Clustering-based sensing equipment distribution method and device, medium and equipment

Abstract

The invention relates to the field of data processing, in particular to a sensing equipment distribution method and device based on clustering, a medium and equipment.The method comprises the steps that according to intersection scores corresponding to all initial intersections, the intersection aggregation degree corresponding to a target area is obtained, and the clustering number N is obtained in combination with shape scores and area areas; the method comprises the following steps: acquiring position coordinates of initial intersections, performing clustering in combination with the position coordinates of the initial intersections to obtain N first clustering sets, acquiring a clustering score according to the number of the initial intersections in each first clustering set to represent the clustering effect of the N first clustering sets, and if the clustering score is smaller than a preset clustering score threshold, determining that the clustering effect of the N first clustering sets is good or bad. If yes, updating the clustering number to W and replacing the clustering number N, repeatedly executing clustering operation until the clustering score is greater than or equal to the clustering score threshold value, obtaining a second clustering set, and arranging preset sensing equipment according to the second clustering set, thereby improving the clustering effect of the initial intersection, and improving the distribution accuracy of the sensing equipment.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a clustering-based sensing device deployment method, device, medium and equipment. Background Art

[0002] Traffic roadside perception is an intelligent analysis technology that follows certain rules and analyzes roadside data sensed by sensors to output information on various road traffic elements, abnormal events, and traffic-related parameters. Through the effective construction of roadside equipment, it can provide highway operation managers with a full range of road condition perception, improve business operation efficiency and management level; at the same time, combined with vehicle-road collaborative technology, it can also provide effective perception redundancy for vehicles running on highways, improve vehicle operation safety, and reduce the cost of autonomous driving construction. Therefore, the advanced roadside perception system is an important foundation and base for smart highways, and is also an important symbol to distinguish the new generation of smart highways.

[0003] In the prior art, sensing devices are usually deployed at intersections to collect information on various road traffic elements, abnormal events, and traffic-related parameters. Therefore, the deployment of sensing devices plays an important role in smart roads. How to improve the accuracy of the deployment of sensing devices has become an urgent problem to be solved. Summary of the invention

[0004] In view of the above technical problems, the technical solution adopted by the present invention is a clustering-based sensing device deployment method, which includes the following steps: S1, obtaining the shape score, area of ​​the target area, intersection score and position coordinates corresponding to each initial intersection in the target area.

[0005] S2, obtaining the intersection aggregation degree corresponding to the target area according to the intersection scores corresponding to all initial intersections.

[0006] S3, obtaining the number of clusters N according to the shape score, the area and the degree of intersection aggregation.

[0007] S4, clustering all the initial intersections according to the number of clusters N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets.

[0008] S5. According to the number of initial intersections included in each first clustering set, clustering scores corresponding to the N first clustering sets are obtained.

[0009] S6: If the cluster score is less than a preset cluster score threshold, the number of clusters N is updated according to a preset step size to obtain an updated number of clusters W.

[0010] S7, using the updated cluster number W to replace the cluster number N, and repeating steps S4-S6 until the cluster score is greater than or equal to a preset cluster score threshold, and determining the corresponding first cluster set as the second cluster set.

[0011] S8, arranging preset sensing devices according to the second cluster set.

[0012] The present invention also provides a clustering-based sensing device deployment device, the clustering-based sensing device deployment device comprising: The data acquisition module is used to obtain the shape score, area of ​​the target area, intersection score and position coordinates corresponding to each initial intersection in the target area.

[0013] The intersection concentration degree acquisition module is used to obtain the intersection concentration degree corresponding to the target area according to the intersection scores corresponding to all initial intersections.

[0014] The cluster number acquisition module is used to obtain the cluster number N according to the shape score, area and intersection aggregation degree.

[0015] The first clustering module is used to cluster all the initial intersections according to the number of clusters N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets.

[0016] The cluster score acquisition module is used to acquire cluster scores corresponding to N first cluster sets according to the number of initial intersections included in each first cluster set.

[0017] The cluster quantity updating module is used to update the cluster quantity according to a preset step size if the cluster score is less than a preset cluster score threshold.

[0018] The second clustering module is used to replace the cluster number N with the updated cluster number, repeatedly execute the first clustering module, the cluster score acquisition module and the cluster number update module until the cluster score is greater than or equal to the preset cluster score threshold, and determine the corresponding first cluster set as the second cluster set.

[0019] The sensing device deployment module is used to deploy preset sensing devices according to the second cluster set.

[0020] The present invention also provides a non-transitory computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the above-mentioned clustering-based perception device deployment method.

[0021] The present invention also provides an electronic device, comprising a processor and the above-mentioned non-transitory computer-readable storage medium.

[0022] The present invention has at least the following beneficial effects: according to the intersection scores corresponding to all initial intersections, the differences between the intersection scores of different initial intersections are measured, the consistency between the intersections and the distribution of the intersections are characterized, so as to obtain the intersection aggregation degree corresponding to the target area, and according to the corresponding relationship between the shape score, the area and the intersection aggregation degree and the cluster number N, the accuracy of the cluster number N is improved; according to the cluster number N and the position coordinates corresponding to all initial intersections, all initial intersections are clustered to obtain N first cluster sets, and according to the number of initial intersections included in each first cluster set, the number of initial intersections included in each first cluster set is obtained. The clustering scores corresponding to the N first clustering sets are obtained to characterize the quality of the clustering effects corresponding to the N first clustering sets. If the clustering score is less than the preset clustering score threshold, the number of clusters is updated according to the preset step size, and the updated number of clusters is used instead of the number of clusters N. The clustering operation is repeated until the clustering score is greater than or equal to the preset clustering score threshold, and the corresponding first clustering set is determined as the second clustering set. The preset perception equipment is arranged according to the second clustering set, which improves the clustering effect of the initial intersection, thereby improving the accuracy of the second clustering set, and further improving the accuracy of the placement of the perception equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 A flowchart of a method for deploying sensing devices based on clustering provided in Embodiment 1 of the present invention; Figure 2 A flowchart of a method for deploying sensing devices based on intersection screening provided in Embodiment 2 of the present invention; Figure 3 A flowchart of a method for deploying sensing devices provided in Embodiment 3 of the present invention; Figure 4 A schematic diagram of the structure of a clustering-based sensing device deployment device provided in Example 4 of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It is understood that, where appropriate, the above-mentioned terms used to distinguish similar objects can be interchanged so that the present invention can also implement other embodiments other than the above-mentioned illustrated embodiments or described embodiments. In addition, the terms "including" and "having" and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Embodiment 1 The first embodiment provides a method for arranging sensing devices based on clustering. The method for arranging sensing devices based on clustering includes the following steps: Figure 1 As shown: S1, obtaining the shape score, area of ​​the target area, intersection score and position coordinates corresponding to each initial intersection in the target area.

