UAV-Based Inspection Path Planning Method, System, and Storage Medium

Abstract

The present invention discloses an inspection path planning method, system and storage medium based on an unmanned aerial vehicle. The method includes: S1, collecting power distribution equipment images of power distribution equipment, performing a first type classification and a second type classification on the power distribution equipment images to determine the secondary type corresponding to the power distribution equipment images; S2, dividing different power distribution equipment into a first group and a second group to generate a basic route; S3, generating the next route, determining whether the eigenvalue of the next route is less than the eigenvalue of the basic route. If so, regarding the next route as the basic route, checking whether the eigenvalue of the basic route is less than the eigenvalue threshold. If not, directly checking whether the eigenvalue of the basic route is less than the eigenvalue threshold. If it is less, regarding the basic route as the final route. If it is greater than or equal to, repeating this step. Through the present invention, a suitable inspection route can be generated.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data analysis, and particularly relates to an inspection path planning method, system and storage medium based on an unmanned aerial vehicle (UAV). Background Art

[0002] Distribution lines are power lines used to deliver electric energy to users. Conducting daily inspections on distribution equipment on the distribution lines can ensure the stability and reliability of power supply, and planning a suitable inspection route is an important prerequisite for carrying out inspection activities.

[0003] A similar prior art is a Chinese patent application with the publication number CN113408774A, which discloses a route planning method, device, storage medium and electronic device. The method includes: dividing target points into at least two sub-regions according to the distances between the target points and the regional division rules; iteratively training the simulated paths between at least two sub-regions according to at least one path evaluation parameter to determine the first path between at least two sub-regions that meets at least one path evaluation parameter; generating the target path between each target point according to the second path of each target point in any sub-region and the first path between at least two sub-regions. However, this patent application does not consider the time issue when generating the route. In addition, a similar prior art is a Chinese patent application with the publication number CN113592122A, which discloses a method and device for route planning, related to the field of big data. The specific implementation solution is: initializing the central point position coordinates of each category in the target locations to be classified; clustering the target locations to be classified based on the central point position coordinates of each category and the category balance constraint condition to determine the set of target locations of each category corresponding to the central point position coordinates of each category, where the category balance constraint condition is that the number of target locations in the set of target locations of each category differs by no more than 1; solving the traveling salesman problem for the set of target locations of each category to determine the walking route of each category corresponding to the set of target locations of each category. However, this patent application does not consider the locations that must be passed through when generating the route. Summary of the Invention

[0004] The present invention collects images of distribution equipment, determines the first-level category and second-level category corresponding to the distribution equipment images, thereby dividing different distribution equipment into a first group and a second group, and then generates and updates the basic route, and searches for the basic route with a feature value less than the feature value threshold as the inspection route. The present invention aims to generate a suitable inspection route.

[0005] To achieve the above invention purpose, the present invention provides the following inspection path planning method based on an unmanned aerial vehicle, which mainly includes the following steps:

[0006] S1. For each power distribution device located on the power distribution line, the UAV module acquires the power distribution device image of the power distribution device, transmits the power distribution device image to the preprocessing module, and the preprocessing module performs the first type classification on the power distribution device image to determine the first-level type corresponding to the power distribution device image. On the basis of determining the first-level type corresponding to the power distribution device image, the preprocessing module continues to perform the second type classification on the power distribution device image to determine the second-level type corresponding to the power distribution device image;

[0007] S2. The preprocessing module sends different power distribution device images and the second-level types respectively corresponding to different power distribution device images to the route planning module, and the route planning module divides the power distribution devices corresponding to the power distribution device images corresponding to the second-level types that meet the preset requirements into the first group, and divides all other power distribution devices into the second group. The route planning module also generates a basic route, calculates the eigenvalue of the basic route, and the basic route includes all the power distribution devices in the first group;

[0008] S3. The route planning module generates the next route based on the basic route, calculates the eigenvalue of the next route, determines whether the eigenvalue of the next route is less than the eigenvalue of the basic route. If so, the next route is regarded as the basic route, and it is checked whether the eigenvalue of the basic route is less than the preset eigenvalue threshold. If not, it is directly checked whether the eigenvalue of the basic route is less than the preset eigenvalue threshold. If it is less, the basic route is used as the final route. If it is greater than or equal to, this step is repeated.

