0053] Accordingly, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but merely represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the scope of protection of the present application.

[0054] It should be noted that like reference numerals and letters denote like items in the following drawings. Therefore, once an item is defined in one drawing, it does not require further definition and explanation in subsequent drawings.

[0055] In the description of the present application, it should be understood that the orientation or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings, or the orientation or positional relationships in which the product of this application is customarily placed during use, or the orientation or positional relationships commonly understood by those skilled in the art. These are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation of the present application.

[0056] In the description of the present application, it should also be noted that, unless otherwise clearly specified and limited, the terms "set", "installed", "connected", "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0057] In addition, in the description of the present application, it can be understood that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variation thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising a..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0058] The following will describe in detail some embodiments of the present application with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments may be combined with each other.

[0059] Please refer to Figure 1 , Figure 1 which is a schematic diagram of the device composition of the computer device 10 provided by an embodiment of the present application. In the embodiment of the present application, the computer device 10 can be communicatively connected to any service robot and obtain relevant operation information of the corresponding service robot for performing complex long-term service tasks in a non-enclosed complex public scenario, so as to use a robot operation knowledge graph adapted to the service robot based on the obtained relevant operation information to perform multi-level index reasoning and evaluation on the task execution quality of the service robot for complex long-term service tasks, and can effectively trace the reasons for poor task execution effects in the case of poor task execution effects, and provide relatively reliable reference optimization suggestions for the traced reasons for poor task execution effects, thereby effectively improving the reliability of the service robot operation quality evaluation results and facilitating the auxiliary improvement of the relevant performance of the service robot for performing complex long-term service tasks.

[0060] In the embodiments of the present application, the computer device 10 may be, but is not limited to, a server, a personal computer, a laptop computer, etc.; the relevant operation information may include the operation environment information, service interaction information, robot state space information, and operation execution log information of the corresponding service robot. The operation environment information includes the sensing data collected by various environmental sensors (including temperature sensors, humidity sensors, etc.) carried by the service robot. The service interaction information includes information such as task instruction records, interactive Q&A, and service object evaluations. The robot state space information includes the state space information records of the service robot at different times; the robot operation knowledge graph records the association relationships between a multi-level robot operation quality evaluation index system related to the task execution of the service robot, various reasons for the decline in operation quality, and various operation optimization suggestions. The multi-level robot operation quality evaluation index system involves the hierarchical relationships among the overall comprehensive index, various measure item comprehensive indexes, and various operation evaluation indexes. Among them, the various measure item comprehensive indexes are all index branches of the overall comprehensive index. Each measure item comprehensive index separately involves various operation evaluation indexes, and the operation evaluation indexes involved in different measure item comprehensive indexes are different from each other. The overall comprehensive index is used to describe the overall quality state of the service task execution result of the corresponding service robot. Each measure item comprehensive index is used to describe the overall working quality status of the corresponding service robot at a certain robot performance level (for example, environmental perception or mobile navigation). Each operation evaluation index is used to describe the ability advantages and disadvantages of the corresponding service robot at a certain performance focus at a specific robot performance level (for example, path planning or obstacle avoidance ability at the mobile navigation performance level).

[0061] For Figure 2Taking the multi-level robot operation quality evaluation index system shown as an example, the comprehensive indicators of various measurement items involved in the multi-level robot operation quality evaluation index system may include, but are not limited to: "(Hardware) performance comprehensive index", "Intention understanding comprehensive index", "Environmental perception comprehensive index", "Navigation comprehensive index" and "(Action) execution comprehensive index". Among them, the various operation evaluation indicators involved in the "(Hardware) performance comprehensive index" may include, but are not limited to: "Time efficiency index (which can be comprehensively determined by statistically averaging the time, maximum time, minimum time, etc. required to complete the same service task multiple times)", "Energy consumption efficiency index", "Endurance capacity index (such as endurance mileage)", "Stability index (which is comprehensively determined by statistically the system oscillation amplitude, system recovery time, etc.)", "Braking capacity index", "Maximum slope index", etc. The various operation evaluation indicators involved in the "Intention understanding comprehensive index" may include, but are not limited to: "Intention analysis index (which can be comprehensively determined by statistically the satisfaction score of the service object, the accuracy score of emotion understanding, etc.)", "Human-computer interaction index (such as speech recognition accuracy)", etc. The various operation evaluation indicators involved in the "Environmental perception comprehensive index" may include, but are not limited to: "Scene classification index", "Target positioning index (which can be comprehensively determined by statistically the position resolution, position accuracy, angle resolution, angle accuracy, etc.)", etc. The various operation evaluation indicators involved in the "Navigation comprehensive index" may include, but are not limited to: "Path planning index", "Obstacle avoidance ability index (which can be comprehensively determined by statistically the success rate of avoiding fixed obstacles, the success rate of avoiding moving obstacles, the success rate of avoiding movable obstacles, the success rate of crossing obstacles, etc.)", etc. The various operation evaluation indicators involved in the "(Action) execution comprehensive index" may include, but are not limited to: "Task execution accuracy index", "Task switching time index" and "Task success rate index", etc.

