Inspection route intelligent planning method and system based on component distribution completion

Abstract

The application discloses a method and system for intelligent planning of inspection routes based on component distribution completion, relates to the technical field of route intelligent planning, and is used for solving the problem of reduced task efficiency and coverage rate, constructing a virtual inspection scene by acquiring inspection area information, setting and receiving interactive instructions when a user enters simulation inspection, detecting the response time and inspection path of the instructions, judging the interactive state and screening and marking the user according to the path deviation rate and the response time, calling the initial route detection stress point distribution and the driving simulation speed, analyzing the route refining degree of the user, setting a monitoring time to collect behavior data, combining the interactive intention trend and the route refining degree to generate an adaptability index, calling the route corresponding to the adaptability level, and dynamically adjusting the route according to the user component operation difference, so that the inspection coverage rate and the execution efficiency are improved, the inspection dead angle and the redundant path are reduced, and efficient inspection of the inspection area is realized.

Description

Technical Field

[0001] This invention relates to the field of intelligent route planning technology, and more specifically, to a method and system for intelligent route planning based on component distribution completion. Background Technology

[0002] With the widespread application of intelligent inspection equipment such as drones and mobile robots, inspection route planning technology has developed rapidly in fields such as power line inspection, industrial equipment testing, and smart city management. Most existing inspection route planning methods rely on preset waypoints, fixed paths, or simple obstacle avoidance algorithms to ensure that inspection equipment can complete its tasks in designated areas.

[0003] The existing technology has the following shortcomings:

[0004] Currently, traditional routes are mostly set unilaterally by maintenance personnel, failing to dynamically optimize based on the operational characteristics, response speed, and interaction status of inspection personnel. Existing solutions fail to refine and adjust routes according to individual differences, easily leading to blind spots or redundant paths in inspections, resulting in reduced task efficiency and coverage, and discrepancies between route planning results and actual needs. Therefore, an intelligent route planning method and system based on component distribution completion is proposed.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an intelligent planning method and system for inspection routes based on component distribution completion, which solves the problems mentioned in the background art by using user interaction feature quantitative analysis, component distribution modeling and adaptability index generation technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent planning method for inspection routes based on component distribution completion, comprising the following steps:

[0008] Step S1: Obtain inspection area information, construct virtual inspection scenario. When the user enters the simulated inspection, set interaction commands and display them through the interactive interface. Detect the response time of the user to complete the interaction commands and the inspection task path. Calculate the path offset rate based on the inspection task path and judge the user's interaction status in combination with the response time.

[0009] Step S2: Based on the user's interaction status, filter and mark the user, call the initial inspection route, detect the distribution of force points of the marked user in the initial inspection route and the driving simulation speed of the simulation device, and analyze the refinement of the user's initial route by combining the distribution of force points and the driving simulation speed.

[0010] Step S3: Set a monitoring time, collect the behavior data of the marked user, analyze the interaction intention trend according to the behavior data of the marked user, and generate an adaptability index combined with the initial path refinement degree;

[0011] Step S4: Mark the adaptability level of the user by using the adaptability index, and call the inspection route corresponding to the adaptability level of the marked user.

[0012] In a preferred embodiment, in step S1, the inspection area information is obtained from the information database, the inspection area information includes the parameters of each component of the inspection tool and the position of the inspection area, modeling is performed through the parameters of each component of the inspection tool, and the virtual inspection scene is entered after combining the position of the inspection area;

[0013] When the user enters the simulated inspection, the simulated cockpit system sets a set of interaction instructions;

[0014] After the user receives the interaction instructions through the interaction interface, the user completes the corresponding task operation in the virtual inspection scene, and the time interval required by the user from receiving the instruction to completing the instruction is recorded by the high-precision sensor as the response time.

[0015] In a preferred embodiment, in step S1, the inspection task path of the user is tracked and compared in real time during the execution of the interaction instruction by the user, the inspection task path refers to the operation trajectory of the user in the simulated cockpit, the system collects and records the position and time sequence of each task operation point of the user through the high-precision sensor, draws the inspection task path of the user, and calculates the difference between the inspection task path of the user and the standard operation path as the path deviation rate;

[0016] After the response time and the path deviation rate are standardized, they are input into the logistic regression model as input features, and the probability value of the user's interaction state being efficient is calculated combined with the weight coefficient;

[0017] The probability value of the user's interaction state being efficient is compared with the preset state judgment threshold value, if the probability value of the user's interaction state being efficient is greater than or equal to the state judgment threshold value, it is determined that the user's interaction state is efficient; if the probability value of the user's interaction state being efficient is less than the state judgment threshold value, it is determined that the user's interaction state is inefficient.

