Path Optimization Method, Device, and Storage Medium for Patrol Robot

Abstract

The present invention provides a path optimization method, device, and storage medium for an inspection robot. The method includes: detecting each point according to an initial detection path to obtain an initial detection result, where the initial detection result includes at least one detection index; re-planning an inspection path according to the initial detection result, optimizing the initial detection path, and determining a target detection path; and re-detecting each point according to the target detection path to obtain a target detection result. This method improves the accuracy and efficiency of path optimization, can achieve fully intelligent and unmanned inspection, and improves safety.

Description

Technical Field

[0001] The present invention relates to the field of hazardous chemical inspection, and particularly to a path optimization method, device, and storage medium for an inspection robot. Background Art

[0002] Currently, the methods for hazardous chemical inspection mainly include manual inspection and robot inspection. Manual inspection requires high qualities of employees. They need to be able to promptly detect the leakage points of toxic gases and the ignition points prone to catching fire, and be able to notify in a timely manner. However, the cost of manual inspection is high, and it cannot conduct inspections 24 hours a day. Moreover, after working for a long time, employees are prone to fatigue, resulting in problems such as omissions or negligence, and the inspection efficiency is low.

[0003] Existing ordinary robot inspections are to repeatedly check in a fixed manner along a pre-designed fixed route in the park to see if there are any special situations. For example, the patent announcement of "An Automatic Hazardous Chemical Inspection Robot" with the publication number CN209774657 discloses several methods for detecting toxic / combustible gases in chemical plants, but it has the following defects: (1) The inspection path of this inspection robot is fixed and cannot change the inspection path according to the changes of various experimental equipment. It does not intelligently process the detection results but transmits them as data, still requiring manual inspection and not being fully intelligent and unmanned. (2) This inspection robot cannot increase or decrease the number of inspections according to the change of the risk coefficient of chemical products. It only detects corresponding parameters on the fixed path and transmits them in real time to update the data dynamically in real time. (3) This inspection robot cannot promptly handle emergencies, cannot promptly issue an alarm and optimize the inspection route, bypass the area where danger is occurring, and may still pass through the area where no danger has occurred, or may cause damage to itself. Summary of the Invention

[0004] In view of the above problems, the present invention provides a path optimization method, device, and storage medium for an inspection robot, which can achieve fully intelligent and unmanned inspection and improve safety.

[0005] To achieve the above object, on the one hand, the present invention provides a path optimization method for an inspection robot, including:

[0006] Detecting each point according to an initial detection path to obtain an initial detection result, where the initial detection result includes at least one detection index;

[0007] Re-planning an inspection path according to the initial detection result, optimizing the initial detection path, and determining a target detection path;

[0008] Detecting each point again according to the target detection path to obtain a target detection result.

[0009] In some embodiments, the method for determining a target detection path includes:

[0010] Conduct safety assessment on each point according to each test indicator, and obtain the first safety factor of each point under the test indicator condition;

[0011] The second safety factor of each point is obtained by integrating the first safety factor of each point under all test indicators;

[0012] Sort the second safety factors of all sites in ascending order,

[0013] According to the sorting order, the points corresponding to the ones with smaller second safety factors are taken as priority inspection targets, and the inspection paths are replanned to obtain the target detection paths.

[0014] In some embodiments, the method for determining a target detection path further comprises:

[0015] Establish a gravitational field for the target point and a repulsive field for the obstacle, and calculate the gravitational and repulsive forces at the current point of the inspection robot;

[0016] Determine whether the inspection robot is currently in a detour behavior,

[0017] If the inspection robot is in a detour behavior, it is further determined whether the inspection robot has left the local minimum area. If it has left, the inspection robot moves along a first path until it reaches the target point. The first path is a non-detour route that moves along the direction of the combined force of attraction and repulsion to reach the target point. If it has not left, the inspection robot moves along a second path and continues the detour behavior until it leaves the local minimum area. The second path is a detour route that moves along the equipotential lines of the repulsive field to leave the local minimum area.

[0018] If the robot is not in a detour, it is further determined whether a local minimum area is encountered. If so, the inspection robot moves along the second path and begins a detour until it leaves the local minimum area. If not, the inspection robot moves along the first path until it reaches the target point.

[0019] The target detection path is updated according to the first path and the second path.

[0020] In some embodiments, a first discrimination condition is constructed using the combined force of the attraction and repulsion at the current position of the inspection robot, and the first discrimination condition is used to determine whether the robot encounters a local minimum area;

[0021] The first discrimination condition is set as:

[0022]

[0023] Among them, Uatt represents gravity, U rep,j Represents repulsive force; ε is a very small positive number, which means that the virtual net force acting on the inspection robot is close to 0. If the net force of attraction and repulsion at the current position of the inspection robot is less than ε, it means that a local minimum area is encountered.

