A method and system for fast road plan routing based on unsupervised learning

Abstract

The present application provides a method and system for fast road plane routing based on unsupervised learning, which relates to the field of computer technology. The method includes: obtaining basic parameters of road plane routing, calculating and planning road routing areas and at least one expected line; in the road routing area, the mouse picks up at least one random point, which is grouped into the initial point set picked up by the mouse, and the random point is close to the expected line; the initial point set is densely processed to obtain a dense point set; the curvature characteristic value of the dense point set is extracted to obtain statistical information; the statistical information is dimensionlessly normalized to obtain processed statistical information; the statistical information is clustered based on unsupervised learning to obtain clustering processing results; the clustering processing results are processed according to straight lines, transition curves and circular curves respectively to obtain initial plane design parameters; the present application solves the problem of low intelligentization of routing and high requirements for users in related technologies.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and system for fast road plane routing based on unsupervised learning. Background Art

[0002] At present, there are many commonly used wiring methods in road plan design, especially in the highway industry, such as: "intersection method", "building block method", "parameter method" and so on.

[0003] The intersection method is that the user inputs the coordinates of the intersection of the line curve and the length, radius, mileage and other information of the corresponding line's transition curve, and then obtains the coordinates of the feature points, stake points, and line points, as well as an intuitive graphic display, so that the line layout and other measurement work can be easily carried out.

[0004] Because the concept of the intersection method is clear, easy to grasp, and the calculation is standardized, the intersection method is currently the most commonly used road design method.

[0005] However, in terms of coordination with the environment, the intersection method requires multiple attempts at data input and graphic generation due to its calculation principle, which makes the design process cumbersome. At present, the intelligent level of road plan routing based on the intersection method is low and the requirements for users are high.

[0006] From the above, we can see that the problems of intelligent wiring and insufficient wiring convenience still need to be solved. Summary of the invention

[0007] The present application provides a method, system and storage medium for fast road plan routing based on unsupervised learning, which can solve the problem of low accuracy of information recommendation in related technologies. The technical solution is as follows:

[0008] According to one aspect of the present application, a method for fast road plane routing based on unsupervised learning includes: obtaining basic parameters of road plane routing, calculating and planning a road routing area and at least one expected line;

[0009] In the road wiring area, the mouse picks up at least one random point, which is collected as the initial point set picked up by the mouse, and the random point is close to the expected line; the initial point set is densely processed to obtain a dense point set; at least one curvature eigenvalue corresponding to the dense point set is extracted to obtain statistical information of the curvature eigenvalue of the dense point set; the statistical information of the curvature eigenvalue includes the first-order central moment, the second-order central moment, the skewness, the kurtosis, the upper and lower tertiles, the linear regression coefficient and the entropy; the statistical information is dimensionlessly normalized to obtain the processed statistical information; the statistical information is clustered based on unsupervised learning to obtain the clustering processing results, and three cluster divisions are output; the road module is divided according to the intersection method, and the road module includes a straight line module, a transition curve module and a circular curve module; the first-order central moment in each cluster is calculated, and sorted from small to large, and the straight line module, the transition curve module and the circular curve module are corresponding one by one in sequence, and the initial plane design parameters are extracted, and the initial plane design parameters include the design parameters of at least one intersection and the corresponding road module;

[0010] For the initial plane design parameters, determine whether the design parameters of the road module corresponding to each intersection point meet the specifications; if so, include them in the optimized plane design parameters; if not, optimize and update the design parameters of the intersection point and the corresponding road module, and include them in the optimized plane design parameters; based on the optimized plane design parameters, perform road plane rapid routing;

[0011] According to one aspect of the present application, a system for rapid road plane routing based on unsupervised learning includes a road routing basic parameter acquisition unit for acquiring basic parameters of road plane routing; a planning unit for planning road routing areas and expected lines; an input unit including a mouse for information input; a preprocessing unit for intensive data processing and feature value extraction; a computing control center for clustering and calculation of plane design parameters; a display unit for displaying data and routing operations, and a storage unit for storing data.

[0012] According to one aspect of the present application, a storage medium stores computer-readable instructions thereon, and the computer-readable instructions are executed by one or more processors to implement the method for fast road plane routing based on unsupervised learning as described above.

[0013] According to one aspect of the present application, a computer program product includes computer-readable instructions, which are stored in a storage medium. One or more processors of an electronic device read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, so that the electronic device implements the method for rapid road plane routing based on unsupervised learning as described above.