[0062] For any robot operation knowledge graph, the graph entities involved may include, but are not limited to: task entities, metric entities, threshold entities, cause entities, and solution entities, etc. Among them, the task entity is used to describe the service tasks that the corresponding service robot can execute, the metric entity is used to describe the specific metric components of the multi-level robot operation quality evaluation index system related to the service task, the threshold entity is used to describe the ideal values or thresholds of different metrics (for example, the ideal values of all operation evaluation metrics respectively, and the threshold of the overall comprehensive metric (i.e., the preset metric score threshold)), the cause entity is used to describe the possible causes for the decline in task execution quality (for example, equipment problems, environmental problems, model problems, etc.), and the solution entity is used to describe feasible robot performance improvement solutions or operation optimization suggestions (for example, alarm, update map, online training model, etc.); at the same time, the graph relationships involved in this robot operation knowledge graph may include, but are not limited to: the hierarchical relationship among the overall comprehensive metric, multiple measure item comprehensive metrics, and multiple operation evaluation metrics, the score evaluation relationship among the overall comprehensive metric, multiple measure item comprehensive metrics, and multiple operation evaluation metrics, the numerical association relationship between the overall comprehensive metric, multiple measure item comprehensive metrics, and multiple operation evaluation metrics and their respective ideal values or thresholds, the logical association relationship among multiple operation quality decline causes, the causal association relationship between multiple operation quality decline causes and multiple operation evaluation metrics, and the mapping association relationship between multiple operation quality decline causes and multiple operation optimization suggestions, etc.

[0063] In the embodiment of the present application, the computer device 10 may include a memory 11, a processor 12, and a communication unit 13. Among them, the memory 11, the processor 12, and the communication unit 13 are electrically connected to each other directly or indirectly to achieve data transmission or interaction. For example, these components of the memory 11, the processor 12, and the communication unit 13 may be electrically connected to each other through one or more communication buses or signal lines.

[0064] In an embodiment of the present application, the memory 11 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electric Erasable Programmable Read-Only Memory (EEPROM), etc. Among them, the memory 11 is used to store a computer program, and after receiving an execution instruction, the processor 12 can execute the computer program accordingly.

[0065] In an embodiment of the present application, the processor 12 may be an integrated circuit chip with signal processing capabilities. The processor 12 may be a general-purpose processor, including at least one of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Network Processor (NP), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc., and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application.

[0066] In an embodiment of the present application, the communication unit 13 is used to establish a communication connection between the computer device 10 and other electronic devices through a network, and transmit and receive data through the network, where the network includes a wired communication network and a wireless communication network. For example, the computer device 10 can obtain relevant operation information of a specific service robot performing complex long-term service tasks in a non-enclosed complex public scenario through the communication unit 13.

[0067] In an embodiment of the present application, the computer device 10 may pre-store a specific computer program related to the service robot operation quality evaluation function at the memory 11, and by driving the processor 12 to execute the specific computer program correspondingly, use the knowledge graph to perform a multi-level index reasoning evaluation on the execution quality of the complex long-term service task of the service robot in a non-enclosed complex public scenario, and simultaneously achieve the traceability effect of the reasons for the poor task effect and the recommendation effect of reference optimization suggestions, so as to improve the reliability of the service robot operation quality evaluation result and facilitate assisting in improving the relevant performance of the service robot in executing complex long-term service tasks.

[0068] It can be understood that Figure 1 The block diagram shown is only a schematic diagram of the composition of the computer device 10, and the computer device 10 may also include more or fewer components than those shown Figure 1 shown, or have a different configuration from that Figure 1 shown. Figure 1 Each component shown can be implemented by hardware, software, or a combination thereof.