[0024] In some embodiments, the movement distance of the inspection robot from a previous point to a current point is used to construct a second discrimination condition, and the second discrimination condition is used to determine whether the robot encounters a local minimum area;

[0025] The second discrimination condition is set as:

[0026] |xx A |<αs A

[0027] Among them, α is a positive number between 0 and 1, x A is a previous point during the movement of the inspection robot, s A Indicates that the inspection robot moves from the previous point x A The total distance to reach the current point x.

[0028] In some embodiments, the gravitational force at the current position of the inspection robot is determined based on the distance between the current position of the inspection robot and the target position. The gravitational force is expressed as:

[0029]

[0030] Among them, U att (x) represents the gravitational force of the inspection robot at the current point x, δ represents the relative influence of the gravitational potential; x represents the current point of the inspection robot, G represents the target point, d(x, G) represents the distance between the current point and the target point, Its function is to weaken the gravitational potential of the target when the inspection robot is far away from the target point.

[0031] In some embodiments, the repulsive force at the current position of the inspection robot is determined based on the distance between the current position of the inspection robot and the obstacle and the repulsive force potential of the obstacle. The repulsive force is expressed as

[0032]

[0033] Among them, U rep,j (x) represents the repulsive force of the inspection robot at the current point x, μ j represents the relative influence of the repulsive potential of the j-th obstacle, d j (x) represents the distance between the robot’s current position x and the j-th obstacle, represents the range of the repulsive potential of the j-th obstacle.

[0034] In some embodiments, the detection index includes one of the information on the concentration of toxic or combustible gases, the temperature information of an object, and the instrument reading information.

[0035] In some embodiments, the information on the concentration of toxic or combustible gases is the first detection index;

[0036] Performing a safety assessment on each site according to each detection index to obtain the first safety factor of each site under the condition of this detection index includes:

[0037] Collecting the information on the concentration of toxic or combustible gases at each site according to the position change of the patrol robot;

[0038] Performing a safety assessment on each site according to the information on the concentration of toxic or combustible gases at each site, and determining the first safety factor of each site under the first detection index.

[0039] In some embodiments, an infrared gas detection method is adopted to collect the information on the concentration of toxic or combustible gases at each site.

[0040] In some embodiments, if, under the first detection index, the first safety factor of a site exceeds a first threshold, a gas concentration alarm message is generated, and the first threshold is a gas concentration threshold.

[0041] In some embodiments, the temperature information of the object is the second detection index;

[0042] Performing a safety assessment on each site according to each detection index to obtain the first safety factor of each site under the condition of this detection index includes:

[0043] Collecting the temperature information of the object at each site according to the position change of the patrol robot;

[0044] Performing a safety assessment on each site according to the temperature information of the object at each site, and determining the first safety factor of each site under the second detection index.

[0045] In some embodiments, the temperature information of the object at each site is collected by a temperature sensor thermal imaging camera arranged on site.

[0046] In some embodiments, if, under the second detection index, the first safety factor of a site exceeds a second threshold, a fire alarm message is generated, and the second threshold is a temperature threshold.

[0047] In some embodiments, the instrument reading information is the third detection index;

[0048] Performing safety assessment on each point according to each detection index to obtain the first safety factor of each point under the condition of this detection index, including:

[0049] Identifying the dashboard images of the relevant equipment arranged at each point from the image library collected at the work site;

[0050] Performing image processing on the dashboard images and reading the instrument indication information displayed in the dashboard images;

[0051] Performing safety assessment on each point according to the instrument indication information of the relevant equipment arranged at each point, and determining the first safety factor of each point under the third detection index.

[0052] In some embodiments, the dashboard images are identified from the image library collected at the work site through the yolov7 algorithm.

[0053] In some embodiments, opencv is used as the image processing method to perform rotation correction on the dashboard images;

[0054] Establishing the center position of the dial, the position of the zero scale line, and the position pointed by the pointer through Hough line detection, measuring the angle between the pointer and the zero scale line, and obtaining the instrument indication information displayed on the dashboard.

[0055] In some embodiments, if the first safety factor of a certain point exceeds a third threshold under the third detection index, an instrument alarm message is generated, and the third threshold is a normal working threshold of the equipment.

[0056] On the other hand, the present invention also provides a path optimization device for an inspection robot, adopting the above path optimization method. The device at least includes:

[0057] An initial detection module, configured to detect each point according to an initial detection path and obtain an initial detection result, where the initial detection result includes at least one detection index;

[0058] A path optimization module, configured to re-plan the inspection path according to the initial detection result, optimize the initial detection path, and determine a target detection path;

[0059] A re-detection module, configured to re-detect each point according to the target detection path and obtain a target detection result.

