Instrument panel identification control system and method based on wheel type inspection robot

Abstract

The invention discloses an instrument panel identification control system and method based on a wheeled inspection robot, and relates to the technical field of instrument panel identification. Acquiring and analyzing instrument parameters of each monitoring device to judge the state of each monitoring device, generating a risk assessment index, and dynamically adjusting an inspection path based on the working state (risk assessment index) of the monitoring device; the wheel type inspection robot is controlled to identify an instrument panel of each point position through the received optimal inspection route, instrument parameters of equipment of each point position are obtained, early warning analysis is carried out on the equipment according to the instrument parameters of the instrument panel, and equipment early warning is triggered accordingly; the equipment which does not trigger the early warning is recorded as the monitoring equipment, and the corresponding instrument parameters are sent to the server for storage and are used as the data basis of the routing inspection path planning of the wheel type routing inspection robot in the next round, so that the rapid abnormal identification and early warning are realized, and the intelligent level of equipment management is obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of instrument panel recognition, and in particular to an instrument panel recognition control system and method based on a wheeled inspection robot. Background Art

[0002] With the development of science and technology, more and more manual instrument inspections in factories are gradually replaced by robots, such as wheeled inspection robots. This is because the working environment conditions in many large factories are harsh, usually manifested as high temperature, high noise, harmful gases, dangerous areas (high voltage areas, confined spaces, etc.), etc. Wheeled inspection robots can effectively reduce personnel risks in the inspection of instruments in factories.

[0003] During the inspection process, current wheeled robots usually identify and upload each instrument according to a fixed route in the background. This long-term fixed inspection route has great limitations. It lacks intelligent automatic planning of the display status of each instrument and cannot achieve adaptive inspection in complex scenarios. Summary of the invention

[0004] The main purpose of the present invention is to provide an instrument panel recognition control system and method based on a wheeled inspection robot to overcome the problems mentioned in the above background technology.

[0005] To achieve the above object, according to one aspect of the present invention, a dashboard recognition control system based on a wheeled inspection robot is provided, the system comprising: a server, an inspection intelligent planning module and an instrument recognition control module;

[0006] The server communicates with the wheeled inspection robot to obtain instrument parameters of each device. The specific instrument parameters include temperature, power, amplitude and noise value at each monitoring time during the monitoring cycle;

[0007] The inspection intelligent planning module analyzes and judges the status of each device based on the instrument parameters at each monitoring moment in the monitoring window to plan the inspection path of the wheeled robot; specifically:

[0008] Step 1: Retrieve the instrument parameters of each monitoring device. The specific instrument parameters include temperature, power, amplitude and noise value. Each device corresponds to a set of standard parameters. The specific standard parameters include standard temperature range and standard power range. Perform quantitative analysis based on the instrument parameters and standard parameters to obtain the status value of the monitoring device at each monitoring time.

[0009] Step 2: Construct a two-dimensional rectangular coordinate system with the monitoring time as the horizontal coordinate and the state value as the vertical coordinate, input the state value into the coordinate system according to its corresponding monitoring time, and record the position of the state value in the coordinate system as the state point, and use line segments to connect each state point in sequence to obtain a state value line graph of the equipment; further analyze the health trend status of the equipment based on the state value line graph to obtain the risk assessment index of the equipment;

[0010] Step 3: Thus, the risk assessment values ​​of several devices can be obtained, the plan layout of the plant area is obtained, and the instrument panels of each device are marked in the plan layout of the plant area according to their corresponding positions, thereby obtaining several points; the point closest to the initial position of the wheeled robot is taken as the starting point, thereby obtaining the constraint conditions of the inspection path: starting point - risk assessment value of each point - starting point; dynamic planning is generated according to the constraint conditions of the inspection path to generate the optimal inspection route, and it is sent to the instrument identification control module;

[0011] The instrument recognition control module controls the wheeled inspection robot to perform instrument panel recognition on the equipment at each point based on the received optimal inspection path to obtain the instrument parameters of the equipment at each point, and performs early warning analysis on the equipment to determine whether to trigger an early warning.

[0012] Furthermore, the specific process of quantitative analysis based on instrument parameters and standard parameters is as follows:

[0013] Retrieve the instrument parameters of each monitoring device, including temperature, power, amplitude and noise value, and record them as Tj, Pj, Dj and Zj respectively; where j = 1, 2, 3 ... J, J is a positive integer, J represents the total number of monitoring moments, and j represents the sequence number of any monitoring moment; each device corresponds to a set of standard parameters, including standard temperature range and standard power range, and record them as [HT1, HT2] and [HP1, HP2] respectively;

[0014] The temperature Tj, power Pj, amplitude Dj, noise value Zj, standard temperature range [HT1, HT2] and standard power range [HP1, HP2] are normalized and their values ​​are taken. The values ​​are calculated and analyzed by formula to obtain the state value Uj of the equipment at each monitoring time. The specific calculation formula is:

[0015]

[0016] Wherein a1, a2, a3, and a4 are respectively set proportional constants, and their values ​​are set by those skilled in the art according to actual conditions; dTj represents the shortest distance between temperature Tj and the standard temperature interval [HT1, HT2]. If temperature Tj belongs to the standard temperature interval [HT1, HT2], dTj=0; if temperature Tj is greater than the maximum value in the standard temperature interval [HT1, HT2], dTj=Tj-HT2; if temperature Tj is less than the standard temperature interval [HT1, HT2], dTj=HT1-Tj; similarly, dPj represents the shortest distance between power Pj and the standard temperature interval [HP1, HP2]. If power Pj belongs to the standard power interval [HP1, HP2], dPj=0; if power Pj is greater than the maximum value in the standard power interval [HP1, HP2], dPj=Pj-HP2; if power Pj is less than the minimum value in the standard power interval [HP1, HP2], dPj=HP1-Pj.

