A method, device, equipment and medium for determining an observation position of a patrol robot

Abstract

This application discloses a method, apparatus, device, and medium for determining the observation position of an inspection robot. The method includes: determining the location and type of anomalies based on global perception information of the inspection area; generating multiple candidate observation positions based on the anomaly locations and determining the observation evaluation information and observation effectiveness parameters of each candidate observation position; identifying whether there are candidate observation positions whose observation coverage exceeds a preset coverage threshold; if so, determining the candidate observation position corresponding to the maximum observation effectiveness parameter as the target observation position; if not, iteratively calculating the comprehensive observation coverage increment and time-related attenuation cost of each candidate observation position, and successively selecting candidate observation positions as target observation positions until the comprehensive observation coverage of all target observation positions exceeds the preset coverage threshold. This technical solution dynamically and adaptively determines the observation position based on the anomaly location and anomaly type, improving the inspection robot's comprehensive observation capability of anomalies.

Description

Technical Field

[0001] This application belongs to the field of electronic digital data processing technology, specifically relating to a method, device, equipment, and medium for determining the observation position of an inspection robot. Background Technology

[0002] Inspection robots can use autonomous movement and sensing devices to observe anomalies in the inspection area at close range, enabling accurate anomaly detection and assessment. The appropriate selection of the observation location directly affects the robot's accuracy in anomaly identification and the quality of the observed information.

[0003] Currently, the commonly used method for determining observation locations involves pre-setting multiple fixed observation locations within the inspection area. When an anomaly occurs, the distance between each fixed observation location and the anomaly location is calculated, and the nearest fixed observation location is selected as the target observation location. However, the pre-set fixed observation locations are not specifically generated for the current anomaly situation, which can easily lead to the inspection robot being unable to fully observe the anomaly and collecting insufficient observation information, thereby reducing the accuracy of anomaly detection and assessment. Summary of the Invention

[0004] This application provides a method, device, equipment, and medium for determining the observation position of an inspection robot, with the aim of dynamically and adaptively determining the observation position based on the abnormal location and type, thereby improving the inspection robot's comprehensive observation capability of abnormalities.

[0005] In a first aspect, embodiments of this application provide a method for determining the observation position of an inspection robot, the method comprising: Step 1: Obtain global perception information of the inspection area, and determine the location and type of anomalies based on the global perception information; Step 2: Generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage; Step 3: Identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; if so, proceed to step 4; if not, proceed to step 5. Step 4: Among the candidate observation locations where the observation coverage exceeds the preset coverage threshold, the candidate observation location corresponding to the maximum observation effectiveness parameter is determined as the target observation location; Step 5: Iteratively calculate the comprehensive observation coverage increment and the time-related attenuation cost of each candidate observation location, and select candidate observation locations as target observation locations one by one according to the comprehensive observation coverage increment and the time-related attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold.

[0006] Furthermore, step 5 includes: Step 51: Calculate the incremental comprehensive observation coverage for each remaining candidate observation location based on the current target observation location set; Step 52: Calculate the path length corresponding to each of the remaining candidate observation locations based on the current target observation location set, and calculate the time decay cost corresponding to each of the remaining candidate observation locations based on the path length. Step 53: Substitute the comprehensive observation coverage increment and the time-degradation cost into the pre-constructed evaluation function to obtain the comprehensive evaluation parameters, and determine the candidate observation position corresponding to the maximum comprehensive evaluation parameter as the target observation position, and add the target observation position to the current target observation position set; Step 54: Identify whether the overall observation coverage of the current target observation location set exceeds the preset coverage threshold; if yes, end the process; if no, repeat steps 51 to 54.

[0007] Furthermore, the step of calculating the time-delay cost corresponding to each remaining candidate observation location based on the path length includes: The timeliness requirement parameters for each candidate observation location are determined based on the anomaly type, and the timeliness prediction parameters for each candidate observation location are calculated based on the path length. For each candidate observation location, the ratio of the timeliness prediction parameter to the timeliness requirement parameter of the candidate observation location is calculated as the timeliness attenuation cost of the candidate observation location.

[0008] Furthermore, determining the lead time requirement parameters for each candidate observation location based on the anomaly type includes: Determine the anomaly propagation rate and the degree of impact on the inspection robot based on the anomaly type; The remaining tolerable time for each candidate observation location is calculated based on the anomalous diffusion rate. Divide the remaining tolerable duration of each candidate observation location by the aforementioned impact degree parameter to obtain the timeliness requirement parameter for each candidate observation location.

[0009] Furthermore, the step of calculating the comprehensive observation coverage increment corresponding to each remaining candidate observation location based on the current target observation location set includes: Based on the abnormal location, determine the effective observation angle range of each target observation location in the current target observation location set and the effective observation angle range of each remaining candidate observation location; The comprehensive observation coverage of the current target observation location set is calculated based on the effective observation angle range of each target observation location in the current target observation location set; Based on the effective observation angle intervals of each target observation position in the current target observation position set and the effective observation angle intervals of each remaining candidate observation position, the effective observation angle interval of the target is determined, and the comprehensive observation coverage of the target is calculated based on the effective observation angle interval of the target. Subtract the overall observation coverage of the current set of target observation locations from the overall observation coverage of each candidate observation location to obtain the overall observation coverage increment for each candidate observation location.

[0010] Furthermore, before calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set, the method further includes: Determine the corresponding preset angle intervals and the angle weights corresponding to each preset angle interval based on the anomaly type. Accordingly, calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set includes: Based on the effective observation angle intervals of each target observation position in the current target observation position set and each preset angle interval, calculate the partition observation coverage of each preset angle region respectively; Based on the angle weights corresponding to each preset angle interval, the weighted summation of the observation coverage of each preset angle region is calculated to obtain the comprehensive observation coverage of the current target observation position set.

[0011] Furthermore, the observation and evaluation information also includes occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters; Accordingly, step 2 includes: Step 21: Generate multiple candidate observation locations based on the anomaly location and the inspection area; Step 22: Obtain environmental information of the inspection area and the types of sensors included in the inspection robot; Step 23: Determine the observation evaluation information for each candidate observation location based on the anomaly location, the anomaly type, the environmental information, and the sensor types; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, observation coverage, occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters. Step 24: Determine the observation validity parameters for each candidate observation location based on the observation evaluation information and the anomaly type.