[0018] In a preferred embodiment, in step S2, the user's interaction state is determined and classified for marking to screen out users with different inspection experience and perform targeted processing, and users with efficient interaction state are screened out and marked;

[0019] The initial inspection route is called, and the force sensor array is used to collect pressure values of each force point on the seat of the user, each pressure value representing the force of the user on the seat surface at a specific position, and a plurality of pressure values form a force sequence as the force point distribution of the user;

[0020] The driving simulation speed of the simulation tool is monitored and recorded in the initial inspection route according to a preset recording frequency, and the driving simulation speed is the running speed of the simulation tool when the user drives the simulation tool.

[0021] In a preferred embodiment, in step S2, the standard deviation of the force point distribution is calculated, and the average gradient difference thereof is calculated according to the change of the driving simulation speed, to obtain the acceleration change of the vehicle;

[0022] After the standard deviation of the force point distribution and the acceleration change of the driving simulation speed are standardized, the initial route refining degree of the user is evaluated by a weighted fusion method.

[0023] In a preferred embodiment, in step S3, the monitoring time is preset and evenly divided into a plurality of sampling time points, the behavior data of the user is collected, including the throttle pressure value, the brake pressure value, and the holding center coordinate of the hand on the steering wheel surface, and the pedal switching frequency and the holding stability coefficient are calculated according to the behavior data of the user.

[0024] The behavior data of the user is collected based on the preset initial inspection route;

[0025] The pressure sensor arranged below the throttle pedal and the brake pedal is used to obtain the throttle pressure value and the brake pressure value at each sampling time point, and the pressure change is obtained by subtracting the throttle pressure value from the brake pressure value at the same sampling time point and taking the absolute value.

[0026] If the pressure change exceeds a preset pressure change threshold, the corresponding sampling time point is marked as a switching time point, the number of switching time points in the monitoring time is counted as the pedal switching frequency, and the ratio of the pedal switching frequency to the monitoring time is taken as the pedal switching frequency.

[0027] In a preferred embodiment, in step S3, the holding center coordinate of the hand of the user on the steering wheel surface is collected by the tactile sensor, the average value of the holding center coordinates of the left hand and the right hand is taken as the holding center coordinate of the sampling time point, and the holding center coordinate of the sampling time point is the midpoint of the line connecting the holding center coordinates of the left hand and the right hand.

[0028] The holding center coordinate of the initial sampling time point is marked as a reference coordinate, the position distance between the reference coordinate and the holding center coordinate of each sampling time point is calculated by the Euclidean distance, and the average value of the position distances is taken as the holding stability coefficient.

[0029] The pedal switching frequency and the holding stability coefficient are normalized respectively, and then the hyperbolic tangent function is used to calculate the adaptability index.

[0030] In a preferred embodiment, in step S4, a first adaptability index threshold and a second adaptability index threshold are preset, and the first adaptability index threshold is greater than the second adaptability index threshold, and the adaptability index is compared to determine the adaptability level of the user;

[0031] If the adaptability index exceeds the first adaptability index threshold, the adaptability level of the user is determined to be the enhanced adaptability level;

[0032] If the adaptability index is between the first adaptability index threshold and the second adaptability index threshold, the adaptability level of the user is determined to be the regular adaptability level;

[0033] If the adaptability index is lower than the second adaptability index threshold, the adaptability level of the user is determined to be the primary adaptability level.

[0034] In a preferred embodiment, in step S4, the scene management module of the simulation driving system calls the inspection route corresponding to the adaptability level of the user.

[0035] The inspection route intelligent planning system based on component distribution completion includes an interaction detection module, a refining evaluation module, an intention analysis module, and a route matching module, and the functions of each module are as follows:

[0036] The interaction detection module is used to set an interaction instruction, the user receives the interaction instruction through an interaction interface, detects the response time of the user completing the interaction instruction and the inspection task path, calculates the path deviation rate according to the inspection task path, and judges the interaction state of the user in combination with the response time;

[0037] The refining evaluation module filters and marks the user according to the interaction state of the user, calls an initial inspection route, detects the stress point distribution of the marked user in the initial inspection route and the driving simulation speed of the simulation tool, analyzes the refining degree of the initial route in combination with the stress point distribution and the driving simulation speed, and marks the user;

[0038] The intention analysis module sets a monitoring time, collects behavior data of the marked user, analyzes the interaction intention trend according to the behavior data of the marked user, and generates an adaptability index in combination with the refining degree of the initial route;

[0039] The route matching module uses the adaptability index to determine the adaptability level of the marked user, and calls the inspection route corresponding to the adaptability level of the marked user.