[0017] Furthermore, the specific process of further analyzing the health trend status of the equipment based on the status value line chart is as follows:

[0018] Two adjacent state points can form a state line segment, so several state line segments can be segmented from the state value line graph, and the state values ​​corresponding to the two end points of the state line segment are averaged to obtain the line segment value recorded as The slope of the state line segment calculated by data fitting is recorded as

[0019] Set the segment value of each state segment and slope Perform normalization and take its value, and perform formulaic calculation and analysis on the value to obtain the risk assessment index KU of the equipment; the specific calculation formula is:

[0020]

[0021] α1 and α2 are respectively set proportional constants, and their specific values ​​are set by those skilled in the art according to actual needs.

[0022] Furthermore, a dynamic planning is generated according to the constraints of the inspection path to generate the best inspection route. The specific steps are as follows:

[0023] 3-1: Define the heuristic function: the starting point is denoted as P start , the current point Pi and the target point P target The straight-line distance h(i) between them, the weighted distance G risk (i);

[0024] 3-2: Cost function definition: actual cost g(i), estimated cost h(i) and total cost f(i);

[0025] 3-3: Dynamic path planning based on heuristic function and cost function;

[0026] 3-4: If the end point P end When it is added to the closed set, it means that the best path has been found. Then, starting from the end point, the parent point is traced back to obtain the best path and send it to the instrument identification control module.

[0027] Furthermore, the specific process of the heuristic function definition is:

[0028] 3-1-1: Let the starting point be P start , and the other points are recorded as Pi, where Pi represents any point in the plant layout diagram, i = 1, 2, 3 ... I, I is a positive integer, and I represents the point after removing the starting point P start The number of all points other than , i represents any one of them; specifically, the corresponding position coordinates are generated according to their positions in the plane layout diagram of the factory area, marked as (xi, yi), and their corresponding risk assessment values ​​are recorded as KUi;

[0029] 3-1-2: Record the next point as the target point. The next point obtained by performing inspection path planning at the current point can be used to dynamically plan the instrument panel recognition path of the wheeled inspection robot. Specifically:

[0030] The current point Pi and the target point P are obtained by the Euclidean distance calculation formula. target The straight-line distance between them is calculated by the specific Euclidean distance formula: Where (x target ,y target ) is the position coordinate of the target point;

[0031] 3-1-3: Define a heuristic function to calculate the weighted distance. The specific formula for calculating the weighted distance is:

[0032] G risk (i) = h(i) × (1 + w × KUi)

[0033] Among them, w is a weight coefficient that adjusts the impact of risk on path selection, and its specific value is set by personnel in this field. KUi is the risk assessment index of the device represented by point i; h(i) is the straight-line distance between the device and the target device.

[0034] Furthermore, the specific process of cost function definition is:

[0035] 3-2-1: Actual cost g(i): from the starting point Pstart The cost to the current point Pi is expressed in the following formula:

[0036]

[0037] Among them, i is the position of the current point, indicating that we want to calculate the total cost from the starting point to the i-th point; k is the index used to accumulate the path cost. By accumulating the path cost of each point, from 1 to i, the total cost from the starting point to the current point i is calculated; k-1 represents the index of the previous point of the current point k;

[0038] 3-2-2: Estimated cost h(i): The heuristic estimated cost from the current point Pi to the target point, defined as g(i) calculated in 3-4 above;

[0039] 3-2-3: Total cost f(i): The final cost is the sum of the actual cost and the heuristic estimated cost. The specific total cost calculation formula is f(i) = g(i) + G risk (i); where g(i) is the actual cost from the starting point to the point Pi, G risk (i) is the heuristic estimated cost from point Pi to the target point.

[0040] Furthermore, the specific process of dynamic path planning based on the heuristic function and the cost function is as follows:

[0041] 3-3-1: Initialize an open set to store the points to be expanded, starting from P start is added to the open set, and a closed set is initialized to store the points that have been visited; for the starting point P start Initialize as: g(P start )=0 and f(P start )=G risk (P start );

[0042] 3-3-2: Expand the points, specifically:

[0043] Select the point P with the minimum cost from the open set current , that is, choose argmin Pi∈openset f(i), P current Moving to the closed set, for P current Each adjacent point P neighbor , do the following:

[0044] If P neighbor If it is already in the closed set, skip it; calculate from the starting point P start To P neighbor The new cost is calculated as follows:new (P neighbor )=g(P current )+G weightde (P current ,P neighbor );

[0045] If P neighbor If it is not in the closed set, or the new cost is smaller, then update its cost and add it to the open set. The specific calculation formula is: new (P neighbor ) = g new (P current )+G risk (P neighbor ), update P neighbor The parent point is P current , from which the path can be traced.

[0046] Furthermore, based on the received optimal inspection path, the wheeled inspection robot is controlled to perform instrument panel recognition on the equipment at each point. The specific recognition process is as follows:

[0047] After receiving the best inspection path, the wheeled inspection robot performs the inspection task according to the planned path. The specific inspection task is to collect and identify data from the instrument panels of the equipment at each point of the best inspection path. The robot's navigation system uses sensors to perceive the environment in real time. During the movement, the wheeled inspection robot will track the path in real time and use obstacle detection algorithms to avoid obstacles.

[0048] When the wheeled inspection robot approaches the target equipment point, it accurately locates the location of the equipment dashboard through the visual recognition system and positioning algorithm. When it confirms that it has reached the target point, the wheeled inspection robot collects images through sensors and identifies the images to extract the parameters of the dashboard. Specific dashboard parameters include temperature, power, amplitude and noise values, thereby obtaining the temperature, power, amplitude and noise values ​​of the equipment at each point within the monitoring window.