[0028] Among them, perception equipment can include sensors such as cameras, laser radars, and millimeter-wave radars, which are used to collect information on various road traffic elements, abnormal events, and traffic-related parameters, and can provide highway operation managers with a full range of road condition perception, improving business operation efficiency and management level. At the same time, combined with vehicle-road collaborative technology, it can also provide effective perception redundancy for vehicles running on highways, improve the safety of vehicle operation, and reduce the cost of autonomous driving construction.

[0029] The target area is the geographical area where sensing equipment needs to be deployed. The shape score of the target area is used to measure the irregularity and narrowness of the shape of the target area. The intersection score of the initial intersection is used to measure the intersection importance and intersection traffic data of the initial intersection. The number of cluster sets corresponding to the target area is determined by combining the shape score of the target area, the area of ​​the area and the intersection score corresponding to the initial intersection. This serves as the basis for clustering the intersections in the target area, thereby improving the accuracy of the intersection clustering results.

[0030] In a specific implementation, S1 further includes the following steps: S11, obtaining geometric shape parameters of the target area, wherein the geometric shape parameters include perimeter, area, rectangular area corresponding to the bounding rectangle, long circumferential line length and short circumferential line length, wherein the bounding rectangle refers to a rectangle formed by two long circumferential lines parallel to the longest axis of the target area and two short circumferential lines perpendicular to the longest axis.

[0031] S12: Obtain a first shape parameter corresponding to the target region according to the perimeter and the region area.

[0032] S13: Determine the ratio of the length of the long external tangent line to the length of the short external tangent line as a second shape parameter corresponding to the target area.

[0033] S14: Determine the ratio of the area of ​​the rectangle to the area of ​​the region as a third shape parameter corresponding to the target region.

[0034] S15, obtaining a shape score of the target area according to the first shape parameter, the second shape parameter and the third shape parameter.

[0035] Among them, the first shape parameter D1=L 2 / (4×π×S), L refers to the circumference, π refers to pi, and correspondingly, the first shape parameter can characterize the shape regularity of the target area. Correspondingly, the larger the first shape parameter is, the lower the shape regularity of the target area is, and the more complex the situation of the intersection in the target area is, then more clustering sets need to be set to cluster the intersections to improve the precision and accuracy of the clustering results.

[0036] Two long circumferential lines parallel to the longest axis of the target area and two short circumferential lines perpendicular to the longest axis are obtained, and the rectangle formed by the two long circumferential lines and the two short circumferential lines is the boundary rectangle. Correspondingly, the boundary rectangle is the minimum matrix surrounding the target area. The second shape parameter can intuitively reflect the flatness of the target area, and the third shape parameter can intuitively reflect the shape regularity of the target area. The larger the second shape parameter, the higher the flatness, the larger the relative distance between intersections, the larger the third shape parameter, the lower the shape regularity, the greater the information difference of traffic elements between intersections, the difference in the situation of abnormal events, and the difference in traffic-related parameters, then more cluster sets need to be set to cluster the intersections to improve the precision and accuracy of the clustering results.

[0037] In a specific implementation, S15 further includes the following steps: S151, obtaining a first parameter weight corresponding to the first shape parameter, a second parameter weight corresponding to the second shape parameter, and a third parameter weight corresponding to the third shape parameter.

[0038] S152, obtaining a shape score of the target area according to the first shape parameter, the first parameter weight, the second shape parameter, the second parameter weight, the third shape parameter and the third parameter weight.

[0039] The specific values ​​of the first parameter weight, the second parameter weight and the third parameter weight may be set by the implementer according to actual conditions.

[0040] In the above, a first shape parameter is obtained according to the perimeter and the area of ​​the region, a second shape parameter is obtained according to the length of the long circumferential tangent and the length of the short circumferential tangent, and a third shape parameter is obtained according to the area of ​​the region and the area of ​​the rectangle. The first shape parameter, the second shape parameter and the third shape parameter are combined to characterize the shape regularity and flatness of the target area, and a shape score of the target area is obtained as a basis for judging the number of cluster sets corresponding to the target area, thereby improving the accuracy of the number of clusters.

[0041] In a specific implementation, S1 further includes the following steps: S16, obtaining the intersection flow corresponding to each initial intersection in the target area, the section type corresponding to each road section, and the section weight corresponding to each section type, where the section types include expressways, trunk roads, secondary trunk roads, and branch roads.

[0042] S17, obtaining a first score corresponding to each initial intersection according to the section types of all sections corresponding to each initial intersection and the section weight corresponding to each section type.

[0043] S18, obtaining a second score corresponding to each initial intersection according to the proportion of the intersection flow corresponding to each initial intersection in the intersection flow corresponding to all initial intersections.

[0044] S19, obtaining an intersection score corresponding to each initial intersection according to the first score and the second score corresponding to each initial intersection.

[0045] The specific value of the road section weight corresponding to each road section type can be set by the implementer according to the actual situation.

[0046] Each initial intersection may correspond to several road sections. According to the road section weight corresponding to each road section type, the sum of the road section weights of all road sections corresponding to each initial intersection is determined as the first score corresponding to each initial intersection.

[0047] Intersection traffic may include motor vehicle traffic, non-motor vehicle traffic, and pedestrian traffic. By combining the first score and the second score corresponding to each initial intersection, the intersection score corresponding to each initial intersection is obtained to characterize the importance of each initial intersection, which serves as the basis for the deployment of sensing equipment.

[0048] In a specific implementation, S19 further includes the following steps: S191, obtaining a first score importance corresponding to the first score and a second score importance corresponding to the second score.

[0049] S192, for any initial intersection, according to the first score, the first score importance, the second score and the second score importance, obtain the intersection score corresponding to the current initial intersection.

[0050] S193, traverse all initial intersections and obtain the intersection score corresponding to each initial intersection.

[0051] The specific values ​​of the first score importance and the second score importance can be set by the implementer according to actual conditions.

[0052] As described above, the intersection score corresponding to each initial intersection is obtained by combining the intersection flow and section type of the initial intersection to measure the importance of the initial intersection, which serves as the basis for the deployment of perception devices and improves the accuracy of the deployment of perception devices.

[0053] S2, obtaining the intersection aggregation degree corresponding to the target area according to the intersection scores corresponding to all initial intersections.

[0054] In a specific implementation, the intersection concentration level corresponding to the target area meets the following conditions: B=M / (Σ i=1 M (A i -A 0 ) 2 ), where B refers to the degree of intersection aggregation corresponding to the target area, M refers to the total number of initial intersections in the target area, and A i It refers to the intersection score corresponding to the i-th initial intersection, i=1, 2, ..., M, A 0 It refers to the average of the intersection scores of the M initial intersections in the target area.