[0012] Secondly, embodiments of this application provide an observation position determination device for an inspection robot, the device comprising: The anomaly detection module is used to acquire global perception information of the inspection area and determine the location and type of anomalies based on the global perception information. The observation and evaluation module is used to generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage; The coverage identification module is used to identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; The first determining module is used to determine the candidate observation position corresponding to the maximum observation effectiveness parameter as the target observation position when there are candidate observation positions where the observation coverage exceeds the preset coverage threshold. The second determining module is used to iteratively calculate the comprehensive observation coverage increment and the time-related decay cost of each candidate observation location when there are no candidate observation locations whose observation coverage exceeds the preset coverage threshold, and select candidate observation locations as target observation locations one by one according to the comprehensive observation coverage increment and the time-related decay cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold.

[0013] Furthermore, the second determining module is specifically used for: Step 51: Calculate the incremental comprehensive observation coverage for each remaining candidate observation location based on the current target observation location set; Step 52: Calculate the path length corresponding to each of the remaining candidate observation locations based on the current target observation location set, and calculate the time decay cost corresponding to each of the remaining candidate observation locations based on the path length. Step 53: Substitute the comprehensive observation coverage increment and the time-degradation cost into the pre-constructed evaluation function to obtain the comprehensive evaluation parameters, and determine the candidate observation position corresponding to the maximum comprehensive evaluation parameter as the target observation position, and add the target observation position to the current target observation position set; Step 54: Identify whether the overall observation coverage of the current target observation location set exceeds the preset coverage threshold; if yes, end the process; if no, repeat steps 51 to 54.

[0014] Furthermore, the step of calculating the time-delay cost corresponding to each remaining candidate observation location based on the path length includes: The timeliness requirement parameters for each candidate observation location are determined based on the anomaly type, and the timeliness prediction parameters for each candidate observation location are calculated based on the path length. For each candidate observation location, the ratio of the timeliness prediction parameter to the timeliness requirement parameter of the candidate observation location is calculated as the timeliness attenuation cost of the candidate observation location.

[0015] Furthermore, determining the lead time requirement parameters for each candidate observation location based on the anomaly type includes: Determine the anomaly propagation rate and the degree of impact on the inspection robot based on the anomaly type; The remaining tolerable time for each candidate observation location is calculated based on the anomalous diffusion rate. Divide the remaining tolerable duration of each candidate observation location by the aforementioned impact degree parameter to obtain the timeliness requirement parameter for each candidate observation location.

[0016] Furthermore, the step of calculating the comprehensive observation coverage increment corresponding to each remaining candidate observation location based on the current target observation location set includes: Based on the abnormal location, determine the effective observation angle range of each target observation location in the current target observation location set and the effective observation angle range of each remaining candidate observation location; The comprehensive observation coverage of the current target observation location set is calculated based on the effective observation angle range of each target observation location in the current target observation location set; Based on the effective observation angle intervals of each target observation position in the current target observation position set and the effective observation angle intervals of each remaining candidate observation position, the effective observation angle interval of the target is determined, and the comprehensive observation coverage of the target is calculated based on the effective observation angle interval of the target. Subtract the overall observation coverage of the current set of target observation locations from the overall observation coverage of each candidate observation location to obtain the overall observation coverage increment for each candidate observation location.

[0017] Furthermore, before calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set, the method further includes: Determine the corresponding preset angle intervals and the angle weights corresponding to each preset angle interval based on the anomaly type. Based on the effective observation angle intervals of each target observation position in the current target observation position set and each preset angle interval, calculate the partition observation coverage of each preset angle region respectively; Based on the angle weights corresponding to each preset angle interval, the weighted summation of the observation coverage of each preset angle region is calculated to obtain the comprehensive observation coverage of the current target observation position set.

[0018] Furthermore, the observation and evaluation information also includes occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters; Accordingly, the observation and evaluation module is specifically used for: Step 21: Generate multiple candidate observation locations based on the anomaly location and the inspection area; Step 22: Obtain environmental information of the inspection area and the types of sensors included in the inspection robot; Step 23: Determine the observation evaluation information for each candidate observation location based on the anomaly location, the anomaly type, the environmental information, and the sensor types; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, observation coverage, occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters. Step 24: Determine the observation validity parameters for each candidate observation location based on the observation evaluation information and the anomaly type.

[0019] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the method described in the first aspect.

[0020] Fourthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the method described in the first aspect.

[0021] In this embodiment, step 1 involves acquiring global perception information of the inspection area and determining the location and type of anomalies based on the global perception information; step 2 involves generating multiple candidate observation locations based on the anomaly locations and the inspection area, determining the observation evaluation information of each candidate observation location based on the anomaly locations, and determining the observation effectiveness parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage; step 3 involves identifying whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; if so, proceed to step 4; if not, proceed to step 5; step 4 involves identifying the candidate observation location with the maximum observation effectiveness parameter among the candidate observation locations whose observation coverage exceeds the preset coverage threshold as the target observation location; step 5 involves iteratively calculating the comprehensive observation coverage increment and time-effect attenuation cost of each candidate observation location, and successively selecting candidate observation locations as target observation locations based on the comprehensive observation coverage increment and the time-effect attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold. The above-mentioned method for determining the observation position of the inspection robot dynamically and adaptively determines the observation position based on the location and type of anomaly, thereby improving the inspection robot's ability to comprehensively observe anomalies. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating a method for determining the observation position of an inspection robot according to an embodiment of this application; Figure 2 This is a flowchart illustrating another method for determining the observation position of an inspection robot provided in an embodiment of this application; Figure 3 This is a flowchart illustrating another method for determining the observation position of an inspection robot provided in this application embodiment; Figure 4 This is a schematic diagram of the observation position determination device for an inspection robot provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this application clearer, specific embodiments of this application will be described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely for explaining this application and not for limiting it. It should also be noted that, for ease of description, only the parts relevant to this application are shown in the drawings, not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe operations (or steps) as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but may also have additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0024] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0025] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0026] The following description, in conjunction with the accompanying drawings, details the method, apparatus, equipment, and medium for determining the observation position of the inspection robot provided in this application, through specific embodiments and application scenarios.

[0027] First, this application is applicable to automated inspection scenarios in complex and structured environments, such as equipment inspection in industrial settings, urban infrastructure inspection, and warehousing and logistics environment inspection. Based on the above application scenarios, it is understood that the executing entity of this application can be an intelligent terminal device with computing processing capabilities, such as an industrial-grade edge computing terminal, an intelligent monitoring host, a portable industrial control computer, a robot-specific controller, and an industrial IoT edge gateway, etc., without further limitations.

[0028] Figure 1This is a flowchart illustrating a method for determining the observation position of an inspection robot according to an embodiment of this application. Figure 1 As shown, the specific steps include the following: Step 1: Obtain global perception information of the inspection area, and determine the location and type of anomalies based on the global perception information.