[0069] In the present application, to ensure that the computer device 10 can reliably evaluate the service robot operation quality and can provide the user with the reasons for the poor task execution effect and reference optimization suggestions, so as to facilitate assisting in improving the relevant performance of the service robot in executing complex long-term service tasks, the embodiment of the present application realizes the foregoing purpose by providing a service robot operation quality evaluation method based on a knowledge graph. The service robot operation quality evaluation method provided by the present application will be described in detail below.

[0070] Please refer to Figure 3 , Figure 3 which is one of the flow diagrams of the service robot operation quality evaluation method based on a knowledge graph provided by an embodiment of the present application. In the embodiment of the present application, the service robot operation quality evaluation method may include steps S210 to S250.

[0071] Step S210, obtain relevant operation information during a single task execution of the target service robot for the target service task.

[0072] In this embodiment, the relevant operation information includes the operation environment information, service interaction information, robot state space information, and operation execution log information of the target service robot during one execution of the target service task; the relevant operation information can be directly uploaded by the target service robot to the computer device 10, and the target service robot is any service robot communicatively connected to the computer device 10.

[0073] Step S220: Quantify the evaluation indicators for relevant operation information to obtain the actual indicator values of various operation evaluation indicators.

[0074] In this embodiment, the computer device 10 can determine all the operation evaluation indicators required in the process of evaluating the quality of the target service robot's operation (i.e., composed of all the operation evaluation indicators involved in the robot operation knowledge graph) by invoking the robot operation knowledge graph adapted to the target service robot. Then, according to the mathematical descriptions of all the determined operation evaluation indicators, perform quantification processing of the evaluation indicators on the relevant operation information of the target service robot for executing the target service task, and obtain the actual indicator values of all the operation evaluation indicators of the target service robot under the corresponding multi-level robot operation quality evaluation index system.

[0075] Step S230: Invoke the robot operation knowledge graph adapted to the target service robot, and perform multi-level index analysis on the actual indicator values of various operation evaluation indicators to obtain the overall comprehensive index score.

[0076] In this embodiment, after the computer device 10 determines the actual indicator values of all the operation evaluation indicators of the target service robot under the corresponding multi-level robot operation quality evaluation index system, it can perform multi-level index analysis on the actual indicator values of all the determined operation evaluation indicators according to the hierarchical relationship and score evaluation relationship among the overall comprehensive index, various measure item comprehensive indexes, and various operation evaluation indicators recorded in the robot operation knowledge graph, so as to infer and evaluate the overall comprehensive index score of the target service robot for executing the target service task this time.

[0077] Optionally, please refer to Figure 4 , Figure 4 Yes Figure 3 is a schematic flowchart of the sub-steps included in step S230. In the embodiment of the present application, step S230 may include sub-steps S231 to S234 to perform multi-level index analysis on various operation evaluation indicators of the target service robot for executing the target service task based on the grey system theory, so as to accurately evaluate the specific operation quality of the target service robot for executing the target service task.

[0078] Sub-step S231: For each measure item comprehensive index, determine all the operation evaluation indicators associated with this measure item comprehensive index according to the hierarchical relationship between this measure item comprehensive index and various operation evaluation indicators.

[0079] Sub-step S232: Calculate the grey relational degree for each actual index value of all job evaluation indicators associated with the comprehensive index of this type of measurement item, and obtain the actual grey relational degree value of the comprehensive index of this type of measurement item.

[0080] In this embodiment, the scoring evaluation relationship between each comprehensive index of the measurement item and multiple associated job evaluation indicators can be reflected by the "grey system theory". Then, for each comprehensive index of the measurement item, the grey relational degree can be calculated using the actual index values of all job evaluation indicators associated with it to obtain the specific index score (i.e., the actual grey relational degree value) of the comprehensive index of this type of measurement item during the execution of the current target service task.

[0081] During this process, for each comprehensive index of the measurement item, the step of calculating the grey relational degree for each actual index value of all job evaluation indicators associated with the comprehensive index of this type of measurement item to obtain the actual grey relational degree value of the comprehensive index of this type of measurement item may include:

[0082] Normalize the ideal index values of all job evaluation indicators associated with the comprehensive index of this type of measurement item at the robot job knowledge graph to obtain the reference value sequence of the comprehensive index of this type of measurement item, and normalize the actual index values of all job evaluation indicators associated with the comprehensive index of this type of measurement item to obtain the actual value sequence of the comprehensive index of this type of measurement item; where the reference value sequence is composed of the normalized ideal index values of all job evaluation indicators, and the actual value sequence is composed of the normalized actual index values of all job evaluation indicators.