[0060] In some embodiments, the path optimization module includes an index constraint optimization module, configured to:

[0061] Performing safety assessment on each point according to each detection index to obtain the first safety factor of each point under the condition of this detection index;

[0062] The second safety factor of each point is obtained by integrating the first safety factor of each point under all test indicators;

[0063] Sort the second safety factors of all sites in ascending order,

[0064] According to the sorting order, the points corresponding to the ones with the smaller second safety factors are taken as priority inspection targets, and the inspection paths are replanned to obtain the target detection paths.

[0065] In some embodiments, the path optimization module includes an indicator constraint optimization module and an obstacle avoidance optimization module, which is used to:

[0066] Establish a gravitational field for the target point and a repulsive field for the obstacle, and calculate the gravitational and repulsive forces at the current point of the inspection robot;

[0067] Determine whether the inspection robot is currently in a detour behavior,

[0068] If the inspection robot is in a detour behavior, it is further determined whether the inspection robot has left the local minimum area. If it has left, the inspection robot moves along a first path until it reaches the target point. The first path is a non-detour route that moves along the direction of the combined force of attraction and repulsion to reach the target point. If it has not left, the inspection robot moves along a second path and continues the detour behavior until it leaves the local minimum area. The second path is a detour route that moves along the equipotential lines of the repulsive field to leave the local minimum area.

[0069] If the robot is not in a detour, it is further determined whether a local minimum area is encountered. If so, the inspection robot moves along the second path and begins a detour until it leaves the local minimum area. If not, the inspection robot moves along the first path until it reaches the target point.

[0070] The target detection path is updated according to the first path and the second path.

[0071] In another aspect, the present invention further provides a readable storage medium having a program or instruction stored thereon, which, when executed by a processor, implements the steps of the above-mentioned path optimization method for the inspection robot.

[0072] From the above scheme, it can be seen that the advantages of the present invention are:

[0073] The path optimization method of the inspection robot provided by the present invention updates data such as toxic / combustible gas detection, thermal imaging detection, and instrument readings in a chemical industrial park, and conducts safety assessments based on each detection result as a detection index, re-plans the inspection path, and combines the improved artificial potential field method for obstacle avoidance settings to further optimize the detection path, improving the accuracy and efficiency of path optimization. This path optimization method can achieve fully intelligent and unmanned inspections, can identify equipment or areas with relatively high risk coefficients to increase the inspection frequency, and can give early warnings for emergencies, greatly improving the safety of chemical industrial parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 FIG. shows the overall flowchart of the path optimization method of the inspection robot provided by an embodiment of the present invention;

[0075] Figure 2 FIG. shows Figure 1 a schematic flowchart of an embodiment of step S2;

[0076] Figure 3 FIG. shows Figure 1 a schematic flowchart of another embodiment of step S2;

[0077] Figure 4 FIG. shows the architecture diagram of the path optimization device of the inspection robot.

[0078] Among them,

[0079] 400 - Path optimization device;

[0080] 401 - Initial detection module;

[0081] 402 - Path optimization module;

[0082] 4021 - Index constraint optimization module;

[0083] 4022 - Obstacle avoidance optimization module;

[0084] 403 - Re - detection module;

[0085] S1 - S3, S211 - S213, S221 - S227: Steps. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0086] To make the above - mentioned features and effects of the present invention more clearly and understandably explained, the following specific embodiments are given and detailed descriptions are made in conjunction with the accompanying drawings of the specification as follows.

[0087] The path optimization method of the inspection robot provided by the present invention will be described in detail below.

[0088] As shown in Figure 1 FIG.Figure 1 The overall flowchart of the path optimization method for the inspection robot is shown.

[0089] The path optimization of an inspection robot at least includes the following steps:

[0090] Step S1: Detect each point according to the initial detection path to obtain an initial detection result, and the initial detection result includes at least one detection index.

[0091] In some embodiments, detecting each point includes collecting the concentration of toxic or combustible gases corresponding to the inspection position, such as detecting the concentration of formaldehyde gas, carbon dioxide gas, ammonia gas, sulfur dioxide gas, hydrogen chloride gas, cyanide, etc., to timely discover the leakage of toxic or combustible gases in the chemical industrial park; collecting the object temperature information corresponding to the inspection position to timely discover the fire situation in the chemical industrial park; and collecting the instrument indication information of the equipment in the corresponding steps of the chemical industrial park to timely discover equipment failures, etc. In specific implementation, the concentration information of toxic or combustible gases, object temperature information, instrument indication information, etc. can be used as detection indexes for detection. The current detection indexes are not limited to these three indexes, and can be specifically set according to actual needs.