[0055] Among them, the degree of intersection aggregation corresponding to the target area can measure the difference between the intersection scores of different initial intersections, thereby characterizing the consistency between intersections and the distribution of intersections, which serves as the basis for determining the number of cluster sets corresponding to the target area.

[0056] Correspondingly, the greater the degree of intersection aggregation, the smaller the information differences of traffic elements between intersections, the differences in the situations of abnormal events, and the differences in traffic-related parameters, and a smaller number of cluster sets are needed to cluster intersections in order to improve the precision and accuracy of the clustering results.

[0057] As described above, by measuring the differences between the intersection scores of different initial intersections, the consistency between intersections and the distribution of intersections can be characterized, thereby obtaining the degree of intersection aggregation, which serves as the basis for determining the number of cluster sets corresponding to the target area, thereby improving the accuracy of the number of clusters.

[0058] S3, obtaining the number of clusters N according to the shape score, the area and the degree of intersection aggregation.

[0059] Among them, the shape score and regional area are positively correlated with the number of clusters, and the degree of intersection aggregation is negatively correlated with the number of clusters.

[0060] In a specific implementation, the number of clusters N meets the following conditions: N = ((α1 × e -B )+(α2×((2 / (1+e -C )))-1)+(α3×(S / S 0 )))×N 0 , where α1 refers to the first preset quantity weight, α2 refers to the second preset quantity weight, α3 refers to the third preset quantity weight, S refers to the area, S 0 Refers to the preset reference area, S 0 >S,N 0 refers to the preset reference cluster number, e refers to the natural constant, and C refers to the shape score, where C meets the following conditions: C=β1×D1+β2×D2+β3×D3, where β1 refers to the first preset scoring weight, β2 refers to the second preset scoring weight, β3 refers to the third preset scoring weight, D1 refers to the first shape parameter, D2 refers to the second shape parameter, D3 refers to the third shape parameter, and D1 meets the following conditions: D1=L 2 / (4×π×S), where L refers to the circumference and π refers to the ratio of a circle to a circle.

[0061] Among them, the first preset quantity weight α1, the second preset quantity weight α2, the third preset quantity weight α3, the preset reference area area S 0 , the preset number of reference clusters N 0 The specific values ​​of the first preset scoring weight β1, the second preset scoring weight β2 and the third preset scoring weight β3 can be set by the implementer according to actual conditions.

[0062] As described above, based on the correspondence between the shape score, the area, the intersection aggregation degree and the cluster number N, the cluster number N is obtained as the basis for clustering the initial intersections, thereby improving the accuracy of the cluster number N and the clustering accuracy of the initial intersections.

[0063] S4, clustering all the initial intersections according to the number of clusters N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets.

[0064] Among them, on the basis of determining the number of clusters N, all initial intersections can be clustered using a K-means clustering algorithm according to the position coordinates corresponding to the initial intersections to obtain N first cluster sets.

[0065] S5. According to the number of initial intersections included in each first clustering set, clustering scores corresponding to the N first clustering sets are obtained.

[0066] Among them, in order to ensure the balanced collection of information on traffic elements, abnormal events and traffic-related parameters between intersections in each first clustering set, this embodiment hopes that the number of initial intersections included in the first clustering set is as equal as possible. Therefore, according to the number of initial intersections included in each first clustering set, the clustering effects of the N first clustering sets are scored. Correspondingly, the higher the clustering score, the better the clustering effect of the N first clustering sets.

[0067] In a specific implementation, the clustering score meets the following conditions: F=(Σ i=1 N (E j -E 0 ) 2 ) / N, where F refers to the clustering scores corresponding to the N first clustering sets, N refers to the total number of first clustering sets, and E j Refers to the number of initial intersections included in the jth first cluster set, j = 1, 2, ..., N, E 0 It refers to the average value of the number of initial intersections included in the N first clustering sets.

[0068] In the above, according to the difference between the numbers of initial intersections included in the first clustering sets, clustering scores corresponding to the N first clustering sets are obtained to measure the clustering effects of the N first clustering sets, which serve as the basis for updating the number of clusters, thereby improving the clustering effect.

[0069] S6: If the cluster score is less than a preset cluster score threshold, the number of clusters N is updated according to a preset step size to obtain an updated number of clusters W.

[0070] The preset step size and the preset clustering score threshold may be set by the implementer according to actual conditions. For example, the preset step size may be +1 or -1, which is used to adjust the number of clusters to improve the clustering effect.

[0071] In a specific embodiment, if the number of clusters is updated according to the preset step size and the corresponding clustering is performed, and the updated clustering score is less than the clustering score before the update, the preset step size can be increased or decreased, or the positive or negative nature of the preset step size can be changed, and clustering can be performed again to achieve the purpose of improving the clustering effect.

[0072] S7, using the updated cluster number W to replace the cluster number N, and repeating steps S4-S6 until the cluster score is greater than or equal to a preset cluster score threshold, and determining the corresponding first cluster set as the second cluster set.

[0073] As described above, by updating the number of clusters and clustering the initial intersections according to the updated number of clusters, the clustering effect of the initial intersections is improved, thereby improving the accuracy of the second cluster set.

[0074] S8, arranging preset sensing devices according to the second cluster set.

[0075] In a specific implementation, S8 further includes the following steps: S81, acquiring H second cluster sets corresponding to all initial intersections in the target area, and the cluster center intersection corresponding to each second cluster set, where H refers to the total number of second cluster sets.

[0076] S82, screening all initial intersections according to the maximum intersection score and the intersection score corresponding to each initial intersection to obtain candidate intersections.

[0077] S83, for the hth second cluster set, according to the total number of preset sensing devices, the ratio of the number of initial intersections in the hth second cluster set to the total number of initial intersections in the H second cluster sets, obtain the number of sensing devices corresponding to the hth second cluster set, h=1, 2, ..., H.

[0078] S84, according to the position coordinates of the cluster center intersection in the hth second cluster set and the position coordinates of each candidate intersection, determine the candidate intersection closest to the cluster center intersection as the first first target intersection corresponding to the hth second cluster set.

[0079] S85, according to the first first target intersection to the kth h The position coordinates of the first target intersection and the position coordinates of each candidate intersection in the current second cluster set will be combined with the previous k h The candidate intersection farthest from the first target intersection is determined as the kth candidate intersection corresponding to the current second cluster set. h +1 first target intersection, where k h ≥1, k is an integer.

[0080] S86, repeat step S85 until k h is equal to the number of sensing devices corresponding to the hth second clustering set.

[0081] S87, traverse all the second cluster sets, and obtain the first target intersection corresponding to each second cluster set.

[0082] S88, arranging a preset sensing device at a preset position corresponding to each first target intersection.

[0083] In a specific implementation, S8 further includes the following steps: S801, obtaining the total number of preset sensing devices and H second cluster sets corresponding to all initial intersections in the target area.