[0083] Based on the reference value sequence and the actual value sequence, calculate the grey relational coefficients of all job evaluation indicators associated with the comprehensive index of this type of measurement item.

[0084] Calculate the arithmetic mean of the grey relational coefficients of all job evaluation indicators associated with the comprehensive index of this type of measurement item to obtain the actual grey relational degree value of the comprehensive index of this type of measurement item.

[0085] Among them, the ideal index value of each job evaluation indicator at the robot job knowledge graph can be represented by the historical optimal index parameter value during the test process of the target service robot, or can be obtained by calculating the average value of the historical index values of this job evaluation indicator during multiple executions of the target service task by the target service robot in the past. The specific calculation process of the ideal index value of various job evaluation indicators in this application is not limited.

[0086] Sub-step S233: Determine the influence weights of each comprehensive index of multiple measurement items on the overall comprehensive index according to the hierarchical relationship between the overall comprehensive index and the comprehensive indexes of multiple measurement items.

[0087] In this embodiment, the score evaluation relationship between the overall comprehensive index and the comprehensive indexes of multiple measurement items associated therewith can be reflected by "the influence weights of the comprehensive indexes of multiple measurement items on the overall comprehensive index respectively". Thus, after the computer device 10 calculates the actual grey correlation degree values of each comprehensive index of the measurement item, it will, based on the hierarchical relationship and score evaluation relationship between the overall comprehensive index and the comprehensive indexes of multiple measurement items recorded in the robot operation knowledge graph, find the influence weights of the comprehensive indexes of multiple measurement items associated with the overall comprehensive index respectively.

[0088] Sub-step S234: According to the influence weights corresponding to the comprehensive indexes of multiple measurement items respectively, perform a weighted summation operation on the actual grey correlation degree values of the comprehensive indexes of multiple measurement items respectively to obtain the overall comprehensive index score.

[0089] Thus, this application can, by executing the above sub-steps S231 to S234, perform a multi-level index analysis on multiple operation evaluation indexes for the target service robot to execute the target service task based on the grey system theory, so as to accurately evaluate the specific operation quality of the target service robot to execute the target service task.

[0090] Step S240: In the case where the overall comprehensive index score is lower than the preset index score threshold, determine at least one evaluation index to be optimized according to the actual index values of multiple operation evaluation indexes, and call the robot operation knowledge graph to infer and generate the quality decline reasons and index optimization suggestions for all the evaluation indexes to be optimized respectively.

[0091] In this embodiment, when the computer device 10 evaluates the overall comprehensive index score of the target service robot for this execution of the target service task, it will call the robot operation knowledge graph to determine the preset index score threshold corresponding to the overall comprehensive index, and then compare the overall comprehensive index score with the preset index score threshold numerically to determine whether the target service task needs to optimize the robot performance.

[0092] Among them, when the overall comprehensive index score is greater than or equal to the preset index score threshold, it indicates that the execution effect of the target service task is good, the robot performance of the target service robot in executing the target service task is in a better state, and the robot performance of the target service task does not need to be optimized. At this time, the computer device 10 can directly generate a job quality report of the target service robot for the target service task based on the overall comprehensive index score for the user's reference. When the overall comprehensive index score is lower than the preset index score threshold, it indicates that the execution effect of the target service task is poor, the robot performance of the target service robot in executing the target service task is relatively poor, and the robot performance of the target service task needs to be optimized. At this time, the computer device 10 will determine at least one evaluation index to be optimized that needs to be optimized currently among many job evaluation indexes based on the actual index values of all job evaluation indexes under the multi-level robot job quality evaluation index system. Then, for each evaluation index to be optimized, the robot job knowledge graph is called to trace the reason for the decline in job quality and infer job optimization suggestions, so as to obtain the reason for the decline in quality of this evaluation index to be optimized and the index optimization suggestions corresponding to the reason for the decline in quality.

[0093] Optionally, in an implementation manner of the embodiment of the present application, in the process of determining the evaluation index to be optimized, the computer device 10 can perform normalization processing on the actual index values of multiple job evaluation indexes belonging to the same measure item comprehensive index, and select the target job evaluation index whose corresponding normalized actual index value is not within a specific index value range (for example, 0.4 to 0.7) from these multiple job evaluation indexes as the evaluation index to be optimized; it can also determine the target measure item comprehensive index with the largest influence weight by considering the influence weights of different measure item comprehensive indexes on the overall comprehensive index, and then perform normalization processing on the actual index values of multiple job evaluation indexes associated with the target measure item comprehensive index, and select the target job evaluation index whose corresponding normalized actual index value is not within a specific index value range from these multiple job evaluation indexes as the evaluation index to be optimized.