[0092] In some embodiments, the infrared gas detection method can be used to collect the concentration information of toxic or combustible gases at each point. This detection method has excellent sensitivity and accuracy, a large measurement range, and a long service life. Its detection principle is that a certain type of gas only corresponds to absorbing the infrared light energy of a certain specific wavelength band. Therefore, the gas detector only needs to detect whether there is a change in the emitted infrared light to detect whether the chemical plant produces toxic gases or flammable and explosive gases during the production process, and whether there is gas leakage. The specific detection principle of the detector can be expressed by the Lambert-Beer law:

[0093] I λ =I 0λ exo{-[α(λ)+β(λ)]L}

[0094] Where: I λ is the infrared ray energy of the wavelength band λ detected by the detector, I 0λ is the infrared ray energy of the wavelength band λ emitted by the detector, α and β are the medium absorption coefficient and scattering coefficient related to the transmission light wavelength and medium density, L is the length of the optical path, and the length of the optical path of the determined point detector is determined.

[0095] In specific implementation, the infrared gas detection method can be used for fixed scenarios and for detecting the concentration of fixed gases. Only relevant filters need to be placed on the infrared radiator so that the detector can only emit rays of a certain specific wavelength band, and detect the gas that absorbs a certain specific wavelength band, that is, detect the concentration of a single gas.

[0096] In some embodiments, the temperature information of the object at each point can be collected by a temperature sensor thermal imaging camera arranged on site. The pan-tilt adjusts its own angle orientation so that the temperature sensor thermal imaging camera can obtain the temperature information of the object from different angles, and the thermal imaging camera judges the temperature by collecting the infrared light emitted from the surface of the object.

[0097] In some embodiments, from the image library collected at the work site, the dashboard image of the relevant equipment arranged at each point can be identified, and then the instrument reading information displayed in the dashboard image can be read.

[0098] Step S2: Re-plan the inspection path based on the initial detection result, optimize the initial detection path, and determine the target detection path.

[0099] In this embodiment, after obtaining the initial detection result through the preliminary detection in step S1, the initial detection result is further used to re-plan the inspection path to optimize the initial detection path and determine the target detection path.

[0100] In this embodiment, the path optimization process includes two parts. One part is to use the previously determined detection indexes as constraint conditions to optimize the detection path by evaluating safety; the other part is to use the improved artificial potential field method to set up obstacle avoidance for the local minimum area that appears during the movement of the robot, and further optimize the detection path.

[0101] In some embodiments, for the first part, using the detection indexes as constraint conditions to optimize the detection path by evaluating safety, specifically as Figure 2 shown in Figure 2 shows the detailed flow diagram of this process, specifically including the following steps:

[0102] Step S211: Perform a safety assessment on each point according to each detection index to obtain the first safety factor of each point under the condition of this detection index. Specifically:

[0103] In some embodiments, the concentration information of toxic or combustible gases is used as the first detection index. First, according to the position change of the patrol robot, the concentration information of toxic or combustible gases at each point is collected; then, based on the concentration information of toxic or combustible gases at each point, a safety assessment is carried out for each point to determine the first safety factor of each point under the first detection index. The smaller the safety factor, the higher the risk of exceeding the standard of toxic or combustible gases at that point. At this time, if under the first detection index, the first safety factor of a certain point exceeds a first threshold (the first threshold is a gas concentration threshold), it indicates that there is a region with a relatively high risk level. At this time, a gas concentration alarm message is generated, and this gas concentration alarm message is sent to relevant staff through remote transmission for timely detection and an immediate alarm to attract attention, so as to avoid more dangers.

[0104] In some embodiments, the object temperature information is used as the second detection index. First, according to the position change of the patrol robot, the object temperature information at each point is collected; then, based on the object temperature information at each point, a safety assessment is carried out for each point to determine the first safety factor of each point under the second detection index. The smaller the safety factor, the higher the object temperature at that point and the higher the risk of fire. At this time, if under the second detection index, special records are made for locations that have not reached the ignition temperature but are higher than the normal object temperature and key safety assessments are carried out. If the first safety factor of a certain point exceeds a second threshold (the second threshold is a temperature threshold), it indicates that there is a dangerous area where a fire may occur. Then, a fire alarm message is generated, and this fire alarm message is sent to relevant staff through remote transmission for timely detection and an immediate alarm to attract attention, so as to avoid more dangers.