[0014] The beneficial effects of the technical solution provided by this application are:

[0015] In the above technical scheme, the present application performs intensive processing through a small number of random points, and then extracts statistical information containing a high-latitude data point set for the curvature eigenvalues, performs unsupervised learning clustering, obtains the road module corresponding to the intersection method of the clustering results, and then obtains the road design parameters, and then standardizes and optimizes the road design parameters to obtain the optimized plane design parameters, and performs rapid road plane wiring; thereby effectively solving the problems of low wiring intelligence and high user requirements existing in related technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can obtain other drawings based on these drawings without creative work.

[0017] Figure 1 It is a structural block diagram of the system involved in this application;

[0018] Figure 2 It is a schematic diagram of the road wiring plan involved in this application;

[0019] Figure 3 It is a schematic diagram of the division of road modules involved in this application;

[0020] Figure 4 is a flow chart of the method involved in this application;

[0021] Figure 5 is a flow chart of S1 according to the method involved in this application;

[0022] Figure 6 is a flow chart of S2 according to the method involved in this application;

[0023] Figure 7 is a flow chart of S3 according to the method involved in this application;

[0024] Figure 8 is a flow chart of S4 according to the method involved in this application;

[0025] Fig. 9 is a flow chart of S5 to S6 according to the method involved in this application;

[0026] Fig.10 is a flow chart of S622 according to the method involved in this application;

[0027] Fig.11It is a flow chart of data processing involved in this application.

[0028] Fig.12 It is a characteristic map of high-latitude data based on curvature eigenvalues ​​involved in this application;

[0029] Fig.13 It is a schematic diagram of the wiring application involved in this application. DETAILED DESCRIPTION

[0030] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present application.

[0031] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present disclosure refers to the presence of the features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0032] With the popularization of computers and the development of CAD, computer-aided road plan design is increasingly used in road design. With the economic development of our country, higher requirements have been put forward for the development of road transportation. The large-scale construction of high-grade highways such as expressways, first-class highways, and special roads for automobiles has challenged the traditional plan design methods.

[0033] Plane alignment design is a key step in road alignment design. The road plane alignment is composed of three parts: straight lines, circular curves, and transition curves, which are the "three elements of plane alignment". Plane alignment design is the design of various combinations of the three elements of plane alignment.

[0034] There are many plane design methods, and there are many commonly used wiring methods in the highway industry, such as: "intersection method", "building block method", "parametric method" and so on.

[0035] The intersection method is that the user inputs the coordinates of the intersection of the line curve and the length, radius, mileage and other information of the corresponding line's transition curve, and then obtains the coordinates of the feature points, stake points, and line points, as well as an intuitive graphic display, so that the line layout and other measurement work can be easily carried out.

[0036] Because the concept of the intersection method is clear, easy to grasp, and the calculation is standardized, the intersection method is currently the most commonly used road design method.

[0037] However, due to the calculation principle of the intersection method, the intersection method cannot be coordinated with the environment and requires multiple "attempts" of data input and graphic generation to complete, which makes the design process cumbersome. Currently, the road plan routing based on the intersection method has a low degree of intelligence and high requirements for users.

[0038] Therefore, it is urgent to study the scheme of rapid road plan layout;

[0039] See also Figure 1-Figure 13 The embodiment of the present application provides a system 100 for fast routing of road planes based on unsupervised learning, comprising:

[0040] The road wiring basic parameter acquisition unit 101 is used to acquire the road plane wiring basic parameters;

[0041] A planning unit 102 for planning road routing areas and expected lines;

[0042] Input unit 105, including a mouse 1051, for information input;

[0043] A preprocessing unit 106, used for intensive data processing and feature value extraction;

[0044] A calculation control center 103 is used for clustering and calculation of plane design parameters;

[0045] Display unit 104, used for displaying data and wiring operations;

[0046] The storage unit 107 is used for storing data.