[0094] Optionally, please refer to Figure 5 , Figure 5 Yes Figure 3 One of the flow diagrams of the sub-steps included in step S240. In the embodiment of the present application, the step "determine at least one evaluation index to be optimized according to the actual index values of multiple job evaluation indexes" in step S240 may include sub-steps S241 to S244 to screen out all evaluation indexes to be optimized with greater optimization value among many job evaluation indexes.

[0095] Sub-step S241: For each comprehensive measure item index, calculate the mathematical expectation of the actual value sequence of this comprehensive measure item index to obtain the associated expected value of this comprehensive measure item index.

[0096] In this embodiment, the actual value sequence of each comprehensive measure item index is composed of the normalized actual index values of all job evaluation indexes associated with this comprehensive measure item index. Then, the associated expected value of this comprehensive measure item index is the average value among the normalized actual index values of all job evaluation indexes associated with this comprehensive measure item index.

[0097] Sub-step S242: According to the influence weights, actual grey correlation degree values, and associated expected values of various comprehensive measure item indexes, calculate the contribution rates of various comprehensive measure item indexes to the decline in job quality.

[0098] In this embodiment, the contribution rate of the i-th comprehensive measure item index among various comprehensive measure item indexes associated with the overall comprehensive index to the decline in job quality is calculated using the following formula:

[0099]

[0100] where, C i is used to represent the contribution rate of the i-th comprehensive measure item index to the decline in job quality, W i is used to represent the influence weight of the i-th comprehensive measure item index, γ i is used to represent the actual grey correlation degree value of the i-th comprehensive measure item index, E i is used to represent the associated expected value of the i-th comprehensive measure item index, W k is used to represent the influence weight of the k-th comprehensive measure item index among various comprehensive measure item indexes, γ k is used to represent the actual grey correlation degree value of the k-th comprehensive measure item index, E k is used to represent the associated expected value of the k-th comprehensive measure item index, and m is used to represent the number of types of indexes of various comprehensive measure item indexes.

[0101] Sub-step S243: From various comprehensive measure item indexes, select all the comprehensive measure item indexes to be optimized whose corresponding contribution rates of decline in job quality meet the first optimization constraint condition.

[0102] In this embodiment, the first optimization constraint condition is used to represent the distribution requirement of the contribution rate of decline in job quality of the comprehensive measure item indexes that need to be optimized in the corresponding multi-level robot job quality evaluation index system, which can be but is not limited to: "the top preset number of seats in the ranking of the contribution rate of decline in job quality (for example, 4)", "the corresponding contribution rate of decline in job quality is in a specific contribution rate interval (for example, 50% - 80%)", etc.

[0103] In an implementation manner of this embodiment, if the first optimization constraint condition is set to "the top 3 seats in terms of the contribution rate of the decline in job quality", the computer device 10 will rank the comprehensive indicators of multiple measurement items associated with the overall comprehensive indicator in descending order according to the contribution rate of the decline in job quality of each comprehensive indicator of the measurement item, and select the 3 comprehensive indicators of the measurement item with the largest corresponding contribution rate of the decline in job quality as a comprehensive indicator of the measurement item to be optimized respectively.

[0104] Sub-step S244: For each comprehensive indicator of the measurement item to be optimized, select the target job evaluation indicator whose corresponding correlation contribution rate meets the second optimization constraint condition as the evaluation indicator to be optimized from all job evaluation indicators associated with this comprehensive indicator of the measurement item to be optimized.

[0105] In this embodiment, for each comprehensive indicator of the measurement item to be optimized, the correlation contribution rate of the j-th job evaluation indicator associated with this comprehensive indicator of the measurement item to be optimized is calculated by the following formula:

[0106]

[0107] where G j is used to represent the correlation contribution rate of the j-th job evaluation indicator, and ξ j is used to represent the grey correlation coefficient of the j-th job evaluation indicator, ξ l is used to represent the grey correlation coefficient of the l-th job evaluation indicator associated with this comprehensive indicator of the measurement item to be optimized, and n is used to represent the number of types of indicators of all job evaluation indicators associated with this comprehensive indicator of the measurement item to be optimized.

[0108] In this embodiment, the second optimization constraint condition is used to represent the distribution requirement of the correlation contribution rate of the job evaluation indicator that needs to be optimized in a single comprehensive indicator of the measurement item to be optimized, and it can be, but is not limited to: "the top preset number of seats in terms of the correlation contribution rate (for example, 3)", "the corresponding correlation contribution rate is in a specific contribution rate range (for example, 60% - 80%)", etc.