[0009] As a preferred technical solution of the present invention, the preprocessing module performs the first type classification on the power distribution device image, including the following steps:

[0010] S111. The preprocessing module determines whether the number of type items currently included in the first-level type is greater than zero. If not, it jumps to S113. If so, the preprocessing module calculates the similarity degree value between the power distribution device image and the average power distribution device image of each type item respectively;

[0011] S112. The preprocessing module determines whether there is a similarity degree value greater than the preset first similarity degree value threshold. If not, it jumps to S113. If so, the preprocessing module divides the power distribution device image into the type item corresponding to the maximum similarity degree value, and updates the average power distribution device image of the type item;

[0012] S113. The preprocessing module divides the power distribution device image into a new type item, and regards the power distribution device image as the average power distribution device image of the new type item.

[0013] As a preferred technical solution of the present invention, the preprocessing module pre-stores a division model for further classifying the power distribution equipment image into a second category on the basis of determining the first category corresponding to the power distribution equipment image. The division model includes different layer models from top to bottom. Each layer model corresponds to a category item included in the second category. Each layer model sequentially includes, from left to right, the identification power distribution equipment image of the corresponding category item, and a number of historical power distribution equipment images. In the order from top to bottom, the similarity value between the identification power distribution equipment image of the corresponding category item in different layer models and the identification power distribution equipment image of the first category decreases sequentially. In each layer model, in the order from left to right, the similarity value between a number of historical power distribution equipment images and the identification power distribution equipment image decreases sequentially.

[0014] As a preferred technical solution of the present invention, the preprocessing module further classifies the power distribution equipment image into a second category on the basis of determining the first category corresponding to the power distribution equipment image, including the following steps:

[0015] S121. In the division model, the preprocessing module determines whether it has reached the end of the model. If so, it jumps to S124. If not, the preprocessing module calculates the similarity value between the power distribution equipment image and the identification power distribution equipment image in the current layer model, and continues with S122;

[0016] S122. The preprocessing module checks whether the similarity value is greater than a preset second similarity value threshold. If not, it moves to the next layer model and jumps to S121. If so, it continues with S123;

[0017] S123. The preprocessing module determines whether it has reached the end of the model. If so, it jumps to S124. If not, it moves to the right in the current layer model. The preprocessing module checks whether the similarity value is greater than the similarity value between the current historical power distribution equipment image and the identification power distribution equipment image in the current layer model. If not, it repeats this step. If so, it adds the power distribution equipment image before the current historical power distribution equipment image, and continues with S125;

[0018] S124. After the preprocessing module adds the power distribution equipment image to the end of the model, it continues with S125;

[0019] S125. The preprocessing module updates the identification power distribution equipment image in the layer model to which the power distribution equipment image is added, and updates the order of different layer models in the division model in the up-down direction.

[0020] As a preferred technical solution of the present invention, the method for the route planning module to generate the next route based on the basic route includes: Method 1: Select any two power distribution devices from all the power distribution devices on the basic route, and exchange the positions of the two power distribution devices on the basic route; Method 2: Select any one power distribution device from all the power distribution devices that belong to the second group but are not on the basic route, and add the power distribution device to the basic route; Method 3: Select any one power distribution device from all the power distribution devices that belong to the second group on the basic route, and delete the power distribution device from the basic route.

[0021] As a preferred technical solution of the present invention, the route planning module calculates the eigenvalue of the route through the following formula:

[0022] θ = χ1 + χ2 + χ3

[0023] Where θ is the eigenvalue, χ1 is the value representing the difference degree between the actual inspection time and the preset standard inspection time. The actual inspection time is the time for inspecting according to the route, χ2 is the value representing the total inspection distance of the route, and χ3 is the value representing the degree to which the power distribution devices with high importance on the route are inspected in advance.

[0024] As a preferred technical solution of the present invention, χ1 includes two cases: when the actual inspection time is greater than the standard inspection time, χ1 = γ*(α - β), where γ is the total number of all power distribution devices that belong to the second group on the route, α is the actual inspection time, and β is the standard inspection time; when the actual inspection time is less than or equal to the standard inspection time, χ1 = ω*(β - α), where ω is the total number of all power distribution devices that belong to the second group but are not on the route, β is the standard inspection time, and α is the actual inspection time.

[0025] As a preferred technical solution of the present invention, Where m is the total number of power distribution devices on the route, n is the appearance order of the power distribution device on the route, E is a function for finding the larger value of two numerical values, κ n is the importance value of the nth power distribution device on the route, and the value of σ is 0.