[0049] Furthermore, the equipment is analyzed for early warning based on the instrument parameters of each point device to determine whether to trigger an early warning. The specific early warning analysis process is as follows:

[0050] The temperature, power, amplitude and noise values ​​are compared and analyzed with the set warning thresholds respectively. Specifically, the temperature, power, amplitude and noise values ​​at each monitoring moment in the monitoring period are calculated with the temperature warning threshold, power warning threshold, amplitude warning value and noise warning value respectively. The differences greater than zero are counted and recorded as the number of temperature warnings, the number of power warnings, the number of amplitude warnings and the number of noise warnings, and are recorded as N1, N2, N3 and N4 respectively; then the corresponding differences greater than zero are calculated by averaging to obtain the mean temperature warning value, the mean power warning value, the mean amplitude warning value and the mean noise warning value at each monitoring moment, and are recorded as M1, M2, M3 and M4 respectively;

[0051] Normalize N1, N2, N3, N4, M1, M2, M3 and M4 and take their values. Perform formulaic calculation and analysis on the values ​​to obtain the warning value Y of each point device. The specific calculation formula is:

[0052]

[0053] Among them, γ1, γ2, γ3, and γ4 are respectively set proportional constants, and the values ​​are set by personnel in this field according to actual needs;

[0054] Compare and analyze the warning value with the set warning threshold. If the warning value is greater than or equal to the set warning threshold, a warning signal is generated, and the location coordinates of the warning point in the plant layout map are obtained and sent to the corresponding engineer;

[0055] If the warning value is less than the set warning threshold, the device at the point will be recorded as a monitoring device, and the dashboard parameters of the monitoring device in the monitoring window will be sent to the server for storage as the data basis for planning the next round of inspection routes of the wheeled inspection robot.

[0056] In order to achieve the above object, according to another aspect of the present invention, a method for controlling instrument panel recognition based on a wheeled inspection robot is provided, the method comprising the following steps:

[0057] S1: Acquire instrument parameters of each device by communicating with the wheeled inspection robot. Specific instrument parameters include temperature, power, amplitude and noise value at each monitoring time during the monitoring cycle;

[0058] S2: Analyze and judge the status of each device based on the instrument parameters at each monitoring moment within the monitoring window to plan the inspection path of the wheeled robot; specifically:

[0059] Step 1: Retrieve the instrument parameters of each monitoring device. The specific instrument parameters include temperature, power, amplitude and noise value. Each device corresponds to a set of standard parameters. The specific standard parameters include standard temperature range and standard power range. Perform quantitative analysis based on the instrument parameters and standard parameters to obtain the status value of the monitoring device at each monitoring time.

[0060] Step 2: Construct a two-dimensional rectangular coordinate system with the monitoring time as the horizontal coordinate and the state value as the vertical coordinate, input the state value into the coordinate system according to its corresponding monitoring time, and record the position of the state value in the coordinate system as the state point, and use line segments to connect each state point in sequence to obtain a state value line graph of the equipment; further analyze the health trend status of the equipment based on the state value line graph to obtain the risk assessment index of the equipment;

[0061] Step 3: Thus, the risk assessment values ​​of several devices can be obtained, the plan layout of the plant area is obtained, and the instrument panels of each device are marked in the plan layout of the plant area according to their corresponding positions, thereby obtaining several points; the point closest to the initial position of the wheeled robot is taken as the starting point, thereby obtaining the constraint conditions of the inspection path: starting point - risk assessment value of each point - starting point; dynamic planning is generated according to the constraint conditions of the inspection path to generate the optimal inspection route, and it is sent to the instrument identification control module;

[0062] S3: Based on the received optimal inspection path, the wheeled inspection robot is controlled to perform instrument panel recognition on the equipment at each point to obtain the instrument parameters of the equipment at each point, and accordingly, an early warning analysis is performed on the equipment to determine whether an early warning is triggered.

[0063] Beneficial effects of the present invention:

[0064] The present invention collects and analyzes the instrument parameters (such as temperature, power, amplitude, noise value, etc.) of each monitoring device to determine the status of each monitoring device, generates a risk assessment index, and dynamically adjusts the inspection path based on the working status (risk assessment index) of the monitoring device. This method abandons the traditional fixed route method and can adjust the inspection priority according to the risk assessment index of the device. The device status is no longer fixed, but changes continuously during the inspection process, making the inspection path more flexible and intelligent.

[0065] The present invention integrates the status value and risk assessment index of the equipment into the dynamic path planning algorithm. When planning the inspection path, not only the spatial position (Euclidean distance) of the equipment is considered, but also the health risk (risk assessment index) of the equipment is incorporated into the decision-making process. Then, based on the definition of weighted distance and heuristic function, high-risk equipment can be dynamically selected and the inspection route can be optimized so that the robot can first reach the location of the equipment that needs urgent inspection, thereby improving the inspection efficiency and response speed, and realizing the flexible adjustment of the inspection strategy according to the health status of the on-site equipment to cope with complex on-site changes.

[0066] The present invention controls the wheeled inspection robot to identify the instrument panel of each point through the received optimal inspection route, and obtains the instrument parameters of the equipment at each point, which not only improves the pertinence of the inspection, but also avoids unnecessary redundant inspections, provides adaptive path adjustment and monitoring functions of the equipment instrument panel, so that the robot can continuously and efficiently perform tasks in complex scenarios; then, early warning analysis is performed on the equipment according to the instrument parameters of the instrument panel, and equipment early warning is triggered accordingly; the equipment that does not trigger the early warning is recorded as monitoring equipment, and its corresponding instrument parameters are sent to the server for storage, which serves as the data basis for the next round of inspection route planning of the wheeled inspection robot, thereby realizing rapid identification of abnormalities and early warning, and significantly improving the intelligent level of equipment management. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0068] Figure 1 is a schematic diagram of system module connection of the present invention;

[0069] Figure 2 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0070] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0071] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0072] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0073] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0074] According to an embodiment of the present invention, Figure 1 As shown, a dashboard recognition control system based on a wheeled inspection robot is provided, the system comprising: a server, an instrument recognition module and an inspection control module;

[0075] The server communicates with the wheeled inspection robot in the factory area and the sensors mounted on the wheeled inspection robot to obtain the instrument parameters of each monitoring device. The specific instrument parameters include the temperature, power, amplitude and noise value at each monitoring time within the monitoring window;