[0109] In an implementation manner of this embodiment, if the second optimization constraint condition is set to "the top 4 seats in terms of the correlation contribution rate", then for each comprehensive indicator of the measurement item to be optimized, the computer device 10 will rank the multiple job evaluation indicators associated with this comprehensive indicator of the measurement item to be optimized in descending order according to the correlation contribution rate of each job evaluation indicator, and select the 4 target job evaluation indicators with the largest corresponding correlation contribution rate as an evaluation indicator to be optimized respectively.

[0110] Thus, by executing the above sub-steps S241 to S244, this application can screen out all the evaluation metrics to be optimized with relatively high optimization value from among a multitude of job evaluation metrics.

[0111] Optionally, please refer to Figure 6 , Figure 6 which Figure 3 is the second schematic flow diagram of the sub-steps included in step S240 in

[0112] Sub-step S245: For each evaluation metric to be optimized, according to the logical association relationship and causal association relationship recorded in the robot job knowledge graph, search for the quality degradation reasons that match this evaluation metric to be optimized.

[0113] In this embodiment, the robot job knowledge graph records the logical association relationship between multiple job quality degradation reasons (for example, the job quality degradation reason "the robot has a hardware problem" will trigger the job quality degradation reasons "low energy consumption efficiency" and "high system oscillation amplitude"), and the causal association relationship between multiple job quality degradation reasons and multiple job evaluation metrics. Therefore, for each evaluation metric to be optimized, the computer device 10 can, by invoking the robot job knowledge graph, based on the causal association relationship, find the job quality degradation reasons directly associated with this evaluation metric to be optimized (for example, "low energy consumption efficiency" and "high system oscillation amplitude"), and then, based on the causal association relationship, find the job quality degradation reasons indirectly associated with this evaluation metric to be optimized (for example, "the robot has a hardware problem"), so as to obtain the quality degradation reasons that match this evaluation metric to be optimized (that is, the aforementioned directly associated job quality degradation reasons and the aforementioned indirectly associated job quality degradation reasons).

[0114] Sub-step S246: For each found quality degradation reason, according to the mapping association relationship recorded in the robot job knowledge graph, search for the job optimization suggestions that match this quality degradation reason as the index optimization suggestions for this evaluation metric to be optimized.

[0115] In this embodiment, the robot operation knowledge graph records the mapping association relationship between various reasons for the decline in operation quality and various operation optimization suggestions (for example, the operation optimization suggestion for the reason of "robot path planning error" in the decline of operation quality is "adjust the path planning model"). The computer device 10 can infer the respective reference optimization suggestions for each quality decline reason based on the mapping association relationship by invoking the robot operation knowledge graph, and use the inferred reference optimization suggestions as one of the index optimization suggestions for the to-be-optimized evaluation index matched with this type of quality decline reason.

[0116] Thus, this application can implement the tracing effect of the reasons for the poor task execution effect and the recommendation effect of the reference optimization suggestions by executing the above sub-steps S245 to sub-step S246 by using the knowledge graph.

[0117] Step S250, generate an operation quality report of the target service robot for the target service task based on the overall comprehensive index score, the quality decline reasons and index optimization suggestions of all the to-be-optimized evaluation indexes.

[0118] Thus, this application can perform multi-level index reasoning and evaluation on the execution quality of the complex long-term service task of the service robot in the non-closed complex public scene by executing the above steps S210 to step S250, and simultaneously implement the tracing effect of the reasons for the poor task execution effect and the recommendation effect of the reference optimization suggestions, so as to improve the reliability of the operation quality evaluation result of the service robot and facilitate assisting in improving the related performance of the service robot in executing the complex long-term service task.

[0119] Optionally, please refer to Figure 7 , Figure 7 which is the second flowchart of the service robot operation quality evaluation method provided by the embodiment of this application. In the embodiment of this application, compared with the service robot operation quality evaluation method shown in Figure 3 , the service robot operation quality evaluation method shown in Figure 7 may further include steps S260 to step S270 to ensure that the updated knowledge graph can be further compatible and adapted with the optimized operation effect of the service robot, and improve the reliability of various reasons for the decline in operation quality and / or operation optimization suggestions recorded in the knowledge graph.

[0120] Step S260, obtain the task execution optimization data of the target service robot for the target service task based on the operation quality report.