[0026] The present invention also provides an inspection path planning system based on an unmanned aerial vehicle, including the following modules:

[0027] The unmanned aerial vehicle module is used to collect images of power distribution devices for each power distribution device located on the power distribution line and transmit the power distribution device images to the preprocessing module;

[0028] A preprocessing module is used to perform a first type classification on the power distribution equipment image to determine the first-level type corresponding to the power distribution equipment image. On the basis of determining the first-level type corresponding to the power distribution equipment image, it continues to perform a second type classification on the power distribution equipment image to determine the second-level type corresponding to the power distribution equipment image, and is used to send different power distribution equipment images and the second-level types respectively corresponding to different power distribution equipment images to the route planning module;

[0029] A route planning module is used to divide the power distribution equipment corresponding to the power distribution equipment image corresponding to the second-level type that meets the preset requirements into the first group, and divide all other power distribution equipment into the second group. It also generates a basic route, calculates the characteristic value of the basic route, and is used to generate the next route based on the basic route, calculate the characteristic value of the next route, and determine whether the characteristic value of the next route is less than the characteristic value of the basic route. If so, the next route is regarded as the basic route, and it checks whether the characteristic value of the basic route is less than the preset characteristic value threshold. If not, it directly checks whether the characteristic value of the basic route is less than the preset characteristic value threshold. If it is less, the basic route is used as the final route. If it is greater than or equal to, it continues to generate the next route.

[0030] The present invention also provides a storage medium storing program instructions, wherein when the program instructions run, they control the device where the storage medium is located to execute the method described in any one of the above.

[0031] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0032] In the present invention, first, for each power distribution device located on a power distribution line, a power distribution device image of the power distribution device is collected, and a first type classification is performed on the power distribution device image to determine the first-level type corresponding to the power distribution device image. Further, on the basis of determining the first-level type corresponding to the power distribution device image, a second type classification is continued for the power distribution device image to determine the second-level type corresponding to the power distribution device image. Secondly, the power distribution devices corresponding to the power distribution device images corresponding to the second-level types that meet the preset requirements are classified into the first group, and all other power distribution devices are classified into the second group. A basic route is also generated, and the characteristic value of the basic route is calculated. The basic route includes all the power distribution devices in the first group. Finally, a next route is generated based on the basic route, the characteristic value of the next route is calculated, and it is determined whether the characteristic value of the next route is less than the characteristic value of the basic route. In the case of yes, the next route is regarded as the basic route, and it is checked whether the characteristic value of the basic route is less than the preset characteristic value threshold. In the case of no, it is directly checked whether the characteristic value of the basic route is less than the preset characteristic value threshold. In the case of less, the basic route is used as the final route. In the case of greater than or equal to, this step is repeated. Through the present invention, a suitable inspection route can be generated as much as possible, where the actual inspection time is not much different from the standard inspection time, the total inspection distance is small, and the power distribution devices with a high degree of importance for early inspection are inspected. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a flowchart of the inspection path planning method based on an unmanned aerial vehicle according to the present invention;

[0034] Figure 2 is a composition structure diagram of the inspection path planning system based on an unmanned aerial vehicle according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and 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.

[0036] It can be understood that the terms "first", "second", etc. used in the present application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish the first element from another element. For example, without departing from the scope of the present application, the first xx script may be referred to as the second xx script, and similarly, the second xx script may be referred to as the first xx script.

[0037] The present invention provides an Figure 1 inspection path planning method based on an unmanned aerial vehicle as shown, which is mainly implemented by performing the following steps:

[0038] S1. For each power distribution device located on the power distribution line, the UAV module collects the power distribution device image of the power distribution device, transmits the power distribution device image to the preprocessing module, and the preprocessing module conducts the first type classification for the power distribution device image to determine the first-level type corresponding to the power distribution device image. On the basis of determining the first-level type corresponding to the power distribution device image, the preprocessing module continues to conduct the second type classification for the power distribution device image to determine the second-level type corresponding to the power distribution device image;

[0039] S2. The preprocessing module sends different power distribution device images and the second-level types respectively corresponding to the different power distribution device images to the route planning module, and the route planning module divides the power distribution devices corresponding to the power distribution device images corresponding to the second-level types meeting the preset requirements into the first group, and divides all other power distribution devices into the second group. The route planning module also generates a basic route and calculates the characteristic value of the basic route. The basic route includes all the power distribution devices in the first group;

[0040] S3. The route planning module generates the next route based on the basic route, calculates the characteristic value of the next route, determines whether the characteristic value of the next route is less than the characteristic value of the basic route. If so, regards the next route as the basic route, checks whether the characteristic value of the basic route is less than the preset characteristic value threshold. If not, directly checks whether the characteristic value of the basic route is less than the preset characteristic value threshold. If it is less, regards the basic route as the final route. If it is greater than or equal to, repeats this step.