[0084] S802, for the hth second cluster set, according to the total number of preset sensing devices, the ratio of the number of initial intersections in the hth second cluster set to the total number of initial intersections in the H second cluster sets, obtain the number of sensing devices corresponding to the hth second cluster set, h=1, 2, ..., H.

[0085] S803: randomly arrange the sensing devices whose number is the same as that corresponding to the h-th second cluster set at the preset positions corresponding to the first target intersection in the h-th second cluster set.

[0086] In the above, according to the intersection scores corresponding to all the initial intersections, the differences between the intersection scores of different initial intersections are measured, and the consistency between the intersections and the distribution of the intersections are characterized, so as to obtain the intersection aggregation degree corresponding to the target area, and according to the corresponding relationship between the shape score, the area and the intersection aggregation degree and the cluster number N, the accuracy of the cluster number N is improved; according to the cluster number N and the position coordinates corresponding to all the initial intersections, all the initial intersections are clustered to obtain N first cluster sets, and according to the number of initial intersections included in each first cluster set, N first cluster sets are obtained. The clustering score corresponding to a cluster set is used to characterize the quality of the clustering effects corresponding to the N first cluster sets. If the clustering score is less than the preset clustering score threshold, the number of clusters is updated according to the preset step size, and the updated number of clusters is used instead of the number of clusters N. The clustering operation is repeated until the clustering score is greater than or equal to the preset clustering score threshold, and the corresponding first cluster set is determined as the second cluster set. The preset perception equipment is arranged according to the second cluster set, which improves the clustering effect of the initial intersection, thereby improving the accuracy of the second cluster set, and further improving the accuracy of the placement of the perception equipment.

[0087] Embodiment 2 The second embodiment provides a method for arranging sensing devices based on intersection screening. The method for arranging sensing devices based on intersection screening includes: Figure 2 As shown: S10, obtaining the intersection score corresponding to each initial intersection in the target area, the location coordinates of each initial intersection, H second cluster sets corresponding to all initial intersections in the target area, and the cluster center intersection corresponding to each second cluster set, where H refers to the total number of second cluster sets.

[0088] In a specific implementation, S10 further includes the following steps: S101, obtaining the intersection flow corresponding to each initial intersection in the target area, the section type corresponding to each road section, and the section weight corresponding to each section type, where the section types include expressways, trunk roads, secondary trunk roads, and branch roads.

[0089] S102: Obtain a first score corresponding to each initial intersection according to the section types of all sections corresponding to each initial intersection and the section weight corresponding to each section type.

[0090] S103, obtaining a second score corresponding to each initial intersection according to the proportion of the intersection flow corresponding to each initial intersection in the intersection flow corresponding to all initial intersections.

[0091] S104, acquiring an intersection score corresponding to each initial intersection according to the first score and the second score corresponding to each initial intersection.

[0092] In a specific implementation, S10 further includes the following steps: S111, obtaining the shape score, area of ​​the target area, and the intersection score and position coordinates corresponding to each initial intersection in the target area.

[0093] S112, obtaining the intersection concentration level corresponding to the target area according to the intersection scores corresponding to all the initial intersections.

[0094] S113, obtaining the number of clusters N according to the shape score, the area and the degree of intersection aggregation.

[0095] S114: Cluster all the initial intersections according to the number of clusters N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets.

[0096] S115 , acquiring clustering scores corresponding to the N first clustering sets according to the number of initial intersections included in each first clustering set.

[0097] S116: If the cluster score is less than a preset cluster score threshold, the number of clusters N is updated according to a preset step size, and an updated number of clusters W is obtained.

[0098] S117, using the updated cluster number W to replace the cluster number N, repeating steps S114-S116 until the cluster score is greater than or equal to the preset cluster score threshold, and determining the corresponding first cluster set as the second cluster set, and obtaining H second cluster sets.

[0099] S20, screening all initial intersections according to the maximum intersection score and the intersection score corresponding to each initial intersection to obtain candidate intersections.

[0100] In a specific implementation, S20 further includes the following steps: S21, obtaining the largest intersection score among all intersection scores and a preset score ratio threshold.

[0101] S22: For any initial intersection, if the ratio of the intersection score of the current initial intersection to the maximum intersection score is greater than the score ratio threshold, the current initial intersection is determined as a candidate intersection.

[0102] S23, traverse all initial intersections and obtain all candidate intersections.

[0103] The specific value of the preset score ratio threshold value can be set by the implementer according to the actual situation. For example, the preset score ratio threshold value is set to 0.5.

[0104] As described above, on the basis that the intersection score can represent the importance of the corresponding initial intersection, the initial intersections are screened according to the intersection scores corresponding to the initial intersections, and the initial intersections with high importance are selected as candidate intersections, providing a basis for equipment deployment.

[0105] S30. For the hth second cluster set, the number of sensing devices corresponding to the hth second cluster set is obtained according to the total number of preset first sensing devices, the ratio of the number of initial intersections in the hth second cluster set to the total number of initial intersections in the H second cluster sets, where h=1, 2, ..., H.

[0106] Among them, the first perception equipment can include sensors such as cameras, laser radars, millimeter-wave radars, etc., which are used to collect information on various road traffic elements, abnormal events and traffic-related parameters, and can provide highway operation managers with a full range of road condition perception, improve business operation efficiency and management level. At the same time, combined with vehicle-road collaborative technology, it can also provide effective perception redundancy for vehicles running on highways, improve the safety of vehicle operation, and reduce the cost of autonomous driving construction.

[0107] In order to ensure balanced collection of information on traffic elements, abnormal events and traffic-related parameters between intersections in each second clustering set, the greater the proportion of initial intersections in the second clustering set in the total number of initial intersections, the more sensing devices are allocated to the corresponding second clustering set.

[0108] The specific value of the total number of preset first sensing devices can be set by the implementer according to actual conditions.

[0109] In a specific implementation, the number of sensing devices corresponding to the hth second clustering set meets the following conditions: P h =ROUNDDOWN(P 0 ×(Gh / (Σ h=1 H G h )), 0), where P h refers to the number of sensing devices corresponding to the hth second cluster set, P 0 Refers to the total number of preset sensing devices, G h refers to the number of initial intersections in the hth second cluster set, ROUNDDOWN refers to a rounding down function, and the decimal place corresponding to the rounding down function is 0.

[0110] As described above, based on the proportion of the number of initial intersections in the second cluster set in the total number of initial intersections, the number of sensing devices corresponding to the second cluster set is obtained to ensure balanced collection of information on traffic elements, abnormal events and traffic-related parameters between intersections in each second cluster set, thereby improving the accuracy of the deployment of sensing devices.