[0121] In this embodiment, after the user obtains the job quality report, the user will refer to various quality decline reasons and various index optimization suggestions recorded in the job quality report, and optimize the specific job details of the target service robot for performing the target service task (including but not limited to: robot hardware composition, robot sensor setting positions, robot positioning algorithms, robot path planning algorithms, robot obstacle avoidance algorithms, robot job space size, etc.), so as to obtain the task execution optimization data of the target service robot for the target service task, where the task execution optimization data is used to describe the job optimization process of the target service robot for the target service task.

[0122] Step S270: Update the graph of the robot job knowledge graph according to the task execution optimization data.

[0123] In this embodiment, the graph update operation of the robot job knowledge graph may include but not be limited to: multi-level robot job quality evaluation index system adjustment operations (including adding / deleting job evaluation indexes, adding / deleting comprehensive indexes of measurement items, adjusting the influence weights of comprehensive indexes of measurement items, etc.), index ideal value / threshold adjustment operations, job quality decline reason adjustment operations (including adding / deleting job quality decline reasons, adjusting logical association relationships, etc.), job optimization suggestion adjustment operations (including adding / deleting / modifying job optimization suggestions, adjusting mapping association relationships, etc.).

[0124] Thus, by executing the above steps S260 to S270, the present application can ensure that the updated knowledge graph can be further compatible and adapted to the optimized service robot job effect, and improve the reliability of various job quality decline reasons and / or job optimization suggestions recorded in the knowledge graph.

[0125] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to the embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0126] In addition, each functional module in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part. If the function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0127] The above is only various implementation manners of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for evaluating the quality of service robot operations based on knowledge graph, characterized in that: The method comprises: Obtain relevant operation information of the target service robot during the execution of a single task for the target service task; Performing evaluation index quantification processing on the relevant operation information to obtain actual index values ​​of various operation evaluation indexes; Calling a robot operation knowledge graph adapted to the target service robot, performing a multi-level indicator analysis on the actual indicator values ​​of each of the multiple operation evaluation indicators, and obtaining an overall comprehensive indicator score; In the case where the overall comprehensive index score is lower than the preset index score threshold, at least one evaluation index to be optimized is determined according to the actual index values ​​of each of the multiple operation evaluation indicators, and the robot operation knowledge graph reasoning is called to generate the reasons for the quality decline of each of the evaluation indicators to be optimized and the index optimization suggestions; Based on the overall comprehensive index score and the reasons for the quality degradation of all the evaluation indicators to be optimized and the index optimization suggestions, a work quality report of the target service robot for the target service task is generated.

2. The method according to claim 1, characterized in that The robot operation knowledge graph records the hierarchical relationship between the overall comprehensive index, the comprehensive index of multiple measurement items and the multiple operation evaluation indexes. The step of calling the robot operation knowledge graph adapted to the target service robot, performing multi-level index analysis on the actual index values ​​of each of the multiple operation evaluation indexes, and obtaining the overall comprehensive index score includes: For each comprehensive indicator of measurement item, according to the hierarchical relationship between the comprehensive indicator of measurement item and multiple operation evaluation indicators, all operation evaluation indicators associated with the comprehensive indicator of measurement item are determined; Perform grey correlation calculation on the actual index values ​​of all job evaluation indexes associated with the comprehensive index of the measurement item to obtain the actual grey correlation value of the comprehensive index of the measurement item; Determining the influence weights of the various measurement item comprehensive indicators on the overall comprehensive indicator according to the hierarchical relationship between the overall comprehensive indicator and the various measurement item comprehensive indicators; According to the influence weights corresponding to the various comprehensive indicators of the measurement items, a weighted sum operation is performed on the actual grey correlation values ​​of the various comprehensive indicators of the measurement items to obtain the overall comprehensive indicator score.

3. The method according to claim 2, characterized in that For each comprehensive indicator of measurement item, the step of performing grey correlation calculation on the actual indicator values ​​of all job evaluation indicators associated with the comprehensive indicator of the measurement item to obtain the actual grey correlation value of the comprehensive indicator of the measurement item includes: Normalizing the ideal index values ​​of all job evaluation indicators associated with the comprehensive index of the measurement item at the robot job knowledge graph to obtain a reference value sequence of the comprehensive index of the measurement item, and normalizing the actual index values ​​of all the job evaluation indicators to obtain an actual value sequence of the comprehensive index of the measurement item; wherein the reference value sequence is composed of the normalized ideal index values ​​of all the job evaluation indicators, and the actual value sequence is composed of the normalized actual index values ​​of all the job evaluation indicators; Based on the reference value sequence and the actual value sequence, calculating the grey correlation coefficient of each of the all operation evaluation indicators; The arithmetic mean value of the grey correlation coefficients of all the job evaluation indicators is calculated to obtain the actual grey correlation value of the comprehensive indicator of the measurement item.