[0041] Specifically, the main steps for generating the inspection route are introduced here. In step S1, for each power distribution device located on the power distribution line, the following processing is performed: the drone module collects the power distribution device image of the power distribution device and transmits the power distribution device image to the preprocessing module. The preprocessing module performs the first type classification on the power distribution device image. The detailed process will be described below. The purpose is to determine the first-level type corresponding to the power distribution device image. The first-level type includes different type items, which can specifically be different types of power distribution devices. On the basis of determining the first-level type corresponding to the power distribution device image, the preprocessing module continues to perform the second type classification on the power distribution device image. The purpose is to determine the second-level type corresponding to the power distribution device image. That is, after determining the first-level type of a power distribution device image, continue to determine the second-level type of this power distribution device image under the first-level type. The second-level type also includes different type items, which can specifically be different problem categories corresponding to a certain type of power distribution device. Examples include damage on the device surface and foreign objects on the device surface. In step S2, the preprocessing module sends different power distribution device images and the second-level types corresponding to different power distribution device images to the route planning module. Then, the route planning module divides the power distribution devices corresponding to the power distribution device images corresponding to the second-level types that meet the preset requirements into the first group. The second-level types that meet the preset requirements refer to the content of the second-level types being the same as the preset type items. For example, the preset type item is damage on the device surface. The preset type item of damage on the device surface instead of foreign objects on the surface is because damage on the device surface is more serious than foreign objects on the surface. Subsequently, the route planning module divides all other power distribution devices into the second group. The route planning module also generates a basic route, calculates the characteristic value of the basic route, and the calculation method will be described below. The basic route includes all the power distribution devices in the first group because all the power distribution devices in the first group need to be inspected more. The basic route includes any number of power distribution devices belonging to the second group. The initial basic route can be generated by the route generation method in the prior art, which will not be elaborated here. In step S3, the route planning module generates the next route based on the basic route, and the generation method will be described below. Calculate the characteristic value of the next route, and the calculation method will also be described below. Determine whether the characteristic value of the next route is less than the characteristic value of the basic route. If so, regard the next route as the basic route. Check whether the characteristic value of the basic route is less than the preset characteristic value threshold. If not, directly check whether the characteristic value of the basic route is less than the preset characteristic value threshold. In the case of being less, regard the basic route as the final route. In the case of being greater than or equal to, repeat step S3 until the final route is generated. The final route is the inspection route. Through the above method, different power distribution devices can be grouped according to the power distribution device images of different power distribution devices, and a suitable inspection route can also be generated according to the grouping result. The inspection route includes all the power distribution devices that must be inspected.

[0042] Further, the preprocessing module performs the first type classification on the power distribution equipment image, including the following steps:

[0043] S111. The preprocessing module determines whether the number of type items currently included in the first-level type is greater than zero. If not, it jumps to S113. If so, the preprocessing module calculates the similarity degree value between the power distribution equipment image and the average power distribution equipment image of each type item respectively.

[0044] S112. The preprocessing module determines whether there is a similarity degree value greater than the preset first similarity degree value threshold. If not, it jumps to S113. If so, the preprocessing module classifies the power distribution equipment image into the type item corresponding to the maximum similarity degree value and updates the average power distribution equipment image of the type item.

[0045] S113. The preprocessing module classifies the power distribution equipment image into a new type item and regards the power distribution equipment image as the average power distribution equipment image of the new type item.

[0046] Specifically, here is how the preprocessing module performs the first type classification on the power distribution equipment image. Before executing steps S111 to S113, multiple historical power distribution equipment images have been accumulated under different type items. In step S111, the preprocessing module determines whether the number of type items currently included in the first-level type is greater than zero. If so, it calculates the similarity degree value between the power distribution equipment image and the average power distribution equipment image of each type item respectively. The greater the similarity degree value, the greater the similarity degree. The average power distribution equipment image of each type item is jointly generated by multiple historical power distribution equipment images belonging to each type item. Specifically, the method of generating the average image of multiple images in the prior art can be used. If not, it jumps to step S113. In step S112, the preprocessing module determines whether there is a similarity degree value greater than the first similarity degree value threshold. If so, it classifies the power distribution equipment image into the type item corresponding to the maximum similarity degree value and regenerates the average power distribution equipment image of the type item. If not, it jumps to step S113. In step S113, since there is no average power distribution equipment image of the type item that is the same as the power distribution equipment image for comparison, the preprocessing module classifies the power distribution equipment image into a new type item and regards the power distribution equipment image as the average power distribution equipment image of the new type item. Through the above method, the first type classification of the power distribution equipment image can be quickly performed.