[0076] The inspection intelligent planning module analyzes and judges the status of each device based on the instrument parameters at each monitoring moment in the monitoring window to plan the inspection path of the wheeled robot, so as to identify and upload the dashboard of the monitoring equipment in a timely manner; specifically:

[0077] Step 1: Obtain instrument parameters of each monitoring device, including temperature, power, amplitude and noise value, and record them as Tj, Pj, Dj and Zj respectively; where j = 1, 2, 3 ... J, J is a positive integer, J represents the total number of monitoring moments, and j represents the sequence number of any monitoring moment;

[0078] Assume that each device corresponds to a set of standard parameters, including standard temperature range and standard power range, which are recorded as [HT1, HT2] and [HP1, HP2] respectively. It should be noted that standard parameters refer to a set of ideal, predefined working ranges or standard values ​​that the device should follow under normal working conditions. The specific standard temperature range refers to the working temperature of the device should be maintained within a predetermined range during normal operation. The temperature range is usually defined by the device manufacturer or industry standards to ensure that the device can operate within a suitable temperature range to avoid damage caused by overheating or low temperature. The standard power range refers to the power consumed or output power of the device should be within a specific power range under normal operation. The standard power range is usually defined based on the design specifications, load requirements and expected working conditions of the device.

[0079] The temperature Tj, power Pj, amplitude Dj, noise value Zj, standard temperature range [HT1, HT2] and standard power range [HP1, HP2] are normalized and their values ​​are taken. The values ​​are calculated and analyzed by formula to obtain the state value Uj of the equipment at each monitoring time. The specific calculation formula is:

[0080]

[0081] Wherein a1, a2, a3, and a4 are respectively set proportional constants, and their values ​​are set by those skilled in the art according to actual conditions; dTj represents the shortest distance between temperature Tj and the standard temperature interval [HT1, HT2]. If temperature Tj belongs to the standard temperature interval [HT1, HT2], dTj=0; if temperature Tj is greater than the maximum value in the standard temperature interval [HT1, HT2], dTj=Tj-HT2; if temperature Tj is less than the minimum value in the standard temperature interval [HT1, HT2], dTj=HT1-Tj; similarly, dPj represents the shortest distance between power Pj and the standard temperature interval [HP1, HP2]. If power When Pj belongs to the standard power interval [HP1, HP2], dPj = 0; if the power Pj is greater than the maximum value in the standard power interval [HP1, HP2], dPj = Pj-HP2; if the power Pj is less than the minimum value in the standard power interval [HP1, HP2], dPj = HP1-Pj; From the formula, it can be seen that when the shortest distance between the temperature Tj and the standard temperature interval [HT1, HT2] is larger, it means that the device state is worse, and the state value is larger; when the shortest distance between the power Pj and the standard power interval [HP1, HP2] is larger, it means that the device state is worse, and the state value is larger; when the amplitude is larger and the noise value is larger, it means that the device operation is unstable, and the state value is larger;

[0082] Step 2: Use the monitoring time as the horizontal coordinate and the state value as the vertical coordinate to construct a two-dimensional rectangular coordinate system. Input the state value into the coordinate system according to its corresponding monitoring time, and record the position of the state value in the coordinate system as the state point. Use line segments to connect each state point in sequence to obtain the state value line graph of the equipment; two adjacent state points can form a state line segment, so that several state line segments can be segmented from the state value line graph, and the state values ​​corresponding to the two endpoints of the state line segment (the two adjacent state points in the original state value line graph) are averaged to obtain the line segment value recorded as The slope of the state line segment calculated by data fitting is recorded as It should be noted that for a certain state line segment, the larger the line segment value, the more abnormal risk the device itself has; the larger the slope, the more the device state is increasing, which means the abnormal risk of the device is increasing;

[0083] Set the segment value of each state segment and slope Perform normalization and take its value, and perform formulaic calculation and analysis on the value to obtain the risk assessment index KU of the equipment; the specific calculation formula is:

[0084]

[0085] Where α1 and α2 are respectively set proportional constants, and their specific values ​​are set by technicians in this field according to actual needs; It can be seen from the formula that when the line segment value When the slope is larger, it means that there is an abnormal risk in the state of the device itself, and the risk assessment index of the device is larger; when the slope is When greater than zero, if The larger the slope, the greater the risk assessment index of the device. When it is less than zero, The larger the absolute value of , the lower the risk assessment index of the device.

[0086] Step 3: Thus, the risk assessment values ​​of several devices can be obtained, and the floor plan of the plant area is obtained. The instrument panels of each device are marked in the floor plan of the plant area according to their corresponding positions. Thus, several points can be obtained. Each specific point represents the instrument panel of a device. The wheeled inspection robot needs to identify the instrument panel of each point; the point closest to the initial position of the wheeled robot is used as the starting point, and thus the constraint conditions of the inspection path can be obtained: starting point - risk assessment value of each point - starting point; according to the constraint conditions of the inspection path, dynamic planning is generated to generate the most suitable inspection route, which is specifically:

[0087] 3-1: Heuristic function definition:

[0088] 3-1-1: Let the starting point be P start , and the other points are recorded as Pi, where Pi represents any point in the plant layout diagram, i = 1, 2, 3 ... I, I is a positive integer, and I represents the point after removing the starting point P start The number of all points other than , i represents any one of them; specifically, the corresponding position coordinates are generated according to their positions in the plane layout diagram of the factory area, marked as (xi, yi), and their corresponding risk assessment values ​​are recorded as KUi;

[0089] 3-1-2: Record the next point as the target point. The next point (i.e. the target point) obtained by performing inspection path planning at the current point can be used to dynamically plan the instrument panel recognition path of the wheeled inspection robot. Specifically:

[0090] The current point Pi and the target point P are obtained by the Euclidean distance calculation formula. target The straight-line distance between them is calculated by the specific Euclidean distance formula: Where (x target ,y target ) is the position coordinate of the target point (i.e., the next point of the current wheeled inspection robot);