4. The method according to claim 2, characterized in that: The step of determining at least one evaluation indicator to be optimized according to the actual indicator values ​​of each of the multiple job evaluation indicators comprises: For each comprehensive indicator of measurement item, mathematical expectation calculation is performed on the actual value sequence of the comprehensive indicator of the measurement item to obtain the associated expected value of the comprehensive indicator of the measurement item; wherein the actual value sequence of each comprehensive indicator of measurement item is composed of the normalized actual indicator values ​​of all job evaluation indicators associated with the comprehensive indicator of the measurement item; Calculate the contribution rate of the work quality decline of each of the multiple comprehensive indicators of the measurement items according to the influence weights, actual grey correlation values ​​and correlation expected values ​​of each of the multiple comprehensive indicators of the measurement items; From the plurality of comprehensive indicators of measurement items, all comprehensive indicators of measurement items to be optimized whose corresponding operation quality degradation contribution rates satisfy the first optimization constraint condition are selected; For each comprehensive indicator of the measurement item to be optimized, a target operation evaluation indicator whose corresponding correlation contribution rate satisfies the second optimization constraint condition is selected from all operation evaluation indicators associated with the comprehensive indicator of the measurement item to be optimized as the evaluation indicator to be optimized.

5. The method according to claim 4, characterized in that The contribution rate of the work quality decline of the i-th comprehensive indicator of the multiple comprehensive indicators is calculated by the following formula: Among them, C i It is used to represent the contribution rate of the comprehensive index of the i-th measurement item to the decline in work quality, W i It is used to represent the influence weight of the comprehensive index of the i-th measurement item, γ i It is used to represent the actual grey relational value of the comprehensive index of the i-th measurement item, E i It is used to represent the expected value of the comprehensive index of the i-th measurement item, W k It is used to represent the influence weight of the kth comprehensive index of the multiple comprehensive indexes, γ k It is used to represent the actual grey relational value of the comprehensive index of the kth measurement item, E k It is used to represent the associated expected value of the kth comprehensive indicator of the measurement item, and m is used to represent the number of indicator types of the multiple comprehensive indicators of the measurement items.

6. The method according to claim 4, characterized in that For each comprehensive indicator of the measurement item to be optimized, the correlation contribution rate of the jth job evaluation indicator associated with the comprehensive indicator of the measurement item to be optimized is calculated using the following formula: Among them, G j It is used to represent the correlation contribution rate of the j-th job evaluation indicator, ξ j The grey correlation coefficient used to represent the j-th job evaluation index, ξ l It is used to represent the grey correlation coefficient of the lth job evaluation indicator associated with the comprehensive indicator of the measurement item to be optimized, and n is used to represent the number of indicator types of all job evaluation indicators associated with the comprehensive indicator of the measurement item to be optimized.

7. The method according to claim 1, characterized in that The robot operation knowledge graph records the logical associations between various reasons for the decline in operation quality, the causal associations between various reasons for the decline in operation quality and various operation evaluation indicators, and the mapping associations between various reasons for the decline in operation quality and various operation optimization suggestions. The step of invoking the robot operation knowledge graph to infer and generate the reasons for the decline in quality and the indicator optimization suggestions for all evaluation indicators to be optimized includes: For each evaluation indicator to be optimized, according to the logical association relationship and causal association relationship recorded in the robot operation knowledge graph, find the quality degradation reason matching the evaluation indicator to be optimized; For each quality degradation reason found, according to the mapping association relationship recorded in the robot operation knowledge graph, find the operation optimization suggestion matching the quality degradation reason as the indicator optimization suggestion for the evaluation indicator to be optimized.

8. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: Acquire task execution optimization data of the target service robot for the target service task based on the operation quality report; The robot operation knowledge graph is updated according to the task execution optimization data.

9. A computer device, characterized in that: It includes a processor and a memory, the memory stores a computer program that can be executed by the processor, and the processor can execute the computer program to implement the service robot operation quality assessment method based on knowledge graph as described in any one of claims 1-8.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a computer device, the service robot operation quality assessment method based on knowledge graph described in any one of claims 1 to 8 is implemented.

Images