[0047] Further, the preprocessing module stores in advance a classification model for further classifying the power distribution equipment image into a second category on the basis of determining the first category corresponding to the power distribution equipment image. The classification model includes different layer models from top to bottom. Each layer model corresponds to one category item included in the second category. Each layer model includes, from left to right in sequence, the identification power distribution equipment image corresponding to the corresponding category item, and a number of historical power distribution equipment images. In the order from top to bottom, the similarity degree values between the identification power distribution equipment images corresponding to the corresponding category items in different layer models and the identification power distribution equipment image of the first category decrease in sequence. In each layer model, in the order from left to right, the similarity degree values between the number of historical power distribution equipment images and the identification power distribution equipment image decrease in sequence.

[0048] Specifically, in order to further classify the power distribution equipment image into a second category on the basis of determining the first category corresponding to the power distribution equipment image, the preprocessing module stores in advance the classification model. In the order from top to bottom, the classification model can be divided into different layer models. Each layer model corresponds to one category item included in the second category. In the order from left to right, each layer model includes, in sequence, the identification power distribution equipment image corresponding to the corresponding category item, which is jointly generated by a number of historical power distribution equipment images, and a number of historical power distribution equipment images. The generation method is the same as the method for generating the above-mentioned average power distribution equipment image. It should be noted that the identification power distribution equipment images corresponding to the corresponding category items in different layer models are connected in sequence in the vertical direction. In addition, in the order from top to bottom, the similarity degree values between the identification power distribution equipment images corresponding to the corresponding category items in different layer models and the identification power distribution equipment image of the first category decrease in sequence. As mentioned above, a number of historical power distribution equipment images have been accumulated under different category items included in the first category. Then, the identification power distribution equipment image of the first category is jointly generated by all the historical power distribution equipment images under the first category, and the generation method is also the same as the method for generating the above-mentioned average power distribution equipment image. In each layer model, in the order from left to right, the similarity degree values between the number of historical power distribution equipment images and the identification power distribution equipment image decrease in sequence.

[0049] Further, the preprocessing module also further classifies the power distribution equipment image into a second category on the basis of determining the first category corresponding to the power distribution equipment image, including the following steps:

[0050] S121. In the classification model, the preprocessing module determines whether it has reached the end of the model. If so, it jumps to S124. If not, the preprocessing module calculates the similarity degree value between the power distribution equipment image and the identification power distribution equipment image in the current layer model, and continues with S122;

[0051] S122. The preprocessing module checks whether the similarity value is greater than a preset second similarity value threshold. If not, it moves to the next layer model and jumps to S121. If yes, it continues with S123;

[0052] S123. The preprocessing module determines whether it has reached the end of the model. If yes, it jumps to S124. If not, it moves right in the current layer model. The preprocessing module checks whether the similarity value is greater than the similarity value between the current historical power distribution equipment image and the identified power distribution equipment image in the current layer model. If not, it repeats this step. If yes, it adds the power distribution equipment image before the current historical power distribution equipment image and continues with S125;

[0053] S124. After the preprocessing module adds the power distribution equipment image to the end of the model, it continues with S125;

[0054] S125. The preprocessing module updates the identified power distribution equipment image in the layer model with the added power distribution equipment image and updates the order of the different layer models in the partitioning model in the up-down direction.

[0055] Specifically, the detailed process of the preprocessing module continuing to perform the second type classification on the power distribution equipment image is introduced here. The preprocessing module starts from the identification power distribution equipment image in the top layer model of the classification model and executes steps S121 to S125. In step S121, in the classification model, the preprocessing module determines whether it has reached the end of the model. If not, the preprocessing module calculates the similarity value between the power distribution equipment image and the identification power distribution equipment image in the current layer model, and continues to step S122. If so, it jumps to step S124. In step S122, the preprocessing module checks whether the similarity value is greater than the second similarity value threshold, and the second similarity value threshold is greater than the above first similarity value threshold. If not, at this time, the power distribution equipment image is no longer compared with the historical power distribution equipment image in the current layer model, and it moves to the next layer model, that is, from the identification power distribution equipment image in the current layer model to the identification power distribution equipment image in the lower layer model. This can reduce the processing time and jumps to step S121. If so, it continues to step S123. In step S123, the preprocessing module determines whether it has reached the end of the model. If so, it jumps to step S124. If not, it moves to the right in the current layer model and can move to the historical power distribution equipment image. The preprocessing module checks whether the similarity value is greater than the similarity value between the current historical power distribution equipment image and the identification power distribution equipment image in the current layer model. If so, it inserts the power distribution equipment image before the current historical power distribution equipment image and continues to step S125. If not, it repeats this step. In step S124, since it has reached the end of the model, the preprocessing module adds the power distribution equipment image after the end of the model and continues to step S125. In step S125, the preprocessing module updates the identification power distribution equipment image in the layer model with the added power distribution equipment image, and updates the order of different layer models in the classification model in the up and down directions.