[0091] 3-1-3: In order to incorporate device risk into the heuristic search, a heuristic function is defined to calculate the weighted distance. The specific weighted distance calculation formula is:

[0092] G risk (i) = h(i) × (1 + w × KUi)

[0093] Among them, w is a weight coefficient for adjusting the impact of risk on path selection, and its specific value is set by the personnel in this field. KUi is the risk assessment index of the device represented by point i; h(i) is the straight-line distance between the device and the target device. Through this heuristic function, it is more inclined to select high-risk devices. Specific technicians can adjust w to adjust the proportion of straight-line distance and risk assessment index to weighted distance;

[0094] 3-2: Cost function definition:

[0095] 3-2-1: Actual cost g(i): from the starting point P start The cost to the current point Pi usually refers to the cost from the starting point P start The cumulative weighted distance to the current point is expressed as follows:

[0096]

[0097] Among them, i is the position of the current point, indicating that we want to calculate the total cost from the starting point to the i-th point; k is the index used to accumulate the path cost. By accumulating the path cost of each point, from 1 to i, the total cost from the starting point to the current point i is calculated; k-1 represents the previous point of the current point k, and the cost between the previous point and the current point is added to the total cost;

[0098] 3-2-2: Estimated cost h(i): The heuristic estimated cost from the current point Pi to the target point, defined as g(i) calculated in 3-4 above;

[0099] 3-2-3: Total cost f(i): The final cost is the sum of the actual cost and the heuristic estimated cost. The specific total cost calculation formula is f(i) = g(i) + G risk (i); where g(i) is the actual cost from the starting point to the point Pi, G risk (i) is the heuristic estimated cost from point Pi to the target point;

[0100] 3-3: Dynamic path planning is performed based on the heuristic function and the cost function, specifically:

[0101] 3-3-1: Initialize an open set to store the points to be expanded, starting from P start is added to the open set, and a closed set is initialized to store the points that have been visited; for the starting point P start Initialize as: g(P start )=0 and f(P start )=G risk (P start );

[0102] 3-3-2: Expand the points, specifically:

[0103] Select the point P with the minimum cost from the open set current , that is, choose argmin Pi∈openset f(i), P current Moving to the closed set, for P current Each adjacent point P neighbor , do the following:

[0104] If P neighbor If it is already in the closed set, skip it; calculate from the starting point P start To P neighbor The new cost is calculated as follows: new (P neighbor )=g(P current )+G weightde (Pcurrent ,P neighbor );

[0105] If P neighbor If it is not in the closed set, or the new cost is smaller, then update its cost and add it to the open set. The specific calculation formula is: new (P neighbor ) = g new (P current )+G risk (P neighbor ), update P neighbor The parent point is P current , from which the path can be traced;

[0106] 3-4: If the end point P end (The specific end point and the starting point are the same point) are added to the closed set, indicating that the best path has been found. Starting from the end point, the parent point is traced back to obtain the best path. It should be noted that if the risk assessment index KUi of a device at a certain point is updated during the inspection process, when the path is replanned, the updated risk assessment index KUi is substituted into the heuristic function G risk (i), and execute 3-3 to dynamically adjust the inspection path and send it to the instrument identification control module;

[0107] The instrument parameters (such as temperature, power, amplitude, noise value, etc.) of each monitoring device are collected and analyzed to determine the status of each monitoring device, generate a risk assessment index, and dynamically adjust the inspection path based on the working status (risk assessment index) of the monitoring device. This method abandons the traditional fixed route method and can adjust the inspection priority according to the risk assessment index of the equipment. The equipment status is no longer fixed, but changes with the inspection process, making the inspection path more flexible and intelligent; the equipment status value and risk assessment index are integrated into the dynamic path planning algorithm. When planning the inspection path, not only the spatial position of the equipment (Euclidean distance) is considered, but also the health risk of the equipment (risk assessment index) is included in the decision-making process. Then, based on the definition of weighted distance and heuristic function, high-risk equipment can be dynamically selected and the inspection route can be optimized, so that the robot can first reach the location of the equipment that needs urgent inspection, thereby improving the inspection efficiency and response speed, and realizing the flexible adjustment of the inspection strategy according to the health status of the on-site equipment to cope with complex on-site changes.

[0108] The instrument recognition control module controls the wheeled inspection robot to recognize and control the instruments of each point device based on the received optimal inspection route. The specific recognition control process is as follows:

[0109] After receiving the best inspection path, the wheeled inspection robot starts to perform the inspection task according to the planned path. The robot's navigation system uses sensors (such as lidar, camera, IMU, etc.) to perceive the environment in real time to ensure smooth movement along the planned path. During the movement, the wheeled inspection robot will track the path in real time and use obstacle detection algorithms (such as SLAM) to avoid obstacles to ensure the correctness of the inspection path and the efficiency of the inspection.

[0110] When the wheeled inspection robot approaches the target equipment point, it accurately locates the location of the equipment dashboard through the visual recognition system (such as laser radar, RGBD camera) and positioning algorithm (such as visual SLAM, GPS positioning, etc.). When it is confirmed that it has reached the target point, the wheeled inspection robot uses sensors (such as cameras, infrared sensors) to collect images (the image collection here is to collect one image at each monitoring moment within a monitoring window), and recognizes the image to extract the parameters of the dashboard. The specific dashboard parameters include temperature, power, amplitude and noise values, thereby obtaining the temperature, power, amplitude and noise values ​​of the equipment at each point within the monitoring window; it should be noted that the specific duration of the monitoring window is set by the technicians in this field according to actual needs;