[0040] Technical effects and advantages of the present application:

[0041] The application constructs a virtual inspection scene by acquiring inspection area information, sets an interactive instruction when a user enters the simulated inspection, the user receives the interactive instruction through an interactive interface, detects the response time of the user completing the interactive instruction and the inspection task instruction, calculates the path offset rate according to the inspection task path, judges the interactive state of the user in combination with the response time, marks the user according to the interactive state of the user, calls the initial inspection route, detects the stress point distribution of the marked user in the initial inspection route and the driving simulation speed of the simulation tool, analyzes the initial route refinement degree of the user by comprehensively analyzing the stress point distribution and the driving simulation speed, sets a monitoring time, collects the behavior data of the marked user, analyzes the interactive intention trend according to the behavior data of the marked user, generates the adaptability index in combination with the initial route refinement degree, calls the inspection route corresponding to the adaptability level of the marked user, dynamically adjusts the inspection route according to the component operation difference of different inspection users for the inspection tool, so as to realize efficient inspection of the inspection area, reduce the inspection dead angle and redundant path, and improve the inspection coverage rate and execution efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The method flowchart of the inspection route intelligent planning method based on component distribution completion of the application.

[0043] Figure 2 The module schematic diagram of the inspection route intelligent planning system based on component distribution completion of the application. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0045] The application obtains inspection area information, constructs a virtual inspection scene, sets an interactive instruction when a user enters simulated inspection, the user receives the interactive instruction through an interactive interface, detects the response time of the user completing the interactive instruction and an inspection task instruction, calculates a path deviation rate according to the inspection task path, judges the interactive state of the user in combination with the response time, marks the user according to the interactive state of the user, calls an initial inspection route, detects the stress point distribution of the marked user in the initial inspection route and the driving simulation speed of the simulation tool, analyzes the initial route refinement degree of the user in combination with the stress point distribution and the driving simulation speed, sets a monitoring time, collects the behavior data of the marked user, analyzes the interactive intention trend according to the behavior data of the marked user, generates an adaptability index in combination with the initial route refinement degree, calls the inspection route corresponding to the adaptability level of the marked user, and dynamically adjusts the inspection route according to the component operation difference of different inspection users for the inspection tool, so as to realize efficient inspection of the inspection area.

[0046] Embodiment 1, an inspection route intelligent planning method based on component distribution completion, as shown in the figure, includes the following steps: Figure 1

[0047] Step S1: Obtain inspection area information, construct a virtual inspection scene, set an interactive instruction when a user enters simulated inspection, and display the interactive instruction through an interactive interface, detect the response time of the user completing the interactive instruction and an inspection task path, calculate a path deviation rate according to the inspection task path, and judge the interactive state of the user in combination with the response time;

[0048] Step S2: Mark the user according to the interactive state of the user, call an initial inspection route, detect the stress point distribution of the marked user in the initial inspection route and the driving simulation speed of the simulation tool, and analyze the initial route refinement degree of the user in combination with the stress point distribution and the driving simulation speed;

[0049] Step S3: Set a monitoring time, collect the behavior data of the marked user, analyze the interactive intention trend according to the behavior data of the marked user, and generate an adaptability index in combination with the initial route refinement degree;

[0050] Step S4: Mark the adaptability level of the user by using the adaptability index, and call the inspection route corresponding to the adaptability level of the marked user.

[0051] The specific implementation is as follows:

[0052] In step S1, access an information database to obtain inspection area information, the inspection area information includes component parameters of an inspection tool and inspection area positions, model through the component parameters of the inspection tool, and enter a virtual inspection scene in combination with the inspection area positions;

[0053] ​When the user enters the simulation inspection, the simulation cockpit system sets a set of interactive instructions, the interactive instructions being a set of predefined operation steps, including steering wheel operation, pedal control, and instrument panel viewing, etc.

[0054] After the user receives the interactive instructions through the interactive interface, the user completes the corresponding task operation in the virtual inspection scene, and the system records the time interval required for the user to receive the instructions to complete the instructions through the high-precision sensor, taking the time interval as the response time. The shorter the response time, the faster the user's reaction to the interactive instructions, and the more efficient the user's interaction state.

[0055] During the execution of the interactive instructions by the user, the user's inspection task path is tracked and compared in real time. The inspection task path refers to the operation trajectory of the user in the simulation cockpit. The system collects and records the position and time sequence of each task operation point of the user through the high-precision sensor, and draws the inspection task path of the user. At the same time, the standard operation path is called. The standard operation path is the user's operation trajectory in the ideal state based on the pre-set driving task standard set by the system, which contains a set of standard operation points.

[0056] It should be noted that the information database is a system for storing, managing, and retrieving information, which in this case is used to call the inspection area information. The high-precision sensor refers to a sensor that can provide very small errors and high-resolution output within its measurement range, accurately capturing and quantifying changes in physical or chemical quantities such as temperature, pressure, displacement, etc.