[0056] Further, the method for the route planning module to generate the next route based on the basic route includes: Method 1: Select any two power distribution equipment from all the power distribution equipment on the basic route and exchange the positions of the two power distribution equipment on the basic route. Method 2: Select any one power distribution equipment from all the power distribution equipment that belongs to the second group but is not on the basic route and add the power distribution equipment to the basic route. Method 3: Select any one power distribution equipment from all the power distribution equipment that belongs to the second group on the basic route and delete the power distribution equipment from the basic route.

[0057] Specifically, a basic route has been generated above. After that, any one of Method 1 to Method 3 can be arbitrarily selected to generate the next route. Method 1 refers to selecting any two power distribution devices from all the power distribution devices on the basic route and swapping the positions of the two power distribution devices on the basic route. Method 2 refers to selecting any one power distribution device from all the power distribution devices that belong to the second group but are not on the basic route and adding the power distribution device to the basic route. Method 3 refers to selecting any one power distribution device from all the power distribution devices that belong to the second group on the basic route and deleting the power distribution device from the basic route. Through the above methods, the next route can be conveniently and quickly generated based on the basic route, reducing the processing burden.

[0058] Further, the route planning module calculates the eigenvalue of the route through the following formula:

[0059] θ = χ1 + χ2 + χ3

[0060] Where θ is the eigenvalue, χ1 is the value representing the difference degree between the actual inspection time and the preset standard inspection time. The actual inspection time is the time for inspection according to the route. χ2 is the value representing the total inspection distance of the route. χ3 is the value representing the degree to which the power distribution devices with a high degree of importance on the route are inspected in advance;

[0061] Further, χ1 includes two cases: when the actual inspection time is greater than the standard inspection time, χ1 = γ * (α - β), where γ is the total number of all power distribution devices that belong to the second group on the route, α is the actual inspection time, and β is the standard inspection time; when the actual inspection time is less than or equal to the standard inspection time, χ1 = ω * (β - α), where ω is the total number of all power distribution devices that belong to the second group but are not on the route, β is the standard inspection time, and α is the actual inspection time;

[0062] Further, Where m is the total number of power distribution devices on the route, n is the appearance order of the power distribution device on the route, E is a function to find the larger value of two numerical values, κ n is the importance value of the nth power distribution device on the route, and the value of σ is 0;

[0063] Specifically, the route planning module calculates the eigenvalue of the route through the above formula θ = χ1 + χ2 + χ3. The smaller the eigenvalue, the more suitable the generated route is. The above formula θ = χ1 + χ2 + χ3 comprehensively considers the difference degree between the actual inspection time and the standard inspection time, the total inspection distance of the route, and the degree to which the distribution equipment with a high degree of importance on the route is inspected in advance. Regarding χ1, the actual inspection time is the time for inspecting according to the route, including the time for moving between different distribution equipment according to the route and the time for maintaining different distribution equipment. The greater the difference degree between the actual inspection time and the standard inspection time, the more inappropriate the generated route means. Regarding χ2, the shorter the total inspection distance of the route, the better. The shorter the inspection distance, the lower the inspection cost. Regarding χ3, the greater the degree to which the distribution equipment with a high degree of importance on the route is inspected in advance, the more it can ensure the stability of the distribution line. The degree to which the distribution equipment with a high degree of importance on the route is inspected in advance is calculated through the formula and the earlier the route for inspecting the distribution equipment with a high degree of importance, the greater the value of the corresponding calculation result. Also, because the values of χ1 and χ2 are both the smaller the better, the opposite number of the calculation result is taken in χ3. Through the above method, it is possible to generate an inspection route with a small difference between the actual inspection time and the standard inspection time, a small total inspection distance, and the inspection of the distribution equipment with a high degree of importance in advance as much as possible.

[0064] According to another aspect of the embodiments of the present invention, as shown in reference to Figure 2 there is also provided an inspection path planning system based on an unmanned aerial vehicle, including an unmanned aerial vehicle module, a preprocessing module, and a route planning module, which are used to implement the inspection path planning method based on an unmanned aerial vehicle as described above.