[0029] The inspection area can be the spatial range in which the inspection robot needs to autonomously perform anomaly observation tasks; the global perception information can be image information covering the entire inspection area.

[0030] Among them, the inspection robot can be an intelligent robot with autonomous mobility and equipped with multiple sensors, used to replace or assist humans in completing abnormal observation tasks, such as an embodied intelligent robot.

[0031] In one embodiment, global perception information of the inspection area can be obtained by using a multi-camera network monitoring system. This multi-camera network monitoring system can be deployed at key locations within the inspection area, enabling a visual monitoring system that achieves full coverage and seamless data collection through the collaborative work of multiple cameras.

[0032] Among them, abnormal location can be a spatial point or spatial sub-region within the inspection area that has an abnormal state; abnormal type can be a classification of the abnormal state corresponding to the abnormal location, such as high temperature / fire, structural damage, medium leakage, personnel violation, and equipment failure.

[0033] In one embodiment, determining the location and type of anomalies based on global perception information can be achieved by inputting the global perception information into a pre-trained anomaly detection model, thereby obtaining the anomaly location and type output by the model. The anomaly detection model can be a deep learning model capable of automatically identifying and locating anomalies from the global perception information and simultaneously classifying anomaly types. Specifically, the anomaly detection model can be a deep learning-based multimodal fusion network, such as a target detection model combining convolutional neural networks and attention mechanisms, or a semantic segmentation model using a U-Net structure.

[0034] Step 2: Generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage.

[0035] Among them, candidate observation locations can be discrete spatial points within the inspection area that meet the reachability constraints of the inspection robot and can be used to perform observations on abnormal locations.

[0036] In one embodiment, the method of generating multiple candidate observation positions based on the abnormal location and the inspection area can be achieved by pre-setting a passable area for the inspection robot within the inspection area, using the Poisson disk sampling method to generate uniformly distributed candidate observation positions within the passable area, and selecting the candidate observation positions that are not obstructed from the abnormal location as the final candidate observation positions.

[0037] Among them, the observation evaluation information can be a set of indicators used to quantitatively evaluate the observation quality of candidate observation locations to anomaly locations. The observation evaluation information may include safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage.

[0038] Specifically, the safety distance evaluation parameter can be a normalized score value used to characterize the safety level of the candidate observation location; the observation distance evaluation parameter can be a normalized score value used to characterize the quality of the observation distance between the candidate observation location and the abnormal location; and the observation coverage can refer to the proportion of the effective observation field of view of the candidate observation location to the spatial coverage of the abnormal location.

[0039] The formula for calculating the safety distance evaluation parameter of the candidate observation location can be: when d < d_safe, When d ≥ d_safe, Where S_safe represents the safety distance evaluation parameter, d represents the observation distance between the candidate observation position and the abnormal position, and d_safe represents the preset safety distance threshold.

[0040] The formula for calculating the observation distance evaluation parameter of the candidate observation location can be: when d < d_min, In the case where d_min ≤ d ≤ d_opt, When d > d_opt, Where S_dist represents the observation distance evaluation parameter, and d_min and d_opt represent the lower and upper limits of the preset optimal observation distance interval, respectively.

[0041] The formula for calculating the observation coverage of candidate observation locations can be: , Where v represents the observation direction vector of the candidate observation position relative to the anomaly position, T represents the center of the anomaly position, P represents the candidate observation position, S_angle represents the observation coverage, n_i represents the normal vector of the i-th convex hull surface of the anomaly position, and A_i represents the area of ​​the i-th convex hull surface of the anomaly position. This represents the total area of ​​all convex hull surfaces at the abnormal location.

[0042] Among them, the observation effectiveness parameter can be a quantitative indicator obtained by combining observation evaluation information and anomaly type, used to characterize the quality of observation at candidate observation locations.

[0043] In one embodiment, the method of determining the observation validity parameters of each candidate observation location based on observation evaluation information and anomaly type can be achieved by determining the corresponding preset allocation weight according to the anomaly type, and then performing a weighted summation calculation on each indicator in the observation evaluation information of the candidate observation location according to the preset allocation weight to obtain the observation validity parameters of the candidate observation location.

[0044] Step 3: Identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; if so, proceed to step 4; if not, proceed to step 5.

[0045] The target observation location can be the observation location that the inspection robot will ultimately use to perform the anomaly observation task.

[0046] The preset coverage threshold can be a pre-set critical value used to determine whether the observation coverage meets the coverage requirements.

[0047] If the observation coverage of a candidate observation location exceeds a preset coverage threshold, it means that the candidate observation location can independently achieve coverage observation of the abnormal location. Therefore, the target observation location can be directly selected from the candidate observation locations whose observation coverage exceeds the preset coverage threshold.

[0048] Step 4: Among the candidate observation locations where the observation coverage exceeds the preset coverage threshold, the candidate observation location corresponding to the maximum observation effectiveness parameter is determined as the target observation location.

[0049] The maximum observation validity parameter is the maximum value among the observation validity parameters of candidate observation locations whose observation coverage exceeds the preset coverage threshold.

[0050] Among the candidate observation locations whose observation coverage exceeds the preset coverage threshold, the candidate observation location corresponding to the maximum observation effectiveness parameter is determined as the target observation location. That is, under the premise of meeting the coverage requirements, the single observation location with the best overall observation quality is selected.

[0051] Step 5: Iteratively calculate the comprehensive observation coverage increment and the time-related attenuation cost of each candidate observation location, and select candidate observation locations as target observation locations one by one according to the comprehensive observation coverage increment and the time-related attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold.

[0052] The incremental comprehensive observation coverage can be the increase in the comprehensive observation coverage of all target observation locations after adding the candidate observation location as the target observation location.

[0053] The overall observation coverage refers to the combined observation coverage of the anomaly location by all target observation locations. Correspondingly, if the overall observation coverage of the target observation locations exceeds a preset coverage threshold, the overall coverage after collaboration of all current target observation locations meets the coverage requirements. Therefore, all current target observation locations can be determined as the final target observation locations.

[0054] Among them, the cost of time decay can be quantitatively represented by the time loss and reduced efficiency of abnormal observation tasks caused by the increase in observation locations and the extension of movement paths.