[0057] The difference between the user's inspection task path and the standard operation path is calculated as the path deviation rate, and the specific calculation formula is as follows:

[0058] ;

[0059] Wherein, is the path deviation rate, is the user operation position at the i-th time point, is the standard operation position at the i-th time point, and n is the total number of steps of the inspection task path. The smaller the path deviation rate, the higher the matching degree between the inspection task path and the standard operation path, and the more efficient the user's interaction state. The larger the path deviation rate, the greater the operation deviation between the inspection task path and the standard operation path, and the less efficient the user's interaction state.

[0060] After standardizing the response time and the path deviation rate, they are input into the logistic regression model as input features. Combined with the weight coefficient, the probability value of the user's interaction state being efficient is calculated through the Sigmoid function, and a threshold is set for binary classification judgment.

[0061] Specifically, the specific calculation formula of the logistic regression model is:

[0062] ;

[0063] wherein, is the probability value of the user's interaction state being efficient, is the bias term, and are the weight coefficients corresponding to the response time and path offset rate respectively, is the response time.

[0064] It should be noted that the logistic regression is a statistical and machine learning model widely used in binary classification problems, which is used to predict a binary result, i.e., to predict the probability of an event occurring; the sigmoid function is an activation function used in machine learning and statistics, which maps any real number to between 0 and 1; the weight coefficients in the logistic regression model are obtained by training and optimizing the historical data by the maximum likelihood estimation method, which makes the value of the parameter maximize the probability of observing data given the observed data; the standardization methods include but are not limited to standard linear transformation based on interval scaling, Z-Score standardization method based on statistics, or normalization method based on nonlinear mapping function, and the application method of the standardization processing is not described here.

[0065] The probability value of the user's interaction state being efficient is compared with the preset state determination threshold value, if the probability value of the user's interaction state being efficient is greater than or equal to the state determination threshold value, it is determined that the user's interaction state is efficient; if the probability value of the user's interaction state being efficient is less than the state determination threshold value, it is determined that the user's interaction state is inefficient.

[0066] It should be noted that the state determination threshold value refers to a probability threshold value set by the model when classifying the user interaction state, which is used to convert the probability value output by the model into a specific classification result. When training the logistic regression model, the best binary classification effect is found through cross-validation as the state determination threshold value, for example: the probability value of the user's interaction state being efficient calculated by the logistic regression model is 0.75, if the preset state determination threshold value is 0.7, the user's interaction state is determined to be efficient.

[0067] In step S2, the user interaction state is determined and classified for screening users with different inspection experience and targeted processing, users with efficient interaction state are screened and marked, and the initial inspection route is called for further analysis.

[0068] The setting of the initial inspection route includes the basic environment and task setting of the simulation driving, and ensures that the basic behavior mode of the user in the initial driving environment can be reflected. The specific parameters of the initial inspection route include, but are not limited to, driving road conditions, vehicle types, road curves, etc., and are uniformly loaded through system preset parameters.

[0069] In the initial inspection route, the force sensor array is used to collect the pressure values of the marking force points of the user on the seat. Each pressure value represents the force of the user on the surface of the seat at a specific position, and a plurality of pressure values form a force sequence as the force point distribution of the user.

[0070] It should be noted that the force sensor array is a structure composed of a plurality of pressure sensors arranged in a two-dimensional or one-dimensional array, which is used to detect and record the pressure distribution of the surface or specific area of an object. The force of the entire surface is formed by collecting the pressure data of each pressure sensor point.

[0071] The driving simulation speed of the simulation tool is monitored and recorded in the initial inspection route according to the preset recording frequency. The driving simulation speed is the running speed of the simulation tool when the user drives the simulation tool.

[0072] The standard deviation of the force point distribution is calculated to measure the uniformity and pressure fluctuation of the force point distribution. The standard deviation of the force point distribution is calculated as follows:

[0073] ;

[0074] wherein, is the standard deviation of the force point distribution, is the i-th pressure value in the force point distribution, is the mean value of the pressure values in the force point distribution, and n is the total number of pressure values in the force point distribution.

[0075] The standard deviation of the force point distribution reflects the dispersion degree of the force point pressure distribution. The smaller the value, the more uniform the force point distribution, the stronger the user's balance control ability, and the higher the user's refinement degree of the initial route. On the contrary, the larger the value, the more uneven the force point pressure distribution, the greater the user's balance control problem, and the lower the user's refinement degree of the initial route.