[0111] The temperature, power, amplitude and noise values ​​are compared and analyzed with the set warning thresholds (specifically, the temperature warning threshold, the power warning threshold, the amplitude warning value and the noise warning value). Specifically, the temperature, power, amplitude and noise values ​​at each monitoring moment in the monitoring period are calculated with the temperature warning threshold, the power warning threshold, the amplitude warning value and the noise warning value, and the differences greater than zero are counted as the number of temperature warnings, the number of power warnings, the number of amplitude warnings and the number of noise warnings, and are recorded as N1, N2, N3 and N4 respectively; the corresponding differences greater than zero are calculated by average to obtain the mean temperature warning, the mean power warning, the mean amplitude warning and the mean noise warning at each monitoring moment, and are recorded as M1, M2, M3 and M4 respectively; N1, N2, N3, N4, M1, M2, M3 and M4 are normalized and their values ​​are taken, and the values ​​are calculated and analyzed by formula to obtain the warning value Y of each point device. The specific calculation formula is:

[0112]

[0113] Among them, γ1, γ2, γ3, and γ4 are respectively set proportional constants, and the values ​​are set by personnel in this field according to actual needs; the warning value is compared and analyzed with the set warning threshold. If the warning value is greater than or equal to the set warning threshold, it means that the abnormal state of the equipment at the point is obvious, then a warning signal is generated, and the position coordinates of the warning point in the plant layout map are obtained and sent to the corresponding engineer; if the warning value is less than the set warning threshold, the equipment at the point is recorded as a monitoring device, and the dashboard parameters of the monitoring device in the monitoring window are sent to the server for storage, so as to serve as the data basis for planning the next round of inspection routes of the wheeled inspection robot;

[0114] By receiving the best inspection route, the wheeled inspection robot is controlled to identify the instrument panel at each point and obtain the instrument parameters of the equipment at each point. This not only improves the targeted inspection, but also avoids unnecessary redundant inspections, provides adaptive path adjustment and monitoring functions for the equipment instrument panel, so that the robot can continue to perform tasks efficiently in complex scenarios; then, the equipment is analyzed for early warning based on the instrument parameters of the instrument panel, and the equipment early warning is triggered accordingly; the equipment that does not trigger the early warning is recorded as monitoring equipment, and its corresponding instrument parameters are sent to the server for storage, which serves as the data basis for the next round of inspection route planning of the wheeled inspection robot, realizing rapid identification of anomalies and early warning, significantly improving the level of intelligent equipment management.

[0115] According to an embodiment of the present invention, Figure 2 As shown, a dashboard recognition control method based on a wheeled inspection robot is also provided, and the method comprises the following steps:

[0116] S1: communicate with each wheeled inspection robot to obtain the instrument parameters of each device, including the temperature, power, amplitude and noise values ​​corresponding to each monitoring moment in the specific instrument parameter monitoring window;

[0117] S2: Perform status analysis on the monitoring equipment based on the instrument parameters of each monitoring equipment to generate a risk assessment index, and plan the inspection route of the wheeled inspection robot accordingly; specifically:

[0118] S1: Acquire instrument parameters of each device by communicating with the wheeled inspection robot. Specific instrument parameters include temperature, power, amplitude and noise value at each monitoring time during the monitoring cycle;

[0119] S2: Analyze and judge the status of each device based on the instrument parameters at each monitoring moment within the monitoring window to plan the inspection path of the wheeled robot; specifically:

[0120] Step 1: Retrieve the instrument parameters of each monitoring device. The specific instrument parameters include temperature, power, amplitude and noise value. Each device corresponds to a set of standard parameters. The specific standard parameters include standard temperature range and standard power range. Perform quantitative analysis based on the instrument parameters and standard parameters to obtain the status value of the monitoring device at each monitoring time.

[0121] Step 2: Construct a two-dimensional rectangular coordinate system with the monitoring time as the horizontal coordinate and the state value as the vertical coordinate, input the state value into the coordinate system according to its corresponding monitoring time, and record the position of the state value in the coordinate system as the state point, and use line segments to connect each state point in sequence to obtain a state value line graph of the equipment; further analyze the health trend status of the equipment based on the state value line graph to obtain the risk assessment index of the equipment;

[0122] Step 3: Thus, the risk assessment values ​​of several devices can be obtained, the plan layout of the plant area is obtained, and the instrument panels of each device are marked in the plan layout of the plant area according to their corresponding positions, thereby obtaining several points; the point closest to the initial position of the wheeled robot is taken as the starting point, thereby obtaining the constraint conditions of the inspection path: starting point - risk assessment value of each point - starting point; dynamic planning is generated according to the constraint conditions of the inspection path to generate the optimal inspection route, and it is sent to the instrument identification control module;

[0123] S3: Based on the received optimal inspection path, the wheeled inspection robot is controlled to perform instrument panel recognition on the equipment at each point to obtain the instrument parameters of the equipment at each point, and accordingly, an early warning analysis is performed on the equipment to determine whether an early warning is triggered.

[0124] The above are only embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. The instrument panel recognition control system based on the wheeled inspection robot is characterized by: include: Server, inspection intelligent planning module and instrument identification control module; The server communicates with the wheeled inspection robot to obtain instrument parameters of each device; The inspection intelligent planning module analyzes and judges the status of each device based on the instrument parameters at each monitoring moment in the monitoring window to plan the inspection path of the wheeled robot; specifically: Step 1: Retrieve the instrument parameters of each monitoring device. The specific instrument parameters include temperature, power, amplitude and noise value. Each device corresponds to a set of standard parameters. The specific standard parameters include standard temperature range and standard power range. Perform quantitative analysis based on the instrument parameters and standard parameters to obtain the status value of the monitoring device at each monitoring time. Step 2: Construct a two-dimensional rectangular coordinate system with the monitoring time as the horizontal coordinate and the state value as the vertical coordinate, input the state value into the coordinate system according to its corresponding monitoring time, and record the position of the state value in the coordinate system as the state point, and use line segments to connect each state point in sequence to obtain a state value line graph of the equipment; further analyze the health trend status of the equipment based on the state value line graph to obtain the risk assessment index of the equipment; Step 3: Thus, the risk assessment values ​​of several devices can be obtained, the plan layout of the plant area is obtained, and the instrument panels of each device are marked in the plan layout of the plant area according to their corresponding positions, thereby obtaining several points; the point closest to the initial position of the wheeled robot is taken as the starting point, thereby obtaining the constraint conditions of the inspection path: starting point - risk assessment value of each point - starting point; dynamic planning is generated according to the constraint conditions of the inspection path to generate the optimal inspection route, and it is sent to the instrument identification control module; The instrument recognition control module controls the wheeled inspection robot to perform instrument panel recognition on the equipment at each point based on the received optimal inspection path to obtain the instrument parameters of the equipment at each point, and performs early warning analysis on the equipment to determine whether to trigger an early warning.