[0065] Among them, the functions of each module are as follows:

[0066] The unmanned aerial vehicle module is used to collect distribution equipment images for each distribution equipment located on the distribution line and transmit the distribution equipment images to the preprocessing module;

[0067] The preprocessing module is used to perform the first type classification on the distribution equipment images to determine the first-level type corresponding to the distribution equipment images, and on the basis of determining the first-level type corresponding to the distribution equipment images, continue to perform the second type classification on the distribution equipment images to determine the second-level type corresponding to the distribution equipment images, and is used to send different distribution equipment images and the second-level types respectively corresponding to different distribution equipment images to the route planning module;

[0068] A route planning module is used to divide the power distribution equipment corresponding to the power distribution equipment image corresponding to the secondary category that meets the preset requirements into the first group, and divide all other power distribution equipment into the second group. It also generates a basic route, calculates the eigenvalue of the basic route, and is used to generate the next route based on the basic route, calculate the eigenvalue of the next route, determine whether the eigenvalue of the next route is less than the eigenvalue of the basic route. If so, the next route is regarded as the basic route, check whether the eigenvalue of the basic route is less than the preset eigenvalue threshold. If not, directly check whether the eigenvalue of the basic route is less than the preset eigenvalue threshold. If it is less, the basic route is used as the final route. If it is greater than or equal to, continue to generate the next route.

[0069] According to another aspect of the embodiments of the present invention, there is also provided a storage medium storing program instructions, wherein when the program instructions run, they control the device where the storage medium is located to execute the method in any one of the above.

[0070] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order limit for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages do not necessarily have to be executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages does not necessarily have to be sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0071] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The above program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0072] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0073] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the invention patent of the present invention should be subject to the appended claims.

[0074] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for inspection path planning based on drone, characterized in that: The method comprises the following steps: S1. For each power distribution equipment located on the power distribution line, the drone module collects the power distribution equipment image of the power distribution equipment, transmits the power distribution equipment image to the preprocessing module, and the preprocessing module performs a first classification on the power distribution equipment image to determine the primary category corresponding to the power distribution equipment image. The preprocessing module also continues to perform a second classification on the power distribution equipment image on the basis of determining the primary category corresponding to the power distribution equipment image, so as to determine the secondary category corresponding to the power distribution equipment image; S2, the preprocessing module sends different power distribution equipment images and the secondary categories corresponding to different power distribution equipment images to the route planning module, and the route planning module divides the power distribution equipment corresponding to the power distribution equipment images corresponding to the secondary categories that meet the preset requirements into the first group, and divides all other power distribution equipment into the second group, and the route planning module also generates a basic route and calculates the characteristic value of the basic route, and the basic route includes all the power distribution equipment in the first group; S3, the route planning module generates a next route based on the basic route, calculates a characteristic value of the next route, and determines whether the characteristic value of the next route is less than the characteristic value of the basic route. If yes, the next route is regarded as the basic route, and the characteristic value of the basic route is checked to be less than a preset characteristic value threshold. If no, the characteristic value of the basic route is directly checked to be less than the preset characteristic value threshold. When the characteristic value of the basic route is less than the preset characteristic value threshold, the basic route is used as the final route. When the characteristic value of the basic route is greater than or equal to the preset characteristic value threshold, this step is repeated.

2. The inspection path planning method based on drone according to claim 1 is characterized in that: The preprocessing module performs a first classification on the power distribution equipment image, including the following steps: S111, the preprocessing module determines whether the number of category items currently included in the first-level category is greater than zero. If not, jump to S113. If yes, the preprocessing module respectively calculates the same degree value between the power distribution equipment image and the average power distribution equipment image of each category item; S112, the preprocessing module determines whether there is a same degree value greater than a preset first same degree value threshold, and if not, jumps to S113, and if so, the preprocessing module divides the power distribution equipment image into category items corresponding to the largest same degree value, and updates the average power distribution equipment image of the category item; S113. The preprocessing module divides the power distribution equipment image into new categories and regards the power distribution equipment image as an average power distribution equipment image of the new categories.

3. The inspection path planning method based on drone according to claim 2 is characterized in that: The preprocessing module stores in advance a division model for continuing to perform a second category division on the distribution equipment image based on the determination of the completion of the first-level category corresponding to the distribution equipment image. The division model includes different layer models from top to bottom, each layer model corresponds to a category item included in the second-level category, and each layer model includes, from left to right, an identification distribution equipment image of the corresponding category item, and a number of historical distribution equipment images. From top to bottom, the degree of similarity between the identification distribution equipment image of the corresponding category item in different layer models and the identification distribution equipment image of the first-level category decreases successively. In each layer model, the degree of similarity between a number of historical distribution equipment images and the identification distribution equipment image decreases successively from left to right.