[0076] The average gradient difference of the driving simulation speed is calculated according to the change of the driving simulation speed, and the acceleration change amount of the vehicle is obtained, so as to evaluate the balance of the user's control of the vehicle. The calculation formula of the acceleration change amount is:

[0077] ;

[0078] wherein, is the acceleration change amount, and Driving simulation speed at two continuous time points, m is the total number of collected driving simulation speeds, when the simulation tool is stationary, i.e. and The speed difference at this time point is not counted in the statistics to avoid misjudgment of acceleration fluctuations in the stationary state.

[0079] The acceleration change quantity is used to quantify the volatility of the driving simulation speed, reflecting the user's acceleration and deceleration ability at different speed stages. The smaller the acceleration change quantity, the more stable the user controls the vehicle, and the stronger the balance control ability. The larger the acceleration change quantity, the worse the user's control ability in the driving process, and the weaker the balance control ability.

[0080] After standardizing the standard deviation of the force point distribution and the acceleration change quantity of the driving simulation speed, the initial route refinement degree of the user is evaluated by a weighted fusion method, and the calculation formula is as follows:

[0081] ;

[0082] wherein, is the initial route refinement degree, satisfying , and is the weighted coefficient, satisfying is used to adjust the influence degree of the standard deviation of the force point distribution and the acceleration change quantity on the final result. The lower the initial route refinement degree of the user, the stronger the balance control ability of the user. The higher the initial route refinement degree of the user, the weaker the balance control ability of the user.

[0083] It should be noted that the weighted coefficient refers to the coefficient allocated to different parameters in the weighted fusion, which aims to adjust the contribution degree of each parameter to the final evaluation result, and is adjusted and determined by an optimization algorithm. For example: after standardizing the standard deviation of the force point distribution and the acceleration change quantity of the driving simulation speed, 0.8 and 0.6 are obtained respectively, and the weighted coefficients are set to 0.7 and 0.3 respectively. Then the initial route refinement degree can be calculated as: .

[0084] In step S3, the preset monitoring time is evenly divided into multiple sampling time points, and the behavior data of the labeled user including the throttle pressure value, the brake pressure value and the hand holding center coordinate on the steering wheel surface are collected. The pedal switching frequency and the holding stability coefficient are calculated according to the behavior data of the labeled user;

[0085] Based on the preset initial inspection route, the behavior data of the labeled user is collected, which is used to evaluate the perception response and operation rhythm of the labeled user to the basic simulation driving task;

[0086] ​​The preset initial inspection route refers to a driving simulation environment with low complexity traffic elements and single road conditions, which is set by professionals to collect the behavior data of the marked user on the basis of control variables.

[0087] The pressure sensor arranged below the accelerator pedal and the brake pedal is used to obtain the accelerator pressure value and the brake pressure value at each sampling time, and the pressure change amount is obtained by taking the absolute value of the difference between the accelerator pressure value and the brake pressure value at the same sampling time.

[0088] If the pressure change amount exceeds the preset pressure change threshold, the corresponding sampling time is marked as a switching time, the number of switching times in the monitoring time is counted as the pedal switching frequency, and the ratio of the pedal switching frequency to the monitoring time is taken as the pedal switching frequency.

[0089] The pedal switching frequency reflects the state switching rhythm and control strategy change of the marked user during the simulation driving process. In the preset primary driving scene, the higher the pedal switching frequency, the more frequent the marked user adjusts the accelerator and brake control, and the higher the interactive intention trend. The lower the pedal switching frequency, the more stable the marked user operates, and the lower the interactive intention trend.

[0090] The touch sensor is used to collect the holding center coordinates of the marked user's hands on the surface of the steering wheel. The average value of the left and right hand holding center coordinates is taken as the holding center coordinates at the sampling time. The holding center coordinates at the sampling time are the midpoints of the lines connecting the left and right hand holding center coordinates, reflecting the overall control center of gravity of the user.

[0091] The holding center coordinates at the initial sampling time are marked as the reference coordinates, and the position distance between the reference coordinates and the holding center coordinates at each sampling time is calculated by the Euclidean distance: wherein, is the reference coordinate, is the holding center coordinates at the i-th sampling time, is the position distance at the i-th sampling time. The average value of each position distance is taken as the holding stability coefficient.

[0092] The larger the holding stability coefficient, the more unstable the marked user's hand operation, and the more unstable the interactive intention. The smaller the holding stability coefficient, the more stable the marked user's hand operation, and the more clear and coherent the interactive intention.

[0093] The pedal switching frequency and the holding stability coefficient are standardized and mapped to the interval of 0 to 1, and the interactive intention trend is obtained by taking the difference between the standardized pedal switching frequency and the holding stability coefficient.