[0105] In some embodiments, the instrument reading information is used as the third detection index. Since the chemical environment is highly dangerous and some specific environments and equipment are not suitable for personnel to enter and come into contact with, it is not practical to collect the reading information of each instrument panel on-site. Therefore, in this embodiment, in combination with the existing target recognition method, from the image library collected at the work site, the instrument panel images of the relevant equipment arranged at each point are identified; then, through image processing of the instrument panel images, the instrument reading information displayed in the instrument panel images is read; finally, based on the instrument reading information of the relevant equipment arranged at each point, the safety of each point is evaluated, and the first safety factor of each point under the third detection index is determined. The smaller the safety factor, the more the instrument reading information at this point exceeds the normal range, and the higher the possibility of equipment failure or overload. At this time, if under the third detection index, the first safety factor of a certain point exceeds a third threshold (the third threshold is the normal working threshold of a device), an instrument alarm message is generated, and this instrument alarm message is sent to the relevant staff through remote transmission for timely equipment detection and an immediate alarm to attract attention, avoiding more dangers.

[0106] In some embodiments, specifically, the yolov7 algorithm can be used to crop the photos containing only the instrument panel from the image library collected at the work site to identify the instrument panel images. In addition, in some embodiments, opencv is specifically used as the image processing method to perform rotational correction on the instrument panel images; and the center position of the dial, the position of the zero scale line, and the position pointed by the pointer are established through Hough line detection, and the angle between the pointer and the zero scale line is measured, so as to obtain the instrument reading information displayed on the instrument panel.

[0107] Step S212: Combine the first safety factors of each point under all detection indexes to obtain the second safety factor of each point.

[0108] Step S213: Sort the second safety factors of all points in ascending order, and based on this sorting order, take the point corresponding to the smaller second safety factor as the priority inspection target, and re-plan the inspection path to obtain the target detection path.

[0109] In this embodiment, by respectively determining the safety factors of each point under the constraints of each detection index, and then taking all detection indexes as the constraint conditions, the overall safety of each point is evaluated, and the overall safety of each point is re-planned for the inspection path to determine the target detection path.

[0110] Furthermore, since the traditional artificial potential field method is not fully applicable to chemical scenarios, local minima may occur during the movement of the robot. To address this problem, in this embodiment, an improved artificial potential field method is used to construct a more suitable potential function to reduce or avoid the occurrence of local minima. In this part, the improved artificial potential field method is used to set up obstacle avoidance for the local minimum area that appears during the movement of the robot, and further optimize the detection path. Specifically, as shown in Figure 3 shown below, Figure 3 Figure shows a detailed flowchart of this process, which specifically includes the following steps:

[0111] First, establish a gravitational field for the target point and a repulsive field for the obstacle, and calculate the gravitational and repulsive forces at the current position of the inspection robot (S221). Specifically:

[0112] By establishing a gravitational field for the target point, based on the distance between the current position of the inspection robot and the target point, determine the gravitational force at the current position of the inspection robot. This gravitational force is expressed as:

[0113]

[0114] where, U att (x) represents the gravitational force of the inspection robot at the current position x, δ represents the relative influence of the gravitational potential; x represents the current position of the inspection robot, G represents the target point, and d(x, G) represents the distance between the current position and the target point. The function of is to weaken the effect of the target gravitational potential when the inspection robot is far from the target point.

[0115] By establishing a repulsive field for the obstacle, based on the distance between the current position of the inspection robot and the obstacle, as well as the repulsive potential of the obstacle, determine the repulsive force at the current position of the inspection robot. This repulsive force is expressed as

[0116] s

[0117] where, U rep,j (x) represents the repulsive force of the inspection robot at the current position x, μ j represents the relative influence of the repulsive potential of the j-th obstacle, d j (x) represents the distance between the current position x of the robot and the j-th obstacle, represents the action range of the repulsive potential of the j-th obstacle.

[0118] Then, determine whether the inspection robot is currently in a bypass behavior (S222):

[0119] Among them, if in a detouring behavior, it is further determined whether the inspection robot has left the local minimum area (S223). If it has left, the inspection robot moves along the first path until it reaches the target point. The first path is a non-detouring route that moves along the direction of the resultant force of gravity and repulsion to reach the target point (S224); if it has not left, the inspection robot moves along the second path and continues the detouring behavior until it leaves the local minimum area. The second path is a detouring route that moves along the equipotential line of the repulsive force field to leave the local minimum area (S225).

[0120] If not in a detouring behavior, it is further determined whether a local minimum area is encountered (S226). If encountered, the inspection robot moves along the second path and starts the detouring behavior until it leaves the local minimum area (S225); if not encountered, the inspection robot moves along the first path until it reaches the target point (S224).

[0121] Finally, update the target detection path according to the first path and the second path (S227).

[0122] In some embodiments, a first discrimination condition is constructed using the resultant force of gravity and repulsion at the current position of the inspection robot, and the first discrimination condition is used to determine whether the robot encounters a local minimum area. The first discrimination condition is set as:

[0123]

[0124] Among them, ε is a very small positive number, which means that the virtual resultant force received by the inspection robot is close to 0. Through the discrimination condition method, the local minimum area can be judged more directly. If the resultant force of gravity and repulsion at the current position of the inspection robot is less than ε, it means that a local minimum area is encountered.