[0111] S40, according to the position coordinates of the cluster center intersection in the hth second cluster set and the position coordinates of each candidate intersection, determine the candidate intersection closest to the cluster center intersection as the first first target intersection corresponding to the hth second cluster set.

[0112] As described above, in each second cluster set, the candidate intersection closest to each cluster center intersection is determined as the first first target intersection corresponding to each second cluster set as the basis for arranging the first sensing device, thereby improving the effectiveness of collecting information on traffic elements, abnormal events and traffic-related parameters between intersections, and improving the accuracy of the deployment of the first sensing device.

[0113] S50, according to the first first target intersection to the kth h The position coordinates of the first target intersection and the position coordinates of each candidate intersection in the current second cluster set will be combined with the previous k h The candidate intersection farthest from the first target intersection is determined as the kth candidate intersection corresponding to the current second cluster set. h +1 first target intersection, where k h ≥1, k is an integer.

[0114] In a specific implementation, S50 further includes the following steps: S51, for any candidate intersection in the current second clustering set, according to the first first target intersection to the kth h The position coordinates of the first target intersection and the position coordinates of the current candidate intersection are obtained, and the current candidate intersection and the first first target intersection to the kth h The total distance between the first target intersections.

[0115] S52, traverse all candidate intersections in the current second cluster set, and obtain the intersections between each candidate intersection in the current second cluster set and the first first target intersection to the kth first target intersection. h The total distance between the first target intersections.

[0116] S53, determining the candidate intersection corresponding to the largest total distance as the kth candidate intersection corresponding to the current second cluster set. h +1 first target intersection.

[0117] Among them, will be with the previous k h The candidate intersection farthest from the first target intersection is determined as the kth candidate intersection corresponding to the current second cluster set. h +1 first target intersection improves the accuracy of the first perception device's deployment, increases the collection scope of traffic element information, abnormal events, and traffic-related parameters between intersections, and increases the richness of data collected by the first perception device.

[0118] S60, repeat step S50 until k h is equal to the number of sensing devices corresponding to the hth second clustering set.

[0119] S70, traverse all second cluster sets to obtain the first target intersection corresponding to each second cluster set.

[0120] S80: Arrange a preset first sensing device at a preset position corresponding to each first target intersection.

[0121] In the above, on the basis that the intersection score can characterize the importance of the corresponding initial intersection, the initial intersections are screened according to the intersection scores corresponding to the initial intersections, and the initial intersections with high importance are selected as candidate intersections. According to the proportion of the number of initial intersections in the second cluster set in the total number of initial intersections, the number of sensing devices corresponding to the second cluster set is obtained to ensure that the information of traffic elements, abnormal events and traffic-related parameters between intersections in each second cluster set are collected in a balanced manner, thereby improving the accuracy of the placement of sensing devices; in each second cluster set, the candidate intersection closest to each cluster center intersection is determined as the first first target intersection corresponding to each second cluster set as the basis for arranging the first sensing device, thereby improving the effectiveness of collecting the information of traffic elements, abnormal events and traffic-related parameters between intersections; the candidate intersection closest to the previous k cluster center intersections is determined as the first first target intersection corresponding to each second cluster set as the basis for arranging the first sensing device, thereby improving the effectiveness of collecting the information of traffic elements, abnormal events and traffic-related parameters between intersections; h The candidate intersection farthest from the first target intersection is determined as the kth candidate intersection corresponding to the current second cluster set. h +1 first target intersection, which increases the collection scope of information on traffic elements, abnormal events and traffic-related parameters between intersections, and increases the richness of data collected by the first perception device; repeat the steps of determining the first target intersection until k hThe number of sensing devices corresponding to the h-th second clustering set is equal to that of the sensing devices, and the preset first sensing devices are arranged at the preset positions corresponding to each first target intersection, thereby improving the accuracy of the deployment of the first sensing devices, and thus improving the accuracy of the collection of information on traffic elements, abnormal events and traffic-related parameters between intersections.

[0122] Embodiment 3 The third embodiment provides a method for arranging sensing devices, and the method for arranging sensing devices includes: Figure 3 As shown: S100, obtain several candidate intersections in the target area, the road section corresponding to each candidate intersection, several first target intersections, the road section corresponding to each first target intersection, the length of each road section and the total number T of preset second sensing devices, where T is an integer greater than 0.

[0123] Among them, the second perception equipment can include sensors such as cameras, laser radars, millimeter-wave radars, etc., which are used to collect information on various road traffic elements, abnormal events and traffic-related parameters, and can provide highway operation managers with a full range of road condition perception, improve business operation efficiency and management level. At the same time, combined with vehicle-road collaborative technology, it can also provide effective perception redundancy for vehicles running on highways, improve the safety of vehicle operation, and reduce the cost of autonomous driving construction.

[0124] In a specific implementation, S100 further includes the following steps: S110, obtaining the intersection flow corresponding to each initial intersection in the target area, the section type corresponding to each road section, and the section weight corresponding to each road section type, where the section types include expressways, trunk roads, secondary trunk roads, and branch roads.

[0125] S120: Obtain a first score corresponding to each initial intersection according to the section types of all sections corresponding to each initial intersection and the section weight corresponding to each section type.

[0126] S130, obtaining a second score corresponding to each initial intersection according to the proportion of the intersection flow corresponding to each initial intersection in the intersection flow corresponding to all initial intersections.

[0127] S140, acquiring the intersection score corresponding to each initial intersection and the maximum intersection score among all the intersection scores according to the first score and the second score corresponding to each initial intersection.

[0128] S150: For any initial intersection, if the ratio of the intersection score of the current initial intersection to the maximum intersection score is greater than a preset score ratio threshold, the current initial intersection is determined as a candidate intersection.

[0129] S160, traverse all initial intersections to obtain all candidate intersections.

[0130] In a specific implementation, S100 further includes the following steps: S10, obtaining the intersection score corresponding to each initial intersection in the target area, the location coordinates of each initial intersection, H second cluster sets corresponding to all initial intersections in the target area, and the cluster center intersection corresponding to each second cluster set, where H refers to the total number of second cluster sets.

[0131] S30. For the hth second cluster set, the number of sensing devices corresponding to the hth second cluster set is obtained according to the total number of preset first sensing devices, the ratio of the number of initial intersections in the hth second cluster set to the total number of initial intersections in the H second cluster sets, where h=1, 2, ..., H.

[0132] S40, according to the position coordinates of the cluster center intersection in the hth second cluster set and the position coordinates of each candidate intersection, determine the candidate intersection closest to the cluster center intersection as the first first target intersection corresponding to the hth second cluster set.

[0133] S50, according to the first first target intersection to the kth h The position coordinates of the first target intersection and the position coordinates of each candidate intersection in the current second cluster set will be combined with the previous k h The candidate intersection farthest from the first target intersection is determined as the kth candidate intersection corresponding to the current second cluster set. h +1 first target intersection, where k h ≥1, k is an integer.