[0094] The interaction intention trend is used to reflect the interaction stability and scene adaptability of the marked user. Based on the complementary relationship between the pedal switching frequency and the holding stability coefficient in the dynamic interaction and operation stability of the marked user, the difference between them is used to reflect the interaction intention trend of the user;

[0095] The interaction intention trend and the initial route refinement degree are substituted into the hyperbolic tangent function to calculate the adaptability index: wherein, is the interaction intention trend, is the initial route refinement degree, , are preset influence weights corresponding to the interaction intention trend and the initial route refinement degree respectively, and the value range is 0 to 1, is the adaptability index;

[0096] The greater the initial route refinement degree, the more unstable the balance rhythm of the marked user, the smaller the adaptability index, and the greater the interaction intention trend, which means that the interaction action of the marked user is active, and the adaptability index is greater;

[0097] wherein, The function has a nonlinear compression characteristic, which can enhance the discriminant sensitivity at the critical point, and the implementation The function value range is normalized from [-1, 1] to [0, 1]. The preset influence weight is set by a professional person and is used to reflect the influence degree of the interaction intention trend and the initial route refinement degree on the adaptability index.

[0098] It should be noted that the preset monitoring time is used to limit the time range for collecting the behavior data of the marked user. After the analysis of the initial route refinement degree is completed, the preset initial inspection route is set, for example, the preset monitoring time is 60 seconds and is evenly divided into 60 sampling time points; the pressure sensor is a sensor device for real-time sensing of the pressure applied by the stepping action; the preset pressure change threshold is a threshold value set by a professional person according to the statistical results and experience of a large number of pressure change amounts, which is used to distinguish the pressure change limit value of whether the marked user performs the pedal switching operation; the tactile sensor is an electronic device capable of sensing and measuring physical contact information, which is used to detect the position coordinates of the hands of the marked user in real time; the Euclidean distance is a standard measurement method for measuring the straight line distance between two points in space, which is used to calculate the position distance of the hand holding center coordinates of the marked user relative to the reference coordinates at different sampling time points.

[0099] In step S4, the first adaptability index threshold and the second adaptability index threshold are preset, and the first adaptability index threshold is greater than the second adaptability index threshold. The adaptability index is compared to determine the adaptability level of the user.

[0100] If the adaptability index exceeds the first adaptability index threshold, the user's adaptability level is determined to be an enhanced adaptability level;

[0101] If the adaptability index is between the first adaptability index threshold and the second adaptability index threshold, the user's adaptability level is determined to be a regular adaptability level;

[0102] If the adaptability index is below the second adaptability index threshold, the user's adaptability level is determined to be a primary adaptability level;

[0103] The scene management module of the simulation driving system calls the inspection route corresponding to the user's adaptability level;

[0104] The scene management module of the simulation driving system refers to a functional component for calling the inspection route according to the user's adaptability level and rendering a driving picture in real time,

[0105] The division of the inspection route is divided into multiple inspection levels by professionals in the field according to the complexity of the inspection task and the degree of interactive challenge, which matches the adaptability level of the user, and will not be analyzed here.

[0106] It should be noted that the first adaptability index threshold and the second adaptability index threshold are used as the basis for classification and judgment of the user's adaptability index. Professionals can obtain or set values based on actual application requirements according to historical adaptability index sample analysis, for example, calculating the mean and standard deviation of the historical adaptability index sample, summing the mean and standard deviation to obtain the first adaptability index threshold, and subtracting the mean and standard deviation to obtain the second adaptability index threshold.

[0107] Embodiment 2, an inspection route intelligent planning system based on component distribution completion, as shown in Figure 2 The modules have the following functions:

[0108] The interaction detection module is used to set interaction instructions, and the user receives the interaction instructions through the interaction interface, detects the response time of the user completing the interaction instructions and the inspection task path, calculates the path deviation rate according to the inspection task path, and judges the user's interaction state in combination with the response time;

[0109] The refining evaluation module filters and marks the user according to the user's interaction state, calls the initial inspection route, detects the stress point distribution of the marked user in the initial inspection route and the driving simulation speed of the simulation tool, analyzes the refining degree of the initial route in combination with the stress point distribution and the driving simulation speed, and marks the user;

[0110] The intention analysis module sets a monitoring time, collects behavior data of the marked user, analyzes an interaction intention trend according to the behavior data of the marked user, and generates an adaptability index in combination with an initial route refinement degree;

[0111] The route matching module judges an adaptability level of the marked user by using the adaptability index, and retrieves an inspection route corresponding to the adaptability level of the marked user.

[0112] Finally, it should be noted that the relationship terms, such as first and second, are used merely to distinguish one entity or action from another entity or action, without necessarily requiring or implying any actual relationship or order between such entities or actions.