[0125] In some embodiments, a second discrimination condition is constructed using the moving distance of the inspection robot from the previous position x A to the current position x, and the second discrimination condition is used to determine whether the robot encounters a local minimum area. The second discrimination condition is set as:

[0126] |x - x A |< αs A

[0127] Among them, α is a positive number between 0 and 1, x A is a certain state during the movement of the inspection robot, and s A represents the total distance of the inspection robot from the previous position x A to the current position x. The establishment of this discrimination means that after the inspection robot moves a long distance, the displacement is very small, which is used to detect whether the inspection robot vibrates and makes circular motion near the local minimum point.

[0128] In this embodiment, after determining the target detection path through safety assessment under the constraint of each detection index, the improved artificial potential field method is further used to set up obstacle avoidance for the local minimum area that appears during the movement of the robot, further optimizing the detection path.

[0129] Step S3: Re-detect each point according to the target detection path to obtain the target detection result.

[0130] In this embodiment, after further optimizing the detection path through step S2, each point is re-detected according to the target detection path, and the detection result is updated in real time.

[0131] In summary, for the path optimization method of the inspection robot provided in this embodiment, it updates the detection of data such as toxic / combustible gas detection, thermal imaging detection, and instrument readings in the chemical industrial park, and performs safety assessment based on each detection result as a detection index, re-plans the inspection path, and combines the improved artificial potential field method to set up obstacle avoidance, further optimizing the detection path, improving the accuracy and efficiency of path optimization. This path optimization method can achieve fully intelligent and unmanned inspection, can determine equipment or areas with a higher risk coefficient to increase the inspection frequency, and can give early warning for emergencies, greatly improving the safety of the chemical industrial park.

[0132] In addition, the above embodiments of the present invention can be applied to terminal devices with the path optimization function of inspection robots. The terminal devices may include personal terminals, host computer terminals, etc., and the embodiments of the present invention do not limit this. The terminal can support operating systems such as Windows, Android, IOS, Windows Phone, etc.

[0133] Refer to Figure 4 , Figure 4 shows a path optimization device 400 for an inspection robot. Based on the path optimization method of the inspection robot, it can be applied to personal terminals and host computer terminal devices, and it can implement the same as Figures 1-3 The path optimization method of the inspection robot shown can enable the path optimization device of the inspection robot provided in the embodiments of the present invention to implement each process implemented by the above path optimization method of the inspection robot.

[0134] A path optimization device 400 for an inspection robot, adopting the above path optimization method of the inspection robot, at least includes:

[0135] An initial detection module 401, configured to detect each point according to the initial detection path to obtain an initial detection result, where the initial detection result includes at least one detection index;

[0136] A path optimization module 402 is used to replan the inspection path based on the initial detection results, optimize the initial detection path, and determine the target detection path;

[0137] The re-detection module 403 is used to re-detect each point according to the target detection path to obtain a target detection result.

[0138] In some embodiments, the path optimization module 402 includes an indicator constraint optimization module 4021 for:

[0139] Conduct safety assessment on each point according to each test indicator, and obtain the first safety factor of each point under the test indicator condition;

[0140] The second safety factor of each point is obtained by integrating the first safety factor of each point under all test indicators;

[0141] Sort the second safety factors of all sites in ascending order,

[0142] According to the sorting order, the points corresponding to the ones with the smaller second safety factors are taken as priority inspection targets, and the inspection paths are replanned to obtain the target detection paths.

[0143] In some embodiments, the path optimization module 402 includes an indicator constraint optimization module and an obstacle avoidance optimization module 4022 for:

[0144] Establish a gravitational field for the target point and a repulsive field for the obstacle, and calculate the gravitational and repulsive forces at the current point of the inspection robot;

[0145] Determine whether the inspection robot is currently in a detour behavior,

[0146] If it is in a detour behavior, it is further determined whether the inspection robot has left the local minimum area. If it has left, the inspection robot moves along the first path until it reaches the target point. The first path is a non-detour route that moves along the direction of the combined force of attraction and repulsion to reach the target point. If it has not left, the inspection robot moves along the second path and continues the detour behavior until it leaves the local minimum area. The second path is a detour route that moves along the equipotential lines of the repulsive field to leave the local minimum area.

[0147] If the robot is not in a detour, it is further determined whether it encounters a local minimum area. If so, the inspection robot moves along the second path and begins a detour until it leaves the local minimum area. If not, the inspection robot moves along the first path until it reaches the target point.

[0148] The target detection path is updated according to the first path and the second path.