[0134] S60, repeat step S50 until k h is equal to the number of sensing devices corresponding to the hth second clustering set.

[0135] S70, traverse all second cluster sets to obtain the first target intersection corresponding to each second cluster set.

[0136] S200, for any candidate intersection, according to the road section corresponding to each candidate intersection and the road section corresponding to each first target intersection, obtain the shortest path between the current candidate intersection and each first target intersection.

[0137] Among them, for any candidate intersection, according to the road section corresponding to each candidate intersection and the road section corresponding to each first target intersection, several paths between the current candidate intersection and each first target intersection are obtained, and then according to the length of the road section included in each path, the total length corresponding to each path can be obtained, thereby screening out the shortest path between the current candidate intersection and each first target intersection as the basis for characterizing the importance of the road section.

[0138] In the above, according to the road section corresponding to each candidate intersection and the road section corresponding to each first target intersection, the shortest path between the current candidate intersection and each first target intersection is obtained, which provides a data basis for characterizing the importance of the road section.

[0139] S300, obtaining the priority of each road segment in each shortest path according to the length of each road segment, a preset first length threshold and a preset second length threshold.

[0140] The specific values ​​of the preset first length threshold and the preset second length threshold may be set by the implementer according to actual conditions. For example, the preset first length threshold may refer to 1 km, and the preset second length threshold may refer to 3 km.

[0141] In a specific implementation, S300 further includes the following steps: S310: For any shortest path, if the sum of the lengths of all segments in the current shortest path is less than or equal to a preset first length threshold, all segments included in the current shortest path are determined as high priority segments.

[0142] S320: If the sum of the lengths of all the sections in the current shortest path is greater than a preset first length threshold and less than or equal to a preset second length threshold, all the sections included in the current shortest path are determined as medium priority sections.

[0143] S330: If the sum of the lengths of all the sections in the current shortest path is greater than a preset second length threshold, all the sections included in the current shortest path are determined as low priority sections.

[0144] Among them, in the travel of motor vehicles, non-motor vehicles and pedestrians, short-distance paths have a higher priority than long-distance paths. Therefore, on the basis that the preset first length threshold is less than the preset second length threshold, the priority of the road sections contained in each shortest path is determined according to the comparison result of the total length of each path with the preset first length threshold and the preset second length threshold, as the basis for measuring the importance of each intersection.

[0145] As described above, based on the comparison results of the total length of each shortest path with the preset first length threshold and the preset second length threshold, the priority of each section in the shortest path is obtained, which is used to characterize the importance of the corresponding section in collecting information on various road traffic elements, abnormal events and traffic-related parameters, providing a data basis for the layout of perception equipment, thereby improving the accuracy of the layout of the second perception equipment.

[0146] S400: Obtain the coverage score corresponding to the current candidate intersection according to the priorities of all road sections in all shortest paths.

[0147] Among them, according to the priority of each road section, the road section situation between each candidate intersection and the first target intersection where the first sensing device is arranged in the target area when each candidate intersection is used as the source of the shortest path can be obtained, and the corresponding coverage score when each candidate intersection is used as the source of the shortest path can be further obtained, which is used to characterize the importance of the corresponding candidate intersection in collecting information on various road traffic elements, abnormal events and traffic-related parameters, as the basis for screening the location of the second sensing device.

[0148] In a specific implementation, S400 further includes the following steps: S410 , according to the priorities of all road sections in all shortest paths, obtain X high priority road sections, Y medium priority road sections, the length of each high priority road section and the length of each medium priority road section, where X and Y are integers greater than or equal to 0.

[0149] S420, according to the total length of all road sections in the target area, the X high-priority road sections, the Y medium-priority road sections, the length of each high-priority road section, and the length of each medium-priority road section, obtain a coverage score Q corresponding to the current candidate intersection, where Q meets the following conditions: Q=(R 1x +R 2y / U) / R 0 , where R 1x is the length of the xth high priority section, x=1, 2, ..., X, X is the total number of high priority sections, R 2y refers to the length of the y-th medium-priority road section, y=1, 2, ..., Y, Y refers to the total number of medium-priority road sections, R 0 It refers to the total length of all road sections in the target area, and U refers to the preset coverage score coefficient.

[0150] Among them, when calculating the coverage score, the weight of the high-priority section is higher than that of the medium-priority section, and the weight of the low-priority section is the lowest. Therefore, this embodiment sets the weight corresponding to the high-priority section to 1, uses the preset coverage score coefficient to reduce the weight of the medium-priority section, and sets the weight of the low-priority section to 0. Based on this, the coverage score corresponding to each candidate intersection is obtained to characterize the importance of each candidate intersection in collecting information on various road traffic elements, abnormal events and traffic-related parameters.

[0151] In one specific embodiment, U=H.

[0152] As described above, based on the priorities of all road sections in all shortest paths, the corresponding coverage score of each candidate intersection as the source of the shortest path is obtained, which is used to characterize the importance of the corresponding candidate intersection in collecting information on various road traffic elements, abnormal events and traffic-related parameters. This serves as the basis for screening the location of the second perception device, thereby improving the accuracy of the location of the second perception device.

[0153] S500, traverse all candidate intersections and obtain the coverage score corresponding to each candidate intersection.

[0154] S600: Filter out T second target intersections according to the total number T of preset second sensing devices and the coverage score corresponding to each candidate intersection.

[0155] In a specific implementation, S600 further includes the following steps: S610: Sort all candidate intersections in descending order of coverage scores to obtain an intersection sorting result.

[0156] S620: Determine the first T candidate intersections in the intersection ranking result as T second target intersections.

[0157] As described above, T candidate intersections with the largest coverage scores are screened out as the second target intersection, which increases the importance of collecting information on various road traffic elements, abnormal events and traffic-related parameters, and improves the accuracy of the second perception device's layout when used as the second perception device's layout location.

[0158] S700: Arrange a preset second sensing device at a preset position corresponding to each second target intersection.

[0159] In the above, according to the road section corresponding to each candidate intersection and the road section corresponding to each first target intersection, the shortest path between the current candidate intersection and each first target intersection is obtained, which provides a data basis for characterizing the importance of the road section; according to the comparison result of the length of each road section with the preset first length threshold and the preset second length threshold, the priority of each road section in the shortest path is obtained, which is used to characterize the importance of the corresponding road section in collecting information, abnormal events and traffic-related parameters of various road traffic elements; according to the priority of all road sections in all shortest paths, the corresponding coverage score of each candidate intersection as the source of the shortest path is obtained, which is used to characterize the importance of the corresponding candidate intersection in collecting information, abnormal events and traffic-related parameters of various road traffic elements; T candidate intersections with the largest coverage scores are selected as the second target intersections, which improves the importance of collecting information, abnormal events and traffic-related parameters of various road traffic elements, and the preset second perception device is arranged at the preset position corresponding to each second target intersection, which improves the accuracy of the layout of the second perception device.