[0113] Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover a non-exclusive inclusion, such that processes, methods, articles, or apparatuses that include a list of elements not only include those elements, but also include other elements not expressly listed or inherent to such processes, methods, articles, or apparatuses. Without further limitation, an element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0114] In this document, the singular forms "a", "an" and "the" can also include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "comprise / comprising" or "have / having" specifies the presence of stated features, integers, steps, operations, components, parts, or combinations thereof, but does not exclude the presence or addition of one or more other features, integers, steps, operations, components, parts, or combinations thereof, the possibility of which exists simultaneously. The term "and / or" used in this specification includes any and all combinations of the associated listed items.

[0115] The various embodiments in the specification are described in a progressive manner, each embodiment focusing on the differences from other embodiments, and the various embodiments can be combined as needed, and the same and similar parts are cross-referenced.

[0116] The above description of the disclosed embodiments enables those skilled in the art to implement or use the various modifications of the embodiments of the present application, and it will be apparent to those skilled in the art that various modifications can be made to the embodiments of the present application without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for intelligent planning of inspection routes based on human-computer interaction behavior evaluation, characterized in that: The method comprises the following steps: Step S1: Obtain the inspection area information, construct a virtual inspection scene, set interaction instructions when the user enters the simulation inspection, and display through the interaction interface, detect the response time of the user completing the interaction instructions and the inspection task path, calculate the path deviation rate according to the inspection task path, and judge the user interaction state according to the response time; Step S2: According to the interaction state of the user, the user is screened and marked, the initial inspection route is called, the stress point distribution of the marked user in the initial inspection route is detected, and the driving simulation speed of the simulation tool is detected, and the initial route refinement degree of the user is analyzed by comprehensively analyzing the stress point distribution and the driving simulation speed; In step S2, the initial inspection route is called, the pressure values of each stress point of the marked user on the seat are collected through the stress sensor array, each pressure value represents the stress of the user at a specific position on the seat surface, and a plurality of pressure values form a stress sequence as the stress point distribution of the user; In step S2, the standard deviation of the stress point distribution is calculated, and the average gradient difference is calculated according to the change of the driving simulation speed, and the acceleration change amount of the vehicle is obtained; After standardizing the standard deviation of the stress point distribution and the acceleration change amount of the driving simulation speed, the initial route refinement degree of the user is evaluated by a weighted fusion method; Step S3: Set the monitoring time, collect the behavior data of the marked user, analyze the interaction intention trend according to the behavior data of the marked user, and generate an adaptability index combined with the initial path refinement degree; In step S3, the monitoring time is preset and evenly divided into a plurality of sampling time points, the behavior data of the marked user includes the throttle pressure value, the brake pressure value and the holding center coordinate of the hand on the steering wheel surface, the pedal switching frequency and the holding stability coefficient are calculated according to the behavior data of the marked user; The behavior data of the marked user is collected based on the preset initial inspection route; In step S3, the holding center coordinate of the hand of the marked user on the steering wheel surface is collected through the tactile sensor, the average value of the left hand and right hand holding center coordinates is taken as the holding center coordinate of the sampling time point, and the holding center coordinate of the sampling time point is the midpoint of the line connecting the left hand and right hand holding center coordinates; The holding center coordinate of the initial sampling time point is marked as the reference coordinate, the position distance between the reference coordinate and the holding center coordinate of each sampling time point is calculated by the Euclidean distance, and the average result of each position distance is taken as the holding stability coefficient; After standardizing the pedal switching frequency and the holding stability coefficient respectively, the adaptability index is calculated by the hyperbolic tangent function; Step S4: Mark the adaptability level of the user by using the adaptability index, and call the inspection route corresponding to the adaptability level of the marked user.

2. The inspection route intelligent planning method based on human-computer interaction behavior evaluation according to claim 1, wherein: In step S1, access the information database to obtain the inspection area information, and the inspection area information includes the parameters of each component of the inspection tool and the position of the inspection area, model the parameters of each component of the inspection tool, and enter the virtual inspection scene combined with the position of the inspection area; When the user enters the simulation inspection, the simulation cockpit system sets a group of interaction instructions; The user receives an interactive instruction through an interactive interface, and completes a corresponding task operation in a virtual inspection scene. A high-precision sensor is used to record a time interval required by the user from receiving the instruction to completing the instruction, and the time interval is taken as a response time.

3. The intelligent planning method of the inspection route based on the human-computer interaction behavior evaluation according to claim 2, characterized in that: In step S1, the inspection task path of the user is tracked and compared in real time during the execution of the interactive instruction by the user. The inspection task path refers to the operation trajectory of the user in the simulated cockpit. The system collects and records the position and time sequence of each task operation point of the user through a high-precision sensor, draws the inspection task path of the user, and calculates the difference between the inspection task path of the user and the standard operation path as the path deviation rate. The response time and the path deviation rate are standardized and taken as input features of a logistic regression model. Combined with a weight coefficient, the probability value of the interactive state of the user being efficient is calculated. The probability value of the interactive state of the user being efficient is compared with a preset state judgment threshold. If the probability value of the interactive state of the user being efficient is greater than or equal to the state judgment threshold, it is determined that the interactive state of the user is efficient. If the probability value of the interactive state of the user being efficient is less than the state judgment threshold, it is determined that the interactive state of the user is inefficient.