2. The instrument panel recognition control system based on the wheeled inspection robot according to claim 1 is characterized in that: The specific process of quantitative analysis based on instrument parameters and standard parameters is as follows: Retrieve the instrument parameters of each monitoring device, including temperature, power, amplitude and noise value, and record them as Tj, Pj, Dj and Zj respectively; where j = 1, 2, 3 ... J, J is a positive integer, J represents the total number of monitoring moments, and j represents the sequence number of any monitoring moment; each device corresponds to a set of standard parameters, including standard temperature range and standard power range, and record them as [HT1, HT2] and [HP1, HP2] respectively; The temperature Tj, power Pj, amplitude Dj, noise value Zj, standard temperature range [HT1, HT2] and standard power range [HP1, HP2] are normalized and their values ​​are taken. The values ​​are calculated and analyzed by formula to obtain the state value Uj of the equipment at each monitoring time. The specific calculation formula is: Wherein a1, a2, a3, and a4 are respectively set proportional constants, and their values ​​are set by those skilled in the art according to actual conditions; dTj represents the shortest distance between temperature Tj and the standard temperature interval [HT1, HT2]. If temperature Tj belongs to the standard temperature interval [HT1, HT2], dTj=0; if temperature Tj is greater than the maximum value in the standard temperature interval [HT1, HT2], dTj=Tj-HT2; if temperature Tj is less than the standard temperature interval [HT1, HT2], dTj=HT1-Tj; similarly, dPj represents the shortest distance between power Pj and the standard temperature interval [HP1, HP2]. If power Pj belongs to the standard power interval [HP1, HP2], dPj=0; if power Pj is greater than the maximum value in the standard power interval [HP1, HP2], dPj=Pj-HP2; if power Pj is less than the minimum value in the standard power interval [HP1, HP2], dPj=HP1-Pj.

3. The instrument panel recognition control system based on the wheeled inspection robot according to claim 2 is characterized in that: The specific process of further analyzing the health trend status of the equipment based on the status value line chart is as follows: Two adjacent state points can form a state line segment, so several state line segments can be segmented from the state value line graph, and the state values ​​corresponding to the two end points of the state line segment are averaged to obtain the line segment value recorded as The slope of the state line segment calculated by data fitting is recorded as Set the segment value of each state segment and slope Perform normalization and take its value, and perform formulaic calculation and analysis on the value to obtain the risk assessment index KU of the equipment; the specific calculation formula is: α1 and α2 are respectively set proportional constants, and their specific values ​​are set by those skilled in the art according to actual needs.

4. The instrument panel recognition control system based on the wheeled inspection robot according to claim 3 is characterized in that: Generate dynamic planning based on the constraints of the inspection path to generate the best inspection route. The specific steps are as follows: 3-1: Define the heuristic function: the starting point is denoted as P start , the current point Pi and the target point P target The straight-line distance h(i) between them, the weighted distance G risk (i); 3-2: Cost function definition: actual cost g(i), estimated cost h(i) and total cost f(i); 3-3: Dynamic path planning based on heuristic function and cost function; 3-4: If the end point P end When it is added to the closed set, it means that the best path has been found. Then, starting from the end point, the parent point is traced back to obtain the best path and send it to the instrument identification control module.

5. The instrument panel recognition control system based on the wheeled inspection robot according to claim 4 is characterized in that: The specific process of heuristic function definition is: 3-1-1: Let the starting point be P start , and the other points are recorded as Pi, where Pi represents any point in the plant layout diagram, i = 1, 2, 3 ... I, I is a positive integer, and I represents the point after removing the starting point P start The number of all points other than , i represents any one of them; specifically, the corresponding position coordinates are generated according to their positions in the plane layout diagram of the factory area, marked as (xi, yi), and their corresponding risk assessment values ​​are recorded as KUi; 3-1-2: Record the next point as the target point. The next point obtained by performing inspection path planning at the current point can be used to dynamically plan the instrument panel recognition path of the wheeled inspection robot. Specifically: The current point Pi and the target point P are obtained by the Euclidean distance calculation formula. target The straight-line distance between them is calculated by the specific Euclidean distance formula: Where (x target ,y target ) is the position coordinate of the target point; 3-1-3: Define a heuristic function to calculate the weighted distance. The specific formula for calculating the weighted distance is: G risk (i)=h(i)×(1+w×KUi) Among them, w is a weight coefficient that adjusts the impact of risk on path selection, and its specific value is set by personnel in this field. KUi is the risk assessment index of the device represented by point i; h(i) is the straight-line distance between the device and the target device.

6. The instrument panel recognition control system based on the wheeled inspection robot according to claim 5 is characterized in that: The specific process of cost function definition is: 3-2-1: Actual cost g(i): from the starting point P start The cost to the current point Pi is expressed in the following formula: Among them, i is the position of the current point, indicating that we want to calculate the total cost from the starting point to the i-th point; k is the index used to accumulate the path cost. By accumulating the path cost of each point, from 1 to i, the total cost from the starting point to the current point i is calculated; k-1 represents the index of the previous point of the current point k; 3-2-2: Estimated cost h(i): The heuristic estimated cost from the current point Pi to the target point, defined as g(i) calculated in 3-4 above; 3-2-3: Total cost f(i): The final cost is the sum of the actual cost and the heuristic estimated cost. The specific total cost calculation formula is f(i) = g(i) + G risk (i); where g(i) is the actual cost from the starting point to the point Pi, G risk (i) is the heuristic estimated cost from point Pi to the target point.