[0149] In addition, it should be understood that in the path optimization device 400 of the inspection robot according to the embodiments of the present invention, only the above division of each functional module is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs. That is, the path optimization device 400 of the inspection robot can be divided into functional modules different from the above-exemplified modules to complete all or part of the functions described above.

[0150] In addition, an embodiment of the present invention further provides an electronic device, including a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, it implements the steps of the above path optimization method of the inspection robot and can achieve the same technical effects.

[0151] Among them, the processor is the processor in the electronic device in the above embodiment. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0152] In addition, an embodiment of the present invention further provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by the processor, it implements the steps of the above path optimization method of the inspection robot and can achieve the same technical effects.

[0153] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present invention is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in a reverse order according to the functions involved. For example, the methods described can be performed in an order different from that described, and various steps can also be added, omitted, or combined. In addition, the features described in reference to certain examples can be combined in other examples.

[0154] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can still make many forms without departing from the spirit of the present invention and the scope protected by the claims, and all of them fall within the protection scope of the present invention.

Claims

1. A path optimization method for an inspection robot, characterized in that: Used for hazardous chemical inspections in chemical parks, this method includes: Based on the initial detection path and according to the position change of the inspection robot, each point is detected in real time to obtain an initial detection result, wherein the initial detection result includes at least one detection indicator, and the detection indicator includes one of toxic or flammable gas concentration information, object temperature information, and instrument reading information; Replanning the inspection path based on the initial detection result, optimizing the initial detection path, and determining the target detection path includes: Conduct safety assessment on each point according to each test indicator, and obtain the first safety factor of each point under each test indicator condition; The second safety factor of each point is obtained by integrating the first safety factor of each point under all test indicators; Sort the second safety factors of all sites in ascending order, According to the sorting order, the points corresponding to the points with the smaller second safety factors are selected as priority inspection targets, and the inspection paths are replanned to obtain the target detection paths; Optimizing the target detection path in combination with obstacles includes: Establish a gravitational field for the target point and a repulsive field for the obstacle, and calculate the gravitational and repulsive forces at the current point of the inspection robot; Determine whether the inspection robot is currently in a detour behavior, If the inspection robot is in a detour behavior, it is further determined whether the inspection robot has left the local minimum area. If it has left, the inspection robot moves along a first path until it reaches the target point. The first path is a non-detour route that moves along the direction of the combined force of attraction and repulsion to reach the target point. If it has not left, the inspection robot moves along a second path and continues the detour behavior until it leaves the local minimum area. The second path is a detour route that moves along the equipotential lines of the repulsive field to leave the local minimum area. If the robot is not in a detour, it is further determined whether a local minimum area is encountered. If so, the inspection robot moves along the second path and begins a detour until it leaves the local minimum area. If not, the inspection robot moves along the first path until it reaches the target point. The target detection path is updated based on the first and second paths. Each point is re-detected according to the optimized target detection path to obtain the target detection result.

2. The method according to claim 1, characterized in that According to the distance between the current position of the inspection robot and the target position, the gravitational force of the current position of the inspection robot is determined. The gravitational force is expressed as: Among them, U att (x) represents the gravitational force of the inspection robot at the current point x, δ represents the relative influence of the gravitational potential; x represents the current point of the inspection robot, G represents the target point, d(x, G) represents the distance between the current point and the target point, Its function is to weaken the gravitational potential of the target when the inspection robot is far away from the target point.

3. The method according to claim 2, characterized in that According to the distance between the inspection robot's current position and the obstacle and the repulsive force potential of the obstacle, the repulsive force at the inspection robot's current position is determined. The repulsive force is expressed as: Among them, U rep,j (x) represents the repulsive force of the inspection robot at the current point x, μ j represents the relative influence of the repulsive potential of the jth obstacle, d j (x) represents the distance between the robot’s current position x and the j-th obstacle, represents the range of the repulsive potential of the j-th obstacle.

4. The method according to claim 3, characterized in that A first discrimination condition is constructed using the combined force of the attraction and repulsion at the current position of the inspection robot, and the first discrimination condition is used to determine whether the robot encounters a local minimum area; The first discrimination condition is set as: Among them, U att represents gravity, U rep,j Represents repulsive force; ε is a very small positive number, which means that the virtual net force acting on the inspection robot is close to 0. If the net force of attraction and repulsion at the current position of the inspection robot is less than ε, it means that a local minimum area is encountered.

5. The method according to claim 1, wherein The second discrimination condition is constructed using the movement distance of the inspection robot from the previous point to the current point, and the second discrimination condition is used to determine whether the robot encounters a local minimum area; The second discrimination condition is set as: |x-x A |<αs A Among them, α is a positive number between 0 and 1, x A is a previous point during the movement of the inspection robot, s A Indicates that the inspection robot moves from the previous point x A The total distance to reach the current point x.