[0055] In one embodiment, the method of iteratively calculating the comprehensive observation coverage increment and the time-related attenuation cost of each candidate observation location, and successively selecting candidate observation locations as target observation locations based on the comprehensive observation coverage increment and the time-related attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds a preset coverage threshold, can be achieved by calculating the comprehensive observation coverage increment and the time-related attenuation cost of each candidate observation location in each iteration, selecting the candidate observation location corresponding to the largest comprehensive observation coverage increment as the target observation location, and, if the largest comprehensive observation coverage increment corresponds to multiple candidate observation locations, selecting the candidate observation location corresponding to the smallest time-related attenuation cost from among the candidate observation locations corresponding to the largest comprehensive observation coverage increment as the target observation location, until the comprehensive observation coverage of all target observation locations exceeds a preset coverage threshold.

[0056] In this embodiment, step 1 involves acquiring global perception information of the inspection area and determining the location and type of anomalies based on the global perception information; step 2 involves generating multiple candidate observation locations based on the anomaly locations and the inspection area, determining the observation evaluation information of each candidate observation location based on the anomaly locations, and determining the observation effectiveness parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage; step 3 involves identifying whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; if so, proceed to step 4; if not, proceed to step 5; step 4 involves identifying the candidate observation location with the maximum observation effectiveness parameter among the candidate observation locations whose observation coverage exceeds the preset coverage threshold as the target observation location; step 5 involves iteratively calculating the comprehensive observation coverage increment and time-effect attenuation cost of each candidate observation location, and successively selecting candidate observation locations as target observation locations based on the comprehensive observation coverage increment and the time-effect attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold. The above-mentioned method for determining the observation position of the inspection robot dynamically and adaptively determines the observation position based on the location and type of anomaly, thereby improving the inspection robot's ability to comprehensively observe anomalies.

[0057] Figure 2 This is a flowchart illustrating another method for determining the observation position of an inspection robot provided in an embodiment of this application. For example... Figure 2 As shown, the specific steps include the following: Step 1: Obtain global perception information of the inspection area, and determine the location and type of anomalies based on the global perception information.

[0058] Step 2: Generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage.

[0059] Step 3: Identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; if so, proceed to step 4; if not, proceed to step 51. Step 4: Among the candidate observation locations where the observation coverage exceeds the preset coverage threshold, the candidate observation location corresponding to the maximum observation effectiveness parameter is determined as the target observation location.

[0060] Step 51: Calculate the incremental comprehensive observation coverage corresponding to each of the remaining candidate observation locations based on the current target observation location set.

[0061] The current target observation location set can be a set of selected target observation locations; the remaining candidate observation locations can be candidate observation locations that have not yet been added to the current target observation location set.

[0062] In one embodiment, the method for calculating the incremental comprehensive observation coverage corresponding to each remaining candidate observation location based on the current target observation location set can be as follows: calculate the comprehensive observation coverage of the current target observation location set, calculate the comprehensive observation coverage of the union of the current target observation location set and each candidate observation location, and subtract the comprehensive observation coverage of the current target observation location set from the comprehensive observation coverage of the union of each candidate observation location to obtain the incremental comprehensive observation coverage corresponding to each candidate observation location.

[0063] The formula for calculating the comprehensive observation coverage of the current target observation location set can be: Where S_angle_total represents the comprehensive observation coverage, and v_j represents the observation direction vector of the j-th target observation position relative to the anomaly position; during the traversal of the current target observation position set, k takes the value of 0 if the i-th convex hull surface has been marked, and takes the value of 1 if the i-th convex hull surface has not been marked; when the i-th convex hull surface is first encountered... When the value is less than 0, the i-th convex hull is marked, and the convex hull is not counted again in subsequent traversals.

[0064] In one embodiment, calculating the incremental comprehensive observation coverage corresponding to each remaining candidate observation location based on the current target observation location set includes: determining the effective observation angle intervals of each target observation location in the current target observation location set and the effective observation angle intervals of each remaining candidate observation location based on the abnormal location; calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set; determining the effective observation angle interval of the target based on the effective observation angle intervals of each target observation location in the current target observation location set and the effective observation angle intervals of each remaining candidate observation location, and calculating the comprehensive observation coverage of the target based on the effective observation angle intervals of the target; subtracting the comprehensive observation coverage of the current target observation location set from the comprehensive observation coverage of the target corresponding to each candidate observation location to obtain the incremental comprehensive observation coverage corresponding to each candidate observation location.

[0065] The effective observation angle range can be the angular range that the target observation position / candidate observation position can cover, centered on the abnormal position and determined by the line connecting the target observation position / candidate observation position and the abnormal position and the sensor field of view.

[0066] In one embodiment, the method of determining the effective observation angle intervals of each target observation position and the remaining candidate observation positions in the current target observation position set based on the abnormal position can be achieved by taking the center of the abnormal position as the origin and the line connecting the target observation position / candidate observation position and the abnormal position as the central axis, and calculating the effective observation angle interval corresponding to the target observation position / candidate observation position at the abnormal position based on the horizontal field of view of the sensor.

[0067] In one embodiment, the method for calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set can be by calculating the union of the effective observation angle intervals of each target observation location in the current target observation location set, and dividing the total angle span of the union by 360 degrees to obtain the comprehensive observation coverage of the current target observation location set.

[0068] In one embodiment, before calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set, the method further includes: determining each preset angle interval and the angle weight corresponding to each preset angle interval according to the anomaly type; correspondingly, calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set includes: calculating the partitioned observation coverage of each preset angle region based on the effective observation angle intervals of each target observation location in the current target observation location set and each preset angle interval; and performing a weighted summation calculation on the partitioned observation coverage of each preset angle region based on the angle weight corresponding to each preset angle interval to obtain the comprehensive observation coverage of the current target observation location set.

[0069] Among them, the preset angle interval can be the angle range corresponding to different directional areas divided by azimuth angle; the angle weight can be used to characterize the proportion of the preset angle interval in the comprehensive evaluation.

[0070] In one embodiment, the current preset angle intervals and their corresponding angle weights can be determined based on the current anomaly type and a pre-built mapping relationship between anomaly types and preset angle intervals and their corresponding angle weights. As an example, the mapping relationship between anomaly types and preset angle intervals and their corresponding angle weights can include: if the anomaly type is high temperature / fire, then the range of flame spread angles is determined as a preset angle interval and assigned an angle weight of 0.7; other angle ranges are determined as preset angle intervals and assigned an angle weight of 0.3.

[0071] Among them, the zonal observation coverage can be the effective observation coverage ratio of the current target observation position set for a single preset angle interval.