[0047] In an exemplary embodiment, the road wiring basic parameter acquisition unit 101 is an image acquisition unit, which is used to acquire satellite images of roads to be wired and design parameters; the planning unit 102 plans the boundary control points of the road wiring area based on the satellite images of the roads to be wired and the design parameters, and constructs the road wiring area and the expected line according to the boundary control points of the road wiring area;

[0048] In an exemplary embodiment, the input unit 105 further includes a keyboard; it may also include a microphone for voice input;

[0049] The present application embodiment provides a method for fast routing of road planes based on unsupervised learning, including:

[0050] S1. Obtain basic parameters of road plan layout, calculate and plan the road layout area and at least one expected line;

[0051] S2. In the road wiring area, the mouse picks up at least one random point, which is aggregated into an initial point set picked up by the mouse, and the random point is close to the expected line;

[0052] S3, performing dense processing on the initial point set to obtain a dense point set;

[0053] S4, extracting at least one curvature eigenvalue corresponding to the dense point set, and obtaining statistical information of the curvature eigenvalue of the dense point set; the statistical information of the curvature eigenvalue includes the first-order central moment , second-order central moment , skewness , Kurtosis , upper and lower tertiles , linear regression coefficient and Entropy ; Perform dimensionless normalization processing on the statistical information to obtain processed statistical information;

[0054] S5. Perform clustering processing on statistical information based on unsupervised learning, obtain clustering processing results, and output three cluster divisions; divide the road modules according to the intersection method, and the road modules include straight line modules, transition curve modules, and circular curve modules; calculate the first-order central moment in each cluster. , and sort them from small to large, corresponding to the straight line module, the transition curve module and the circular curve module one by one in sequence, and extracting the initial plane design parameters, the initial plane design parameters including at least one intersection point and the design parameters of the corresponding road module;

[0055] S6. For the initial plane design parameters, determine whether the design parameters of the road module corresponding to each intersection point meet the specifications; if so, include them in the optimized plane design parameters; if not, optimize and update the design parameters of the intersection point and the corresponding road module, and include them in the optimized plane design parameters;

[0056] S7. Based on the optimized plane design parameters, the road plane is quickly routed.

[0057] In an exemplary embodiment, the method of S1, obtaining basic parameters of road plane wiring, calculating and planning road wiring areas and at least one expected line includes: S11, obtaining basic parameters of road plane wiring, calculating and determining the boundary control point zc of the road wiring area; S12, constructing the road wiring area S according to the boundary control point zc of the road wiring area B; Based on the road routing area S B , in the road wiring area S B Plan at least one expected line L along the inner road plane layout direction Y The expected line L Y The curvature changes little;

[0058] In an exemplary embodiment, S2, in the planned road routing area S B The mouse picks up at least one random point, which is collected as an initial point set picked up by the mouse. The method for the random point to approach the expected line includes: S21, constructing a picking expected control line on both sides of the expected line and , pick the expected control line and the vertical distance between the expected line is D, D is related to the road width; S22, the mouse is between the two expected control lines and Randomly pick at least one random point PO q , which is collected as the initial point set picked by the mouse.

[0059] In an exemplary embodiment, S3, performing dense processing on the initial point set to obtain a dense point set includes:

[0060] S31, interpolate the initial point set to obtain at least one dense point PO c ; S32, at least one dense point PO c The sequential connection lines with the initial point set form a dense connection line L M ; Calculate dense connections L M Curvature; S33, determine the dense connection line L M Is the curvature of continuous? If so, S341, then the dense point PO c If not, S342 discard the dense point PO c .

[0061] In an exemplary embodiment, S4, the method of extracting the curvature feature value of the dense point set and obtaining statistical information includes: S41, obtaining a dense line y with continuous curvature ;

[0062] S42: extracting curvature feature values ​​based on the curvature continuous dense lines.

[0063] ; where k is the curvature eigenvalue, and is the first-order differential of f with respect to x or y; , is the second-order differential of f with respect to x or y;

[0064] In an exemplary embodiment, the method for statistical information processing includes:

[0065] Extract a dense point PO c The curvature eigenvalue k i , i is its data position; get data k i The 7 data before and after are grouped as data points , ; Based on the dense point set, extract the data point set K= ; is the data set corresponding to the i-th dense point;

[0066] Extracting a subset of raw statistics The corresponding first-order central moment , second-order central moment , skewness , Kurtosis , upper and lower thirds , linear regression coefficient and Entropy , for the original statistical data subset Perform normalized dimensionless processing to make the data dimensionless and between [0,1], obtain the processed statistical information, and group it into the first High-latitude data points with dense points ;

[0067] After sorting and collecting, we get a dense high-latitude data point set T= , For the A high-dimensional data point set with dense points, n is the number of dense points.