[0160] Embodiment 4 The fourth embodiment provides a clustering-based sensing device deployment device, and the clustering-based sensing device deployment device includes: Figure 4 As shown: The data acquisition module 41 is used to acquire the shape score, area of ​​the target area, and the intersection score and position coordinates corresponding to each initial intersection in the target area.

[0161] The intersection concentration degree acquisition module 42 is used to acquire the intersection concentration degree corresponding to the target area according to the intersection scores corresponding to all the initial intersections.

[0162] The cluster number acquisition module 43 is used to acquire the cluster number N according to the shape score, the area and the intersection aggregation degree.

[0163] The first clustering module 44 is used to cluster all the initial intersections according to the number of clusters N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets.

[0164] The cluster score acquisition module 45 is used to acquire cluster scores corresponding to N first cluster sets according to the number of initial intersections included in each first cluster set.

[0165] The cluster quantity updating module 46 is configured to update the cluster quantity according to a preset step size if the cluster score is less than a preset cluster score threshold.

[0166] The second clustering module 47 is used to replace the cluster number N with the updated cluster number, repeatedly execute the first clustering module 44, the cluster score acquisition module 45 and the cluster number update module 46 until the cluster score is greater than or equal to the preset cluster score threshold, and determine the corresponding first cluster set as the second cluster set.

[0167] The sensing device placement module 48 is used to place preset sensing devices according to the second cluster set.

[0168] In a specific implementation, the intersection concentration degree acquisition module 42 further includes: The first data acquisition submodule is used to obtain the geometric shape parameters of the target area, wherein the geometric shape parameters include the perimeter, the area of ​​the area, the rectangular area corresponding to the bounding rectangle, the length of the long circumtangent line and the length of the short circumtangent line, wherein the bounding rectangle refers to a rectangle formed by two long circumtangent lines parallel to the longest axis of the target area and two short circumtangent lines perpendicular to the longest axis.

[0169] The first shape parameter acquisition submodule is used to acquire the first shape parameter corresponding to the target area according to the perimeter and the area.

[0170] The second shape parameter acquisition submodule is used to determine the ratio of the long external tangent line length to the short external tangent line length as the second shape parameter corresponding to the target area.

[0171] The third shape parameter acquisition submodule is used to determine the ratio of the rectangular area to the regional area as the third shape parameter corresponding to the target region.

[0172] The shape score acquisition submodule is used to acquire the shape score of the target area according to the first shape parameter, the second shape parameter and the third shape parameter.

[0173] In a specific implementation, the intersection concentration level corresponding to the target area meets the following conditions: B=M / (Σ i=1 M (A i -A 0 ) 2 ), where B refers to the degree of intersection aggregation corresponding to the target area, M refers to the total number of initial intersections in the target area, and A i It refers to the intersection score corresponding to the i-th initial intersection, i=1, 2, ..., M, A 0 It refers to the average of the intersection scores of the M initial intersections in the target area.

[0174] In a specific implementation, the number of clusters N meets the following conditions: N = ((α1 × e -B )+(α2×((2 / (1+e-C )))-1)+(α3×(S / S 0 )))×N 0 , where α1 refers to the first preset quantity weight, α2 refers to the second preset quantity weight, α3 refers to the third preset quantity weight, S refers to the area, S 0 Refers to the preset reference area, S 0 >S,N 0 refers to the preset reference cluster number, e refers to the natural constant, and C refers to the shape score, where C meets the following conditions: C=β1×D1+β2×D2+β3×D3, where β1 refers to the first preset scoring weight, β2 refers to the second preset scoring weight, β3 refers to the third preset scoring weight, D1 refers to the first shape parameter, D2 refers to the second shape parameter, D3 refers to the third shape parameter, and D1 meets the following conditions: D1=L 2 / (4×π×S), where L refers to the circumference and π refers to the ratio of a circle to a circle.

[0175] In a specific implementation, the intersection concentration degree acquisition module 42 further includes: The second data acquisition submodule is used to obtain the intersection flow corresponding to each initial intersection in the target area, the section type corresponding to each road section and the section weight corresponding to each section type. The section types include expressways, trunk roads, secondary trunk roads and branch roads.

[0176] The first score acquisition submodule is used to acquire a first score corresponding to each initial intersection according to the section types of all sections corresponding to each initial intersection and the section weight corresponding to each section type.

[0177] The second score acquisition submodule is used to acquire a second score corresponding to each initial intersection according to the proportion of the intersection flow corresponding to each initial intersection in the intersection flow corresponding to all initial intersections.

[0178] The intersection score acquisition submodule is used to acquire the intersection score corresponding to each initial intersection according to the first score and the second score corresponding to each initial intersection.

[0179] In a specific implementation, the clustering score meets the following conditions: F=(Σ i=1 N (E j -E 0 ) 2 ) / N, where F refers to the clustering scores corresponding to the N first clustering sets, N refers to the total number of first clustering sets, and E j Refers to the number of initial intersections included in the jth first cluster set, j = 1, 2, ..., N, E0 It refers to the average value of the number of initial intersections included in the N first clustering sets.

[0180] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0181] Embodiment 5 Embodiment 5 of the present invention provides a non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, and the at least one instruction or at least one program is loaded and executed by a processor to implement the steps: S1, obtaining the shape score, area of ​​the target area, intersection score and position coordinates corresponding to each initial intersection in the target area.

[0182] S2, obtaining the intersection aggregation degree corresponding to the target area according to the intersection scores corresponding to all initial intersections.

[0183] S3, obtaining the number of clusters N according to the shape score, the area and the degree of intersection aggregation.

[0184] S4, clustering all the initial intersections according to the number of clusters N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets.

[0185] S5. According to the number of initial intersections included in each first clustering set, clustering scores corresponding to the N first clustering sets are obtained.

[0186] S6: If the cluster score is less than a preset cluster score threshold, the number of clusters N is updated according to a preset step size to obtain an updated number of clusters W.

[0187] S7, using the updated cluster number W to replace the cluster number N, and repeating steps S4-S6 until the cluster score is greater than or equal to a preset cluster score threshold, and determining the corresponding first cluster set as the second cluster set.

[0188] S8, arranging preset sensing devices according to the second cluster set.

[0189] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchronization Link) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0190] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0191] Embodiment 6 A sixth embodiment of the present invention provides an electronic device, which includes a processor and the non-transitory computer-readable storage medium in the fifth embodiment of the present invention.