7. The instrument panel recognition control system based on the wheeled inspection robot according to claim 6 is characterized in that: The specific process of dynamic path planning based on heuristic function and cost function is as follows: 3-3-1: Initialize an open set to store the points to be expanded, starting from P start is added to the open set, and a closed set is initialized to store the points that have been visited; for the starting point P start Initialize as: g(P start )=0 and f(P start )=G risk (P start ); 3-3-2: Expand the points, specifically: Select the point P with the minimum cost from the open set current , that is, choose argmin Pi∈openset f(i), P current Moving to the closed set, for P current Each adjacent point P neighbor , do the following: If P neighbor If it is already in the closed set, skip it; calculate from the starting point P start To P neighbor The new cost is calculated as follows: new (P neighbor )=g(P current )+G weightde (P current ,P neighbor ); If P neighbor If it is not in the closed set, or the new cost is smaller, then update its cost and add it to the open set. The specific calculation formula is: new (P neighbor ) = g new (P current )+G risk (P neighbor ), update P neighbor The parent point is P current , from which the path can be traced.

8. The instrument panel recognition control system based on the wheeled inspection robot according to claim 1 is characterized in that: Based on the received optimal inspection path, the wheeled inspection robot is controlled to identify the instrument panel of the equipment at each point. The specific identification process is as follows: After receiving the best inspection path, the wheeled inspection robot performs the inspection task according to the planned path. The specific inspection task is to collect and identify data from the instrument panels of the equipment at each point of the best inspection path. The robot's navigation system uses sensors to perceive the environment in real time. During the movement, the wheeled inspection robot will track the path in real time and use obstacle detection algorithms to avoid obstacles. When the wheeled inspection robot approaches the target equipment point, it accurately locates the location of the equipment dashboard through the visual recognition system and positioning algorithm. When it confirms that it has reached the target point, the wheeled inspection robot collects images through sensors and identifies the images to extract the parameters of the dashboard. Specific dashboard parameters include temperature, power, amplitude and noise values, thereby obtaining the temperature, power, amplitude and noise values ​​of the equipment at each point within the monitoring window.

9. The instrument panel recognition control system based on the wheeled inspection robot according to claim 8 is characterized in that: According to the instrument parameters of each point equipment, the equipment is analyzed for early warning to determine whether to trigger an early warning. The specific early warning analysis process is as follows: The temperature, power, amplitude and noise values ​​are compared and analyzed with the set warning thresholds respectively. Specifically, the temperature, power, amplitude and noise values ​​at each monitoring moment in the monitoring period are calculated with the temperature warning threshold, power warning threshold, amplitude warning value and noise warning value respectively. The differences greater than zero are counted and recorded as the number of temperature warnings, the number of power warnings, the number of amplitude warnings and the number of noise warnings. Then, the corresponding differences greater than zero are averaged to obtain the temperature warning mean, power warning mean, amplitude warning mean and noise warning mean at each monitoring moment; The number of temperature warnings, the number of power warnings, the number of amplitude warnings, the number of noise warnings, the mean value of temperature warnings, the mean value of power warnings, the mean value of amplitude warnings, and the mean value of noise warnings are normalized and their values ​​are taken, and the values ​​are calculated and analyzed by formula to obtain the warning value of each point device; Compare and analyze the warning value with the set warning threshold. If the warning value is greater than or equal to the set warning threshold, a warning signal is generated, and the location coordinates of the warning point in the plant layout map are obtained and sent to the corresponding engineer; If the warning value is less than the set warning threshold, the device at the point will be recorded as a monitoring device, and the dashboard parameters of the monitoring device in the monitoring window will be sent to the server for storage as the data basis for planning the next round of inspection routes of the wheeled inspection robot.

10. The instrument panel recognition control method based on the wheeled inspection robot is characterized by Applied to the dashboard recognition control system based on the wheeled inspection robot according to any one of claims 1 to 9, the method comprises the following steps: S1: Acquire instrument parameters of each device by communicating with the wheeled inspection robot. Specific instrument parameters include temperature, power, amplitude and noise value at each monitoring time during the monitoring cycle; S2: Analyze and judge the status of each device based on the instrument parameters at each monitoring moment within the monitoring window to plan the inspection path of the wheeled robot; specifically: Step 1: Retrieve the instrument parameters of each monitoring device. The specific instrument parameters include temperature, power, amplitude and noise value. Each device corresponds to a set of standard parameters. The specific standard parameters include standard temperature range and standard power range. Perform quantitative analysis based on the instrument parameters and standard parameters to obtain the status value of the monitoring device at each monitoring time. Step 2: Construct a two-dimensional rectangular coordinate system with the monitoring time as the horizontal coordinate and the state value as the vertical coordinate, input the state value into the coordinate system according to its corresponding monitoring time, and record the position of the state value in the coordinate system as the state point, and use line segments to connect each state point in sequence to obtain a state value line graph of the equipment; further analyze the health trend status of the equipment based on the state value line graph to obtain the risk assessment index of the equipment; Step 3: Thus, the risk assessment values ​​of several devices can be obtained, the plan layout of the plant area is obtained, and the instrument panels of each device are marked in the plan layout of the plant area according to their corresponding positions, thereby obtaining several points; the point closest to the initial position of the wheeled robot is taken as the starting point, thereby obtaining the constraint conditions of the inspection path: starting point - risk assessment value of each point - starting point; dynamic planning is generated according to the constraint conditions of the inspection path to generate the optimal inspection route, and it is sent to the instrument identification control module; S3: Based on the received optimal inspection path, the wheeled inspection robot is controlled to perform instrument panel recognition on the equipment at each point to obtain the instrument parameters of the equipment at each point, and accordingly, an early warning analysis is performed on the equipment to determine whether an early warning is triggered.

Images