[0068] In an exemplary embodiment, clustering is performed on statistical information based on unsupervised learning, and a method for obtaining a clustering result includes:

[0069] Step 1: Randomly select a point from the dense point high-latitude data point set T as the cluster center ;

[0070] Step 2: For data points , calculate the distance to the nearest cluster center to the selected cluster center:

[0071]

[0072] Step 3: Use roulette wheel to select a new cluster center from the data. The selection principle is: The larger it is, the greater the probability of it being selected as the next cluster center;

[0073] Step 4: Repeat step 2 until 3 cluster centers are selected ; They are the first to third cluster centers respectively;

[0074] Step 5: Select As the initial cluster center as the initial cluster ; Initialize clusters for the first to third time respectively;

[0075] Step 6: Calculate each Distance from the three cluster centers: , For the said indivual The jth distance of Mark the smallest The corresponding category is ,renew ;

[0076] Step 7: For Recalculate new cluster centers using samples from : ;

[0077] Step 8: Repeat steps 6 to 7 until the cluster center does not change;

[0078] Step 9: Output three cluster divisions. ;

[0079] Step 10: Divide the road modules according to the intersection method. The road modules include straight line modules, transition curve modules and circular curve modules. Calculate the first-order central moment of each cluster. , and sort them from small to large, corresponding to the straight line module, the transition curve module and the circular curve module one by one in sequence, and extracting the initial plane design parameters, the initial plane design parameters including at least one intersection point and the design parameters of the corresponding road module;

[0080] Initial graphic design parameters:

[0081] ;

[0082] in JD i coordinate; is the corresponding intersection point JD i The road design parameters include the length of the first side transition curve, the radius of the circular curve and the length of the second side transition curve;

[0083] In an exemplary embodiment, the S6 includes the following steps: S61, for the initial plane design parameters, judging whether the design parameters of the road module corresponding to each intersection point meet the specification; if so, S621, classifying the plane design parameters after optimization; if not, S622, the method for optimizing and updating the design parameters of the intersection point and the corresponding road module includes:

[0084] S6221, calculating the radius of the circular curve module of the intersection according to the limit value, and calculating the length of the transition curve; S6222, updating the plane design parameters, and including them in the optimized plane design parameters;

[0085] In an exemplary embodiment: the road module is divided according to the intersection method, and the road module includes a straight line module and , Transition Curve Module and and circular curve module :

[0086] The clustering results correspond to two straight line modules and Calculation of

[0087] Based on the obtained pairwise straight line parameters, the intersection coordinates can be obtained: ;

[0088] In an exemplary embodiment, the general analytical expression of a straight line may be used. According to the least squares method, Minimum, ( ) is the point coordinate of the linear module; according to ; Calculate and sort out The coefficients are as follows: It should be noted that this part involves The equal coefficients and symbols are only the mathematical coefficients and symbols in the geometry of the line, and are not to be confused with other parts;

[0089] Use general analytical expressions to fit the coordinates of intersection points and other straight line points;

[0090] Circular curve module :Using the general analytical formula, Due to the small number of measurement points in the circular curve part, Taubin fitting is used to ensure the stability and accuracy of the fitting, where: ; Here ( ) is the point coordinate on the circular curve module; It should be noted that this part involves The characters are only mathematical coefficients and calculation symbols and should not be confused with other parts;

[0091] ;

[0092] ;

[0093] ;

[0094] ;

[0095] ;

[0096] Combining the above formula, we can get the coefficient ; and then calculate:

[0097] Center coordinates

[0098] radius ;

[0099] It should be noted that the error weight of the original point picked up by the mouse is set to the error of the densely processed point. times, of which to be processed.

[0100] Transition Curve Module and :

[0101] The transition curve is processed as a clothoid, and the length of the transition curve is iteratively approximated; the clothoid is expanded, and the coordinate formula is arranged as follows:

[0102] ;

[0103] The remainder is:

[0104] ; ;in is a positive integer, is the length of the clothoid, is the clothoid angle;

[0105] The above-mentioned coordinate formula can guarantee accuracy. It should be noted that the X, Y, The characters are just mathematical coefficients and calculation symbols and should not be confused with other parts.

[0106] Known circle radius And the distance d from the center of the circle to the straight line is given by approximation:

[0107] Available The value is roughly Between, take , then:

[0108]

[0109] Depend on The distance from the center of the corresponding radius of curvature to the straight line can be obtained ,if (known precision), the iteration is terminated, otherwise if ,Pick , otherwise , repeat the above process. By: , you can get the corresponding transition curve length .