4. The intelligent planning method of the inspection route based on the human-computer interaction behavior evaluation according to claim 1, characterized in that: In step S2, the interactive state of the user is determined and classified for screening users with different inspection experience and targeted processing. Users with an efficient interactive state are screened and marked. In step S3, the driving simulation speed of the simulation tool is monitored and recorded at a preset recording frequency in the initial inspection route. The driving simulation speed is the driving speed of the simulation tool when the user drives the simulation tool.

5. The intelligent planning method of the inspection route based on the human-computer interaction behavior evaluation according to claim 1, characterized in that: In step S3, pressure sensors arranged below the accelerator pedal and the brake pedal are used to obtain the accelerator pressure value and the brake pressure value at each sampling time. The absolute value of the difference between the accelerator pressure value and the brake pressure value at the same sampling time is taken as the pressure change. If the pressure change exceeds a preset pressure change threshold, the corresponding sampling time is marked as a switching time. The number of switching times within the monitoring time is taken as the pedal switching frequency, and the ratio of the pedal switching frequency to the monitoring time is taken as the pedal switching frequency.

6. The intelligent planning method of the inspection route based on the human-computer interaction behavior evaluation according to claim 1, characterized in that: In step S4, a first adaptation index threshold and a second adaptation index threshold are preset, and the first adaptation index threshold is greater than the second adaptation index threshold. The adaptation force index is compared with the first adaptation index threshold and the second adaptation index threshold to determine the adaptation level of the user. If the adaptation force index exceeds the first adaptation index threshold, the adaptation level of the user is determined to be the enhanced adaptation level. If the adaptation force index is between the first adaptation index threshold and the second adaptation index threshold, the adaptation level of the user is determined to be the regular adaptation level. If the adaptability index is lower than the second adaptability threshold, the user's adaptability level is determined as a primary adaptability level.

7. The intelligent planning method of inspection route based on human-computer interaction behavior evaluation according to claim 6, characterized in that: In step S4, the scenario management module of the simulation driving system calls the inspection route corresponding to the adaptability level of the user.

8. The intelligent planning system of inspection route based on human-computer interaction behavior evaluation, for realizing the intelligent planning method of inspection route based on human-computer interaction behavior evaluation according to any one of claims 1-7, characterized in that: The interactive detection module is used to set interactive instructions, and the user receives the interactive instructions through the interactive interface, detects the response time of the user completing the interactive instructions and the inspection task path, calculates the path deviation rate according to the inspection task path, and judges the interactive state of the user in combination with the response time. The refining evaluation module marks the user according to the interactive state of the user, calls the initial inspection route, detects the stress point distribution of the marked user in the initial inspection route and the driving simulation speed of the simulation tool, analyzes the refining degree of the initial route in combination with the stress point distribution and the driving simulation speed, and marks the user. The stress sensor array is used to collect the pressure values of each stress point of the marked user on the seat, and each pressure value represents the stress of the user at a specific position on the surface of the seat. The standard deviation of the stress point distribution and the average gradient difference of the driving simulation speed are calculated, and the acceleration change of the vehicle is obtained. The standard deviation of the stress point distribution and the acceleration change of the driving simulation speed are standardized, and the initial route refining degree of the user is evaluated by a weighted fusion method. The intention analysis module sets a monitoring time, collects the behavior data of the marked user, analyzes the interactive intention trend according to the behavior data of the marked user, and generates an adaptability index in combination with the initial route refining degree. The monitoring time is preset and evenly divided into multiple sampling time points, and the behavior data of the marked user including the throttle pressure value, the brake pressure value and the holding center coordinate of the hand on the steering wheel surface are collected. The behavior data of the marked user are collected based on the preset initial inspection route. The holding center coordinate of the hand on the steering wheel surface is collected by the tactile sensor, and the average value of the left and right holding center coordinates is taken as the holding center coordinate of the sampling time point. The holding center coordinate of the initial sampling time point is marked as the reference coordinate, and the position distance between the reference coordinate and the holding center coordinate of each sampling time point is calculated by the Euclidean distance. The average result of each position distance is taken as the holding stability coefficient. The pedal switching frequency and the holding stability coefficient are standardized respectively, and the adaptability index is calculated by the hyperbolic tangent function. The route matching module uses the adaptability index to determine the adaptability level of the marked user, and calls the inspection route corresponding to the adaptability level of the marked user.

Images