6. The method according to claim 1, wherein The toxic or combustible gas concentration information is the first detection indicator; The safety evaluation of each point according to each detection indicator is performed to obtain a first safety factor of each point under the detection indicator condition, including: According to the position changes of the inspection robot, the concentration information of toxic or combustible gas at each point is collected; A safety assessment is performed on each point based on the toxic or combustible gas concentration information at each point, and a first safety factor of each point under the first detection index is determined.

7. The method according to claim 6, characterized in that The infrared gas detection method is used to collect toxic or combustible gas concentration information at each point.

8. The method according to claim 6, characterized in that If, under the first detection indicator, there is a point where the first safety factor exceeds a first threshold, a gas concentration alarm message is generated, where the first threshold is a gas concentration threshold.

9. The method according to claim 1, characterized in that The object temperature information is the second detection indicator; The safety evaluation of each point according to each detection indicator is performed to obtain a first safety factor of each point under the detection indicator condition, including: According to the position change of the inspection robot, the temperature information of the object at each point is collected; A safety assessment is performed on each point based on the temperature information of the object at each point, and a first safety factor of each point under the second detection index is determined.

10. The method according to claim 9, characterized in that The temperature information of objects at each point is collected through the temperature sensor thermal imaging camera arranged on site.

11. The method according to claim 9, characterized in that If, under the second detection indicator, the first safety factor of a point exceeds a second threshold, a fire alarm message is generated, where the second threshold is a temperature threshold.

12. The method according to claim 1, characterized in that The instrument numerical information is the third detection indicator; The safety evaluation of each point according to each detection indicator is performed to obtain a first safety factor of each point under the detection indicator condition, including: Identify the instrument panel images of the relevant equipment at each location from the image library collected at the work site; performing image processing on the instrument panel image and reading instrument numerical information displayed in the instrument panel image; A safety assessment is performed on each location based on the instrument reading information of the relevant equipment arranged at each location, and a first safety factor of each location under the third detection index is determined.

13. The method according to claim 12, characterized in that The dashboard image is identified from the image library collected from the work site using the yolov7 algorithm.

14. The method according to claim 12 or 13, characterized in that Using OpenCV as an image processing method, the instrument panel image is subjected to rotation correction; The center position of the dial, the position of the zero scale line, and the position of the pointer are determined through Hough line detection, and the angle between the pointer and the zero scale line is measured to obtain the instrument reading information displayed on the instrument panel.

15. The method according to claim 12, characterized in that If, under the third detection index, the first safety factor of a point exceeds a third threshold, an instrument alarm message is generated, and the third threshold is a normal operation threshold of the equipment.

16. A path optimization device for an inspection robot, characterized in that: The path optimization method according to any one of claims 1 to 15 is adopted, wherein the device at least comprises: An initial detection module is configured to detect each point according to the initial detection path and the position change of the inspection robot to obtain an initial detection result, wherein the initial detection result includes at least one detection indicator, and the detection indicator includes one of toxic or flammable gas concentration information, object temperature information, and instrument reading information; A path optimization module is used to replan the inspection path based on the initial detection results, optimize the initial detection path, and determine the target detection path; wherein the path optimization module includes an indicator constraint optimization module for: Conduct safety assessment on each point according to each test indicator, and obtain the first safety factor of each point under the test indicator condition; The second safety factor of each point is obtained by integrating the first safety factor of each point under all test indicators; Sort the second safety factors of all sites in ascending order, According to the sorting order, the points corresponding to the ones with the smaller second safety factors are selected as priority inspection targets, and the inspection paths are replanned to obtain the target detection paths; and Optimizing the target detection path in combination with obstacles includes: Establish a gravitational field for the target point and a repulsive field for the obstacle, and calculate the gravitational and repulsive forces at the current point of the inspection robot; Determine whether the inspection robot is currently in a detour behavior, If the inspection robot is in a detour behavior, it is further determined whether the inspection robot has left the local minimum area. If it has left, the inspection robot moves along a first path until it reaches the target point. The first path is a non-detour route that moves along the direction of the combined force of attraction and repulsion to reach the target point. If it has not left, the inspection robot moves along a second path and continues the detour behavior until it leaves the local minimum area. The second path is a detour route that moves along the equipotential lines of the repulsive field to leave the local minimum area. If the robot is not in a detour, it is further determined whether a local minimum area is encountered. If so, the inspection robot moves along the second path and begins a detour until it leaves the local minimum area. If not, the inspection robot moves along the first path until it reaches the target point. The target detection path is updated based on the first and second paths. The re-detection module is used to re-detect each point according to the optimized target detection path to obtain the target detection result.

17. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by a processor, the steps of the path optimization method of the inspection robot according to any one of claims 1 to 15 are implemented.

Images