[0110] In summary, the intersection method is used to divide the road modules, and the three clusters output in the clustering results are divided into one-to-one corresponding straight line modules, transition curve modules and circular curve modules to calculate: Extract the initial plane design parameters:

[0111]

[0112] in: It is the intersection point JD i coordinate, i is the circular curve radius, are the lengths of the transition curves on both sides respectively;

[0113] The method for determining whether the design parameters of the intersection and the corresponding road module meet the specification; if not, optimizing and updating the design parameters of the intersection and the corresponding road module includes:

[0114] The radius of the circular curve module of the intersection is calculated according to the limit value, and the length of the transition curve is calculated and included in the optimized plane design parameters;

[0115] It is found that the specification of the transition curve [circle] in the road plane design specification is:

[0116] Table 1 Minimum length of the clothoid

[0117]

[0118] The radius of the circular curve is:

[0119] Table 2 Minimum radius of circular curve

[0120]

[0121] Based on the designed road speed, the corresponding design parameters are matched.

[0122] The above road plane parameters are calculated based on the point set clicked by the mouse after the discrete point set is densified based on the requirement of plane curvature continuity. However, the parameters of the generated road are uncertain. The point clicked by the mouse should be placed on the plane line position that meets the road specifications as much as possible:

[0123] Specifically, if it does not fall on a plane line position that meets road specifications, the acceptable circular curve radius is calculated based on the initial plane design parameters. Value range: ; If the circular curve radius is acceptable If the general value in the road design specification is not empty, the calculation is performed according to the general value of the road design. If the general value is empty, the calculation is performed according to the limit value. If the limit value is empty, that is: ,but For non-empty By binary search, calculate the length of the corresponding transition curve ; If it meets the specification, it is accepted and the distance between the original mouse point in the intersection interval and the nearest point of the plane line generated in the interval is recorded at the same time:

[0124] ;

[0125] in is the distance between the nearest points of the plane line in the interval, is the distance calculation function, For the The square of the distance between points, I is the interval.

[0126] Will The smallest value Accepts the optimized parameters.

[0127] At this time, fix the intersection coordinates and traverse the various design parameters. or , , the optimized road design parameters can be obtained , which facilitates rapid routing based on optimized road design parameters.

[0128] In addition, a storage medium is provided in an embodiment of the present application, on which computer-readable instructions are stored, and the computer-readable instructions are executed by one or more processors to implement the method for rapid road plane routing based on unsupervised learning as described above.

[0129] In an embodiment of the present application, a computer program product is provided, which includes computer-readable instructions. The computer-readable instructions are stored in a storage medium. One or more processors of an electronic device read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, so that the electronic device implements the method for rapid road plane routing based on unsupervised learning as described above.

[0130] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0131] The above is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for fast road plan routing based on unsupervised learning, characterized in that: include: Obtaining basic parameters of road plan layout, calculating and planning the road layout area and at least one expected line; In the road routing area, the mouse picks up at least one random point, which is grouped as an initial point set picked up by the mouse, and the random point is close to the expected line; Performing dense processing on the initial point set to obtain a dense point set; Extract at least one curvature eigenvalue corresponding to the dense point set to obtain statistical information of the curvature eigenvalue of the dense point set; the statistical information of the curvature eigenvalue includes first-order central moment, second-order central moment, skewness, kurtosis, upper and lower tertiles, linear regression coefficient and entropy; Performing dimensionless normalization processing on the statistical information to obtain processed statistical information; Based on unsupervised learning, statistical information is clustered to obtain clustering results, and three cluster divisions are output; road modules are divided according to the intersection method, and the road modules include straight line modules, transition curve modules, and circular curve modules; the first-order central moment in each cluster is calculated and sorted from small to large, and the straight line modules, transition curve modules, and circular curve modules are corresponding one by one in sequence, and initial plane design parameters are extracted, and the initial plane design parameters include at least one intersection and the design parameters of the corresponding road module; For the initial plane design parameters, determine whether the design parameters of the road module corresponding to each intersection point meet the specifications; if so, include them in the optimized plane design parameters; if not, optimize and update the design parameters of the intersection point and the corresponding road module, and include them in the optimized plane design parameters; Based on the optimized plane design parameters, the road plane is quickly routed; Methods for clustering statistical information based on unsupervised learning and obtaining clustering results include: Step 1: Randomly select a point from the dense point high-latitude data point set T as the cluster center u r ; Step 2: For the high-latitude data point set of the s-th dense point, calculate the distance to the nearest cluster center of the selected cluster center: Where, dis(k) is the calculated distance; is the calculation function of cluster center; Step 3: Use roulette wheel to select a new cluster center from the data. The selection principle is: the larger the newly calculated dis(k), the greater the probability of being selected as the next cluster center; Step 4: Repeat step 2 until 3 cluster centers are selected They are the first to third cluster centers respectively; Step 5: Select u r As the initial cluster center, the initialization cluster C = [C1, C2, C3]; C1, C2, C3 are the first to third initialization clusters respectively; Step 6: Calculate each T t Distance from the three cluster centers: d ij is the sth dense point T t The square of the Euclidean distance between the jth cluster center and T t Mark the smallest d ij The corresponding category is C j , update C j =C j ∪T t ; Step 7: For C j Recalculate new cluster centers using samples from : Among them C q Initialize clustering for the qth cluster; Step 8: Repeat steps 6 to 7 until the cluster center does not change; Step 9: Output three cluster partitions E, E = [E1, E2, E3]; Step 10: Divide the road modules according to the intersection method. The road modules include straight line modules, transition curve modules and circular curve modules. Calculate the first-order central moment of each cluster. and sorting them from small to large, corresponding to the straight line module, the transition curve module and the circular curve module one by one, extracting the initial plane design parameters, the initial plane design parameters including at least one intersection point and the design parameters of the corresponding road module; Initial graphic design parameters: Where [X i ,Y i ] is the intersection point JD i Coordinates; pi ,R i ,l si ] is the corresponding intersection point JD i The road design parameters include the length of the first side transition curve, the radius of the circular curve and the length of the second side transition curve.