[0072] In one embodiment, the method of calculating the partitioned observation coverage of each preset angle region based on the effective observation angle intervals of each target observation position in the current target observation position set and each preset angle interval can be adopted by calculating the intersection of the effective observation angle intervals of each target observation position with the preset angle interval for a preset angle region, and performing a union calculation on the intersections corresponding to each target observation position to obtain the coverage angle interval corresponding to the preset angle interval, and dividing the total angle span of the coverage angle interval by the total angle span of the preset angle interval to obtain the partitioned observation coverage of the preset angle region.

[0073] The comprehensive observation coverage of the current target observation location set is the sum of the products between the partition observation coverage corresponding to each preset angle interval and the angle weight.

[0074] The advantage of this scheme is that by dynamically adjusting the weights of different directional areas according to the anomaly type, the assessment of comprehensive observation coverage is more aligned with the detection needs of different anomalies, thereby improving the targeting and effectiveness of observation location selection.

[0075] The effective observation angle interval corresponding to the candidate observation position can be the union of the effective observation angle intervals of each target observation position in the current target observation position set and the effective observation angle interval of the candidate observation position.

[0076] One method for calculating the comprehensive observation coverage of a target based on its effective observation angle range is to divide the total angular span of the effective observation angle range of the target by 360 degrees to obtain the comprehensive observation coverage of the target.

[0077] The incremental comprehensive observation coverage corresponding to each candidate observation location is the difference between the comprehensive observation coverage of the target corresponding to each candidate observation location and the comprehensive observation coverage of the current set of target observation locations.

[0078] The advantage of this scheme is that it directly eliminates observation overlap through the union operation of angle intervals, avoiding duplicate counting. At the same time, it uses the proportion of angle span as the coverage index, which is computationally efficient and has an intuitive physical meaning. It reduces computational complexity while ensuring the accuracy of coverage assessment.

[0079] Step 52: Calculate the path length corresponding to each of the remaining candidate observation locations based on the current target observation location set, and calculate the time decay cost corresponding to each of the remaining candidate observation locations based on the path length.

[0080] The path length can be the shortest path length for the inspection robot to move from any target observation position in the current set of target observation positions to the remaining candidate observation positions.

[0081] In one embodiment, the method of calculating the path length corresponding to each of the remaining candidate observation positions based on the current target observation position set can be as follows: for each of the remaining candidate observation positions, the path length from each target observation position in the current target observation position set to the candidate observation position can be calculated based on the environmental information of the inspection area and the robot motion path planning algorithm, and the shortest path length among them can be determined as the path length corresponding to the candidate observation position in this iteration.

[0082] In one embodiment, the method of calculating the time-related attenuation cost corresponding to each remaining candidate observation location based on the path length can be to directly use the path length corresponding to each remaining candidate observation location as the time-related attenuation cost corresponding to each candidate observation location.

[0083] In one embodiment, calculating the time-delay cost corresponding to each remaining candidate observation location based on the path length includes: determining the time-delay requirement parameter for each candidate observation location based on the anomaly type, and calculating the time-delay prediction parameter for each candidate observation location based on the path length; for each candidate observation location, calculating the ratio of the time-delay prediction parameter to the time-delay requirement parameter as the time-delay cost of the candidate observation location.

[0084] Among them, the timeliness requirement parameter for candidate observation locations can be a pre-set threshold for the maximum allowable time for the inspection robot to reach the candidate observation location based on the urgency of the anomaly type.

[0085] In one embodiment, determining the timeliness requirement parameters for each candidate observation location based on the anomaly type can be achieved by using the current anomaly type and a pre-built correspondence between anomaly types and timeliness requirement parameters to determine the corresponding target timeliness requirement parameters, which serve as the timeliness requirement parameters for all candidate observation locations. As an example, the correspondence between anomaly types and timeliness requirement parameters can include: if the anomaly type is high temperature / fire, then the timeliness requirement parameter for each candidate observation location is 60 seconds; if the anomaly type is personnel violation, then the timeliness requirement parameter for each candidate observation location is 300 seconds.

[0086] In one embodiment, determining the timeliness requirement parameters for each candidate observation location based on the anomaly type includes: determining the anomaly propagation rate and the degree of impact on the inspection robot based on the anomaly type; calculating the remaining tolerable duration for each candidate observation location based on the anomaly propagation rate; and dividing the remaining tolerable duration for each candidate observation location by the degree of impact parameters to obtain the timeliness requirement parameters for each candidate observation location.

[0087] Among them, the abnormal diffusion rate can be the distance the abnormal area expands outward per unit time; the parameter of the degree of influence on the inspection robot can be a normalized coefficient set based on the degree of interference of the abnormality type on the robot's observation function and motion safety.

[0088] In one embodiment, determining the anomaly diffusion rate and its impact on the inspection robot based on the anomaly type can be achieved by querying a pre-built mapping table of anomaly types, diffusion rates, and impact parameters to obtain the corresponding anomaly diffusion rate and impact parameters. As an example, this mapping table includes: if the anomaly type is high temperature / fire, the anomaly diffusion rate is 0.5 m / s and the impact parameter is 4.0; if the anomaly type is media leakage, the anomaly diffusion rate is 0.25 m / s and the impact parameter is 2.5.

[0089] The remaining tolerable duration can be the longest allowable duration from the corresponding moment of this iteration, before the anomaly spreads to the candidate observation position, to ensure that the robot can still make effective observations without any safety risks.

[0090] In one embodiment, the remaining tolerable duration for each candidate observation location can be calculated based on the abnormal diffusion rate by dividing the straight-line distance from the candidate observation location to the center of the abnormal location by the abnormal diffusion rate to obtain the abnormal diffusion duration, and then subtracting the sum of the cumulative movement time and observation time of the current target observation location set corresponding to this iteration from the abnormal diffusion duration to obtain the remaining tolerable duration.

[0091] Among them, the higher the impact degree parameter, the larger the denominator of the ratio of the remaining tolerable time to the impact degree parameter, and the shorter the final time requirement parameter. This can prompt the inspection robot to prioritize safe observation locations. Therefore, the ratio of the remaining tolerable time to the impact degree parameter can be used as the time requirement parameter.

[0092] The advantage of this scheme is that it ensures the timeliness of observations in high-diffusion-rate, high-threat areas while reserving a reasonable timeliness buffer for low-threat, distant areas.

[0093] Among them, the timeliness prediction parameter can be the shortest time required for the inspection robot to move to the candidate observation location.

[0094] In one embodiment, the method for calculating the time-prediction parameters of each candidate observation location based on the path length can be to divide the path length by the preset moving speed of the inspection robot to obtain the time-prediction parameters.