[0192] The above are only preferred embodiments of the present invention, and are not intended to limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any technician familiar with the profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A clustering-based sensing device deployment method, characterized in that: The clustering-based sensing device deployment method comprises the following steps: S1, obtaining a shape score, an area of ​​a target area, an intersection score and a position coordinate corresponding to each initial intersection in the target area; S2, obtaining the intersection concentration level corresponding to the target area according to the intersection scores corresponding to all initial intersections; S3, obtaining the number of clusters N according to the shape score, the area of ​​the region and the degree of intersection aggregation; S4, clustering all the initial intersections according to the cluster number N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets; S5, acquiring clustering scores corresponding to the N first clustering sets according to the number of initial intersections included in each first clustering set; S6, if the cluster score is less than a preset cluster score threshold, updating the number of clusters N according to a preset step size, and obtaining an updated number of clusters W; S7, using the updated cluster number W to replace the cluster number N, and repeating steps S4-S6 until the cluster score is greater than or equal to the preset cluster score threshold, and determining the corresponding first cluster set as the second cluster set; S8: Arrange preset sensing devices according to the second cluster set.

2. The clustering-based sensing device deployment method according to claim 1, characterized in that: S1 also includes the following steps: S11, obtaining geometric shape parameters of the target area, wherein the geometric shape parameters include perimeter, area, rectangular area corresponding to the bounding rectangle, long circumferential line length, and short circumferential line length, wherein the bounding rectangle refers to a rectangle formed by two long circumferential lines parallel to the longest axis of the target area and two short circumferential lines perpendicular to the longest axis; S12, acquiring a first shape parameter corresponding to the target region according to the perimeter and the region area; S13, determining the ratio of the length of the long external tangent line to the length of the short external tangent line as a second shape parameter corresponding to the target area; S14, determining the ratio of the area of ​​the rectangle to the area of ​​the region as a third shape parameter corresponding to the target region; S15: Obtain a shape score of the target area according to the first shape parameter, the second shape parameter, and the third shape parameter.

3. The clustering-based sensing device deployment method according to claim 2, characterized in that: The intersection concentration level corresponding to the target area meets the following conditions: B=M / (Σ i=1 M (A i -A 0 ) 2 ), where B refers to the degree of intersection concentration corresponding to the target area, M refers to the total number of initial intersections in the target area, and A i It refers to the intersection score corresponding to the i-th initial intersection, i=1, 2, ..., M, A 0 It refers to the average of the intersection scores of the M initial intersections in the target area.

4. The clustering-based sensing device deployment method according to claim 3, characterized in that: The number of clusters N meets the following conditions: N = ((α1 × e -B )+(α2×((2 / (1+e -C )))-1)+(α3×(S / S 0 )))×N 0 , where α1 refers to the first preset quantity weight, α2 refers to the second preset quantity weight, α3 refers to the third preset quantity weight, S refers to the area of ​​the region, S 0 Refers to the preset reference area, S 0 >S,N 0 refers to the preset reference cluster number, e refers to the natural constant, C refers to the shape score, and C meets the following conditions: C=β1×D1+β2×D2+β3×D3, where β1 refers to the first preset scoring weight, β2 refers to the second preset scoring weight, β3 refers to the third preset scoring weight, D1 refers to the first shape parameter, D2 refers to the second shape parameter, D3 refers to the third shape parameter, and D1 meets the following conditions: D1=L 2 / (4×π×S), wherein L refers to the circumference and π refers to the pi.

5. The clustering-based sensing device deployment method according to claim 1, characterized in that: S1 also includes the following steps: S16, obtaining the intersection flow corresponding to each initial intersection in the target area, the section type corresponding to each road section, and the section weight corresponding to each road section type, wherein the road section types include expressways, trunk roads, secondary trunk roads, and branch roads; S17, obtaining a first score corresponding to each initial intersection according to the section types of all sections corresponding to each initial intersection and the section weight corresponding to each section type; S18, obtaining a second score corresponding to each initial intersection according to the proportion of the intersection flow corresponding to each initial intersection in the intersection flow corresponding to all initial intersections; S19, obtaining an intersection score corresponding to each initial intersection according to the first score and the second score corresponding to each initial intersection.

6. The clustering-based sensing device deployment method according to claim 1, characterized in that: The clustering score meets the following conditions: F=(Σ i=1 N (E j -E 0 ) 2 ) / N, where F refers to the clustering scores corresponding to the N first clustering sets, N refers to the total number of first clustering sets, and E j Refers to the number of initial intersections included in the jth first cluster set, j = 1, 2, ..., N, E 0 It refers to the average value of the number of initial intersections included in the N first clustering sets.

7. A clustering-based sensing device deployment device, characterized in that: The clustering-based sensing device deployment device includes: A data acquisition module, used to acquire the shape score, area of ​​the target area, and the intersection score and position coordinates corresponding to each initial intersection in the target area; A road intersection concentration degree acquisition module is used to acquire the road intersection concentration degree corresponding to the target area according to the road intersection scores corresponding to all initial road intersections; A cluster number acquisition module, used to acquire the cluster number N according to the shape score, the area of ​​the region and the degree of intersection aggregation; A first clustering module, used for clustering all the initial intersections according to the cluster number N and the position coordinates corresponding to all the initial intersections to obtain N first cluster sets; A clustering score acquisition module, used to acquire clustering scores corresponding to the N first clustering sets according to the number of initial intersections included in each first clustering set; A cluster number updating module, configured to update the cluster number N according to a preset step size if the cluster score is less than a preset cluster score threshold, and obtain an updated cluster number W; a second clustering module, configured to replace the cluster number N with the updated cluster number W, repeatedly execute the first clustering module, the cluster score acquisition module and the cluster number update module until the cluster score is greater than or equal to the preset cluster score threshold, and determine the corresponding first cluster set as the second cluster set; A sensing device deployment module is used to deploy preset sensing devices according to the second cluster set.

8. The clustering-based sensing device deployment device according to claim 7, characterized in that: The intersection aggregation degree acquisition module also includes: A first data acquisition submodule is used to acquire geometric shape parameters of the target area, wherein the geometric shape parameters include perimeter, area, rectangular area corresponding to the boundary rectangle, long external tangent length and short external tangent length; A first shape parameter acquisition submodule, used to acquire a first shape parameter corresponding to the target area according to the perimeter and the area of ​​the area; A second shape parameter acquisition submodule, used for determining the ratio of the length of the long external tangent line to the length of the short external tangent line as a second shape parameter corresponding to the target area; A third shape parameter acquisition submodule, used to determine the ratio of the rectangular area to the area of ​​the region as a third shape parameter corresponding to the target region; The shape score acquisition submodule is used to acquire the shape score of the target area according to the first shape parameter, the second shape parameter and the third shape parameter.

9. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the clustering-based perception device deployment method as described in any one of claims 1-6.

10. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 9.

Images