4. The inspection path planning method based on drone according to claim 3 is characterized in that: The preprocessing module further performs a second classification of the power distribution equipment image based on the determination of the first-level category corresponding to the power distribution equipment image, including the following steps: S121, in the partition model, the preprocessing module determines whether it reaches the end of the model, and if so, jumps to S124, and if not, the preprocessing module calculates the same degree value between the power distribution equipment image and the identified power distribution equipment image in the current layer model, and continues to S122; S122, the preprocessing module checks whether the same degree value is greater than a preset second same degree value threshold, if not, moves to the next layer model and jumps to S121, if yes, continues to S123; S123, the preprocessing module determines whether it has reached the end of the model. If yes, jump to S124. If no, move right in the current layer model. The preprocessing module checks whether the same degree value is greater than the same degree value between the current historical power distribution device image and the identified power distribution device image in the current layer model. If no, repeat this step. If yes, add the power distribution device image to the current historical power distribution device image and continue to S125. S124, after the preprocessing module adds the distribution equipment image to the end of the model, proceed to S125; S125. The preprocessing module updates the identified power distribution equipment image in the layer model to which the power distribution equipment image is added, and updates the order of different layer models in the partition model in the up-down direction.

5. The inspection path planning method based on drone according to claim 1 is characterized in that: The method for the route planning module to generate the next route based on the basic route includes: method one: selecting any two distribution devices from all the distribution devices on the basic route, and exchanging the positions of the two distribution devices on the basic route; method two: selecting any one distribution device from all the distribution devices belonging to the second group but not on the basic route, and adding the distribution device to the basic route; method three: selecting any one distribution device from all the distribution devices belonging to the second group on the basic route, and deleting the distribution device from the basic route.

6. The inspection path planning method based on drone according to claim 5 is characterized in that: The route planning module calculates the characteristic value of the route through the following formula: in, is the characteristic value, χ1 is the value representing the difference between the actual inspection time and the preset standard inspection time, the actual inspection time is the time for inspection along the route, χ2 is the value representing the total inspection distance of the route, and χ3 is the value representing the degree to which the important distribution equipment on the route is inspected in advance.

7. The inspection path planning method based on drone according to claim 6 is characterized in that: χ1 includes two cases: when the actual inspection time is greater than the standard inspection time, χ1=γ*(α-β), where γ is the total number of all distribution equipment belonging to the second group on the route, α is the actual inspection time, and β is the standard inspection time; when the actual inspection time is less than or equal to the standard inspection time, χ1=ω*(β-α), where ω is the total number of all distribution equipment belonging to the second group but not on the route, β is the standard inspection time, and α is the actual inspection time.

8. The inspection path planning method based on drone according to claim 7 is characterized in that: Where m is the total number of distribution equipment on the route, n is the order of appearance of the distribution equipment on the route, E is the function to find the larger of two values, κ n is the importance value of the nth distribution equipment on the route, and the value of σ is 0.

9. A drone-based inspection path planning system, characterized in that: Includes the following modules: The drone module is used to collect images of the power distribution equipment for each power distribution equipment located on the power distribution line, and transmit the images of the power distribution equipment to the preprocessing module; a preprocessing module, configured to perform a first classification on the power distribution equipment image to determine the primary category corresponding to the power distribution equipment image, and further perform a second classification on the power distribution equipment image to determine the secondary category corresponding to the power distribution equipment image on the basis of determining the primary category corresponding to the power distribution equipment image, and to send different power distribution equipment images and the secondary categories respectively corresponding to different power distribution equipment images to the route planning module; a route planning module, for classifying the power distribution equipment corresponding to the power distribution equipment image corresponding to the secondary category that meets the preset requirements into the first group, and classifying all other power distribution equipment into the second group, and further generating a basic route, calculating the characteristic value of the basic route, and for generating a next route based on the basic route, calculating the characteristic value of the next route, and judging whether the characteristic value of the next route is less than the characteristic value of the basic route, and if so, treating the next route as the basic route, and checking whether the characteristic value of the basic route is less than a preset characteristic value threshold, and if not, directly checking whether the characteristic value of the basic route is less than the preset characteristic value threshold; When the characteristic value of the basic route is less than the preset characteristic value threshold, the basic route is used as the final route. When the characteristic value of the basic route is greater than or equal to the preset characteristic value threshold, the next route is generated.

10. A storage medium, characterized in that: The storage medium stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 8.

Images