[0095] Among them, the ratio of the timeliness prediction parameter to the timeliness requirement parameter of the candidate observation location is used. The larger the ratio, the closer the predicted time for the inspection robot to reach the candidate observation location is to the allowable threshold, the more severe the timeliness loss, and the higher the attenuation cost. The smaller the ratio, the better the timeliness performance and the lower the attenuation cost. Therefore, the ratio of the timeliness prediction parameter to the timeliness requirement parameter of the candidate observation location can be used as the timeliness attenuation cost of the candidate observation location.

[0096] The advantage of this approach is that it dynamically matches timeliness requirements with anomaly types and accurately predicts timeliness consumption by combining path length and robot movement speed, making the selection of observation locations more targeted in balancing coverage gain and timeliness cost.

[0097] Step 53: Substitute the comprehensive observation coverage increment and the time-related decay cost into the pre-constructed evaluation function to obtain the comprehensive evaluation parameters, and determine the candidate observation position corresponding to the maximum comprehensive evaluation parameter as the target observation position, and add the target observation position to the current target observation position set.

[0098] Among them, the comprehensive evaluation parameter can be a quantitative value that is calculated by the evaluation function and characterizes the overall advantages and disadvantages of the candidate observation location in terms of coverage gain and time cost.

[0099] The evaluation function can be a pre-defined mapping relationship that uses the increase in comprehensive observation coverage and the cost of time-related degradation as independent variables to quantify and evaluate the comprehensive value of candidate observation locations, for example: Where J represents the comprehensive evaluation parameter, This represents the increase in comprehensive observation coverage, and C represents the cost of time-degradation. and These are the preset weights corresponding to the increase in comprehensive observation coverage and the cost of time-related degradation, respectively.

[0100] The maximum comprehensive evaluation parameter is the maximum value among the comprehensive evaluation parameters corresponding to the remaining candidate observation locations.

[0101] Among them, the candidate observation position corresponding to the maximum comprehensive evaluation parameter is determined as the target observation position. That is, among the remaining candidate observation positions, the candidate observation position with the best comprehensive performance in terms of coverage gain and timeliness cost is selected as the new target observation position.

[0102] In this process, the target observation location is added to the current target observation location set to update the composition of the current target observation location set, providing a basis for the next round of iterative calculation.

[0103] Step 54: Identify whether the overall observation coverage of the current target observation location set exceeds the preset coverage threshold; if yes, end the process; if no, repeat steps 51 to 54.

[0104] The advantage of this scheme is that, through an iterative greedy selection mechanism, while ensuring that the observation coverage of abnormal locations meets the coverage requirements, the timeliness cost of the inspection robot is reduced as much as possible, thus achieving a balance between coverage integrity and inspection efficiency.

[0105] Figure 3 This is a flowchart illustrating another method for determining the observation position of an inspection robot provided in this application embodiment. For example... Figure 3 As shown, the specific steps include the following: Step 1: Obtain global perception information of the inspection area, and determine the location and type of anomalies based on the global perception information.

[0106] Step 21: Generate multiple candidate observation locations based on the anomaly location and the inspection area.

[0107] Step 22: Obtain environmental information of the inspection area and the types of sensors included in the inspection robot.

[0108] Among them, environmental information can be a collection of various objective environmental data that affect the observation and movement of the inspection robot within the inspection area, including terrain structure information, lighting data, meteorological data, and information on obstructions within the inspection area.

[0109] Among them, the sensor type can be a category of various devices carried by the inspection robot for environmental perception and anomaly detection, classified according to their working principle and function, such as visual sensors and millimeter-wave radar sensors.

[0110] In one embodiment, the method for obtaining environmental information of the inspection area and the various sensor types included in the inspection robot can be to collect real-time light and meteorological data through the environmental perception sensors carried by the inspection robot, and to read terrain structure information, obstruction information, and various sensor types through the system configuration file or preset parameter library of the inspection robot.

[0111] Step 23: Determine the observation evaluation information for each candidate observation location based on the anomaly location, the anomaly type, the environmental information, and the sensor types; wherein the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, observation coverage, occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters.

[0112] Among them, the occlusion evaluation parameter can be a normalized score value used to characterize the degree of interference of the occlusion object on the sensor observation along the observation path between the candidate observation position and the abnormal position.

[0113] In one embodiment, the method for determining the occlusion evaluation parameters for each candidate observation location can be as follows: for each sensor included in the inspection robot, the occlusion score corresponding to each sensor is determined based on environmental information, candidate observation locations, and abnormal locations, and the occlusion scores corresponding to each sensor are weighted and summed to obtain the occlusion evaluation parameters.

[0114] Among them, the illumination evaluation parameters can be normalized score values ​​used to characterize the illumination conditions at candidate observation locations.

[0115] In one embodiment, the illumination evaluation parameters for each candidate observation location can be determined by calculating the illumination adaptation score for each sensor type based on illumination data and the illumination data threshold corresponding to each sensor type, and then performing a weighted summation of the illumination adaptation scores for each sensor type to obtain the illumination evaluation parameters.

[0116] Among them, the sensor adaptation evaluation parameters can be normalized score values ​​used to characterize the degree of matching between the observation distance, anomaly type and the performance of each sensor of the inspection robot corresponding to the candidate observation position.

[0117] In one embodiment, the sensor adaptation evaluation parameters for each candidate observation location can be determined by calculating the observation distance between each candidate observation location and the abnormal location, determining the adaptation score of each sensor included in the inspection robot based on the abnormality type, observation distance, sensor type, and pre-built adaptation mapping relationship, and then performing a weighted summation of the adaptation scores of each sensor to obtain the sensor adaptation evaluation parameters.

[0118] Step 24: Determine the observation validity parameters for each candidate observation location based on the observation evaluation information and the anomaly type.

[0119] In one embodiment, the method of determining the observation validity parameters of each candidate observation location based on observation evaluation information and anomaly type can be achieved by determining the corresponding preset allocation weight according to the anomaly type, and then performing a weighted summation calculation on the safety distance evaluation parameters, observation distance evaluation parameters, observation coverage, occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters of the candidate observation location according to the preset allocation weight, so as to obtain the observation validity parameters of the candidate observation location.

[0120] Step 3: Identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; if so, proceed to step 4; if not, proceed to step 5. Step 4: If there are candidate observation locations where the observation coverage exceeds the preset coverage threshold, the candidate observation location corresponding to the maximum observation effectiveness parameter is determined as the target observation location among the candidate observation locations where the observation coverage exceeds the preset coverage threshold.

[0121] Step 5: Iteratively calculate the comprehensive observation coverage increment and the time-related attenuation cost of each candidate observation location, and select candidate observation locations as target observation locations one by one according to the comprehensive observation coverage increment and the time-related attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold.

[0122] The advantage of this scheme is that by introducing occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters, it is possible to comprehensively evaluate the observation feasibility and data quality of candidate observation locations in the actual environment, avoid observation failures caused by environmental obstacles, illumination changes, or sensor mismatch, thereby improving the accuracy and robustness of anomaly detection.

[0123] Figure 4 This is a schematic diagram of the observation position determination device for an inspection robot provided in an embodiment of this application. Figure 4 As shown, the device includes: The anomaly perception module 410 is used to acquire global perception information of the inspection area and determine the location and type of anomaly based on the global perception information. The observation and evaluation module 420 is used to generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage; The coverage identification module 430 is used to identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; The first determining module 440 is used to determine the candidate observation position corresponding to the maximum observation effectiveness parameter as the target observation position when there are candidate observation positions where the observation coverage exceeds the preset coverage threshold. The second determining module 450 is used to iteratively calculate the comprehensive observation coverage increment and the time-related decay cost of each candidate observation location when there are no candidate observation locations whose observation coverage exceeds the preset coverage threshold, and select candidate observation locations as target observation locations one by one according to the comprehensive observation coverage increment and the time-related decay cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold.

[0124] Furthermore, the second determining module 450 is specifically used for: Step 51: Calculate the incremental comprehensive observation coverage for each remaining candidate observation location based on the current target observation location set; Step 52: Calculate the path length corresponding to each of the remaining candidate observation locations based on the current target observation location set, and calculate the time decay cost corresponding to each of the remaining candidate observation locations based on the path length. Step 53: Substitute the comprehensive observation coverage increment and the time-degradation cost into the pre-constructed evaluation function to obtain the comprehensive evaluation parameters, and determine the candidate observation position corresponding to the maximum comprehensive evaluation parameter as the target observation position, and add the target observation position to the current target observation position set; Step 54: Identify whether the overall observation coverage of the current target observation location set exceeds the preset coverage threshold; if yes, end the process; if no, repeat steps 51 to 54.

[0125] Furthermore, the step of calculating the time-delay cost corresponding to each remaining candidate observation location based on the path length includes: The timeliness requirement parameters for each candidate observation location are determined based on the anomaly type, and the timeliness prediction parameters for each candidate observation location are calculated based on the path length. For each candidate observation location, the ratio of the timeliness prediction parameter to the timeliness requirement parameter of the candidate observation location is calculated as the timeliness attenuation cost of the candidate observation location.

[0126] Furthermore, determining the lead time requirement parameters for each candidate observation location based on the anomaly type includes: Determine the anomaly propagation rate and the degree of impact on the inspection robot based on the anomaly type; The remaining tolerable time for each candidate observation location is calculated based on the anomalous diffusion rate. Divide the remaining tolerable duration of each candidate observation location by the aforementioned impact degree parameter to obtain the timeliness requirement parameter for each candidate observation location.

[0127] Furthermore, the step of calculating the comprehensive observation coverage increment corresponding to each remaining candidate observation location based on the current target observation location set includes: Based on the abnormal location, determine the effective observation angle range of each target observation location in the current target observation location set and the effective observation angle range of each remaining candidate observation location; The comprehensive observation coverage of the current target observation location set is calculated based on the effective observation angle range of each target observation location in the current target observation location set; Based on the effective observation angle intervals of each target observation position in the current target observation position set and the effective observation angle intervals of each remaining candidate observation position, the effective observation angle interval of the target is determined, and the comprehensive observation coverage of the target is calculated based on the effective observation angle interval of the target. Subtract the overall observation coverage of the current set of target observation locations from the overall observation coverage of each candidate observation location to obtain the overall observation coverage increment for each candidate observation location.

[0128] Furthermore, before calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set, the method further includes: Determine the corresponding preset angle intervals and the angle weights corresponding to each preset angle interval based on the anomaly type. Based on the effective observation angle intervals of each target observation position in the current target observation position set and each preset angle interval, calculate the partition observation coverage of each preset angle region respectively; Based on the angle weights corresponding to each preset angle interval, the weighted summation of the observation coverage of each preset angle region is calculated to obtain the comprehensive observation coverage of the current target observation position set.

[0129] Furthermore, the observation and evaluation information also includes occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters; Accordingly, the observation and evaluation module 420 is specifically used for: Step 21: Generate multiple candidate observation locations based on the anomaly location and the inspection area; Step 22: Obtain environmental information of the inspection area and the types of sensors included in the inspection robot; Step 23: Determine the observation evaluation information for each candidate observation location based on the anomaly location, the anomaly type, the environmental information, and the sensor types; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, observation coverage, occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters. Step 24: Determine the observation validity parameters for each candidate observation location based on the observation evaluation information and the anomaly type.

[0130] In this embodiment, an anomaly perception module is used to acquire global perception information of the inspection area and determine the anomaly location and anomaly type based on the global perception information; an observation evaluation module is used to generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes a safety distance evaluation parameter, an observation distance evaluation parameter, and an observation coverage; a coverage identification module is used to identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; first The first determination module is used to, when there are candidate observation locations whose observation coverage exceeds a preset coverage threshold, determine the candidate observation location corresponding to the maximum observation effectiveness parameter as the target observation location. The second determination module is used to, when there are no candidate observation locations whose observation coverage exceeds the preset coverage threshold, iteratively calculate the comprehensive observation coverage increment and time-related attenuation cost of each candidate observation location, and successively select candidate observation locations as target observation locations based on the comprehensive observation coverage increment and the time-related attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold. The above-mentioned observation location determination device for the inspection robot dynamically and adaptively determines the observation location based on the anomaly location and anomaly type, improving the inspection robot's comprehensive anomaly observation capability.

[0131] The observation position determination device for the inspection robot in this application embodiment can be a device, or a component, integrated circuit, or chip in a terminal. This device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, PDAs, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network-attached storage (NAS), personal computers (PCs), televisions (TVs), ATMs, or self-service machines, etc. This application embodiment does not impose specific limitations.

[0132] The observation position determination device for the inspection robot in this embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this embodiment does not specifically limit its use.

[0133] The observation position determination device for the inspection robot provided in this application embodiment can realize the various processes implemented in the above embodiments. To avoid repetition, it will not be described again here.

[0134] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. For example... Figure 5 As shown, this application embodiment also provides an electronic device 500, including a processor 501, a memory 502, and a program or instructions stored in the memory 502 and executable on the processor 501. When the program or instructions are executed by the processor 501, they implement the various processes of the above-mentioned inspection robot observation position determination embodiment and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0135] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0136] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described inspection robot observation position determination embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0137] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0138] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0139] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0140] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0141] The above description is merely a preferred embodiment and the technical principles employed in this application. This application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions that can be made by those skilled in the art will not depart from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the claims.

Claims

1. A method for determining the observation position of an inspection robot, characterized in that, The method includes: Step 1: Obtain global perception information of the inspection area, and determine the location and type of anomalies based on the global perception information; Step 2: Generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage; Step 3: Identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; if so, proceed to step 4; if not, proceed to step 5. Step 4: Among the candidate observation locations where the observation coverage exceeds the preset coverage threshold, the candidate observation location corresponding to the maximum observation effectiveness parameter is determined as the target observation location; Step 5: Iteratively calculate the comprehensive observation coverage increment and the time-related attenuation cost of each candidate observation location, and select candidate observation locations as target observation locations one by one according to the comprehensive observation coverage increment and the time-related attenuation cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold.

2. The method for determining the observation position of the inspection robot according to claim 1, characterized in that, Step 5 includes: Step 51: Calculate the incremental comprehensive observation coverage for each remaining candidate observation location based on the current target observation location set; Step 52: Calculate the path length corresponding to each of the remaining candidate observation locations based on the current target observation location set, and calculate the time decay cost corresponding to each of the remaining candidate observation locations based on the path length. Step 53: Substitute the comprehensive observation coverage increment and the time-degradation cost into the pre-constructed evaluation function to obtain the comprehensive evaluation parameters, and determine the candidate observation position corresponding to the maximum comprehensive evaluation parameter as the target observation position, and add the target observation position to the current target observation position set; Step 54: Identify whether the overall observation coverage of the current target observation location set exceeds the preset coverage threshold; if yes, end the process; if no, repeat steps 51 to 54.

3. The method for determining the observation position of the inspection robot according to claim 2, characterized in that, The step of calculating the time-delay cost corresponding to each remaining candidate observation location based on the path length includes: The timeliness requirement parameters for each candidate observation location are determined based on the anomaly type, and the timeliness prediction parameters for each candidate observation location are calculated based on the path length. For each candidate observation location, the ratio of the timeliness prediction parameter to the timeliness requirement parameter of the candidate observation location is calculated as the timeliness attenuation cost of the candidate observation location.

4. The method for determining the observation position of the inspection robot according to claim 3, characterized in that, The step of determining the timeliness requirement parameters for each candidate observation location based on the anomaly type includes: Determine the anomaly propagation rate and the degree of impact on the inspection robot based on the anomaly type; The remaining tolerable time for each candidate observation location is calculated based on the anomalous diffusion rate. Divide the remaining tolerable duration of each candidate observation location by the aforementioned impact degree parameter to obtain the timeliness requirement parameter for each candidate observation location.

5. The method for determining the observation position of the inspection robot according to claim 2, characterized in that, The step of calculating the incremental comprehensive observation coverage corresponding to each remaining candidate observation location based on the current target observation location set includes: Based on the abnormal location, determine the effective observation angle range of each target observation location in the current target observation location set and the effective observation angle range of each remaining candidate observation location; The comprehensive observation coverage of the current target observation location set is calculated based on the effective observation angle range of each target observation location in the current target observation location set; Based on the effective observation angle intervals of each target observation position in the current target observation position set and the effective observation angle intervals of each remaining candidate observation position, the effective observation angle interval of the target is determined, and the comprehensive observation coverage of the target is calculated based on the effective observation angle interval of the target. Subtract the overall observation coverage of the current set of target observation locations from the overall observation coverage of each candidate observation location to obtain the overall observation coverage increment for each candidate observation location.

6. The method for determining the observation position of the inspection robot according to claim 5, characterized in that, Before calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set, the method further includes: Determine the corresponding preset angle intervals and the angle weights corresponding to each preset angle interval based on the anomaly type. Accordingly, calculating the comprehensive observation coverage of the current target observation location set based on the effective observation angle intervals of each target observation location in the current target observation location set includes: Based on the effective observation angle intervals of each target observation position in the current target observation position set and each preset angle interval, calculate the partition observation coverage of each preset angle region respectively; Based on the angle weights corresponding to each preset angle interval, the weighted summation of the observation coverage of each preset angle region is calculated to obtain the comprehensive observation coverage of the current target observation position set.

7. The method for determining the observation position of the inspection robot according to any one of claims 1-6, characterized in that, The observation and evaluation information also includes occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters; Accordingly, step 2 includes: Step 21: Generate multiple candidate observation locations based on the anomaly location and the inspection area; Step 22: Obtain environmental information of the inspection area and the types of sensors included in the inspection robot; Step 23: Determine the observation evaluation information for each candidate observation location based on the anomaly location, the anomaly type, the environmental information, and the sensor types; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, observation coverage, occlusion evaluation parameters, illumination evaluation parameters, and sensor adaptation evaluation parameters. Step 24: Determine the observation validity parameters for each candidate observation location based on the observation evaluation information and the anomaly type.

8. A device for determining the observation position of an inspection robot, characterized in that, The device includes: The anomaly detection module is used to acquire global perception information of the inspection area and determine the location and type of anomalies based on the global perception information. The observation and evaluation module is used to generate multiple candidate observation locations based on the anomaly location and the inspection area, determine the observation evaluation information of each candidate observation location based on the anomaly location, and determine the observation validity parameters of each candidate observation location based on the observation evaluation information and the anomaly type; wherein, the observation evaluation information includes safety distance evaluation parameters, observation distance evaluation parameters, and observation coverage; The coverage identification module is used to identify whether there are candidate observation locations whose observation coverage exceeds a preset coverage threshold; The first determining module is used to determine the candidate observation position corresponding to the maximum observation effectiveness parameter as the target observation position when there are candidate observation positions where the observation coverage exceeds the preset coverage threshold. The second determining module is used to iteratively calculate the comprehensive observation coverage increment and the time-related decay cost of each candidate observation location when there are no candidate observation locations whose observation coverage exceeds the preset coverage threshold, and select candidate observation locations as target observation locations one by one according to the comprehensive observation coverage increment and the time-related decay cost, until the comprehensive observation coverage of all target observation locations exceeds the preset coverage threshold.

9. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein when the program or instructions are executed by the processor, they implement the observation position determination method of the inspection robot as described in any one of claims 1-7.

10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a processor, implement the observation position determination method for the inspection robot as described in any one of claims 1-7.

Images