2. The method for fast road plan layout based on unsupervised learning as claimed in claim 1, characterized in that: The method of obtaining basic parameters of road planar wiring, calculating and planning road wiring areas and at least one expected line includes: Obtaining basic parameters of road planar wiring, calculating and determining boundary control points of a road wiring area, and constructing a road wiring area according to the boundary control points of the road wiring area; Based on the road routing area, at least one expected line is planned in the road routing area along the road plane routing direction; the expected line has a curvature with a small change.

3. The method for fast road plan routing based on unsupervised learning as claimed in claim 2, characterized in that: In the road wiring area, the mouse picks up at least one random point, which is collected as an initial point set picked up by the mouse. The method for the random point to approach the expected line includes: constructing a pick-up expected control line on both sides of the expected line, the vertical distance between the pick-up expected control line and the expected line is D, and D is related to the road width. The mouse randomly picks up at least one random point between the two pick-up expected control lines, which are collected as an initial point set picked up by the mouse, and the initial point set is S m =[S1,S2,...,S m ]; S m =[x m ,y m ] is the plane coordinate of the random point m.

4. The method for fast road plan routing based on unsupervised learning as claimed in claim 3, characterized in that: The method of performing dense processing on the initial point set to obtain a dense point set includes: The initial point set is interpolated to obtain at least one dense point, and the at least one dense point and the initial point set are sequentially connected to form a dense line; the curvature of the dense line is calculated to determine whether the curvature of the dense line is continuous, and if it is continuous, the dense point is included in the dense point set; otherwise, the dense point is discarded, and the initial point set is saved after interpolation processing; the curvature eigenvalue corresponding to the dense point set is extracted to obtain statistical information of the curvature eigenvalue of the dense point set.

5. The method for fast road plan routing based on unsupervised learning as claimed in claim 4, characterized in that: The method of extracting the curvature characteristic value of the dense point set and obtaining statistical information includes: Get dense lines with continuous curvature; Curvature feature value extraction is performed based on the curvature continuous dense lines.

6. The method for fast road plan layout based on unsupervised learning as claimed in claim 5, characterized in that: The statistical information processing method comprises: Extract the curvature eigenvalue k of a dense point t , t is the data position of the dense point; get data x t The 7 data before and after are grouped into data set K t , K t =[k t-7 ,k t-6 ,…k t ,k t+1 ,…k t+7 ]; Based on the dense point set, extract the data point set K = [K1, K2, ..., K t ,...,K n ]; K t is the data set corresponding to the tth dense point; Extract the original statistics subset K t The corresponding first-order central moment Second-order central moment Var[k] t , skewness Kurtosis Upper and lower tertiles ΔQ t , linear regression coefficient W t and entropy H t , for the original statistical data subset K t Perform normalized dimensionless processing to make the data dimensionless and between [0,1], obtain the processed statistical information, and group it into the high-dimensional data point set of the tth dense point The dense point high-latitude data point set T = [T1, T2, ..., T t ,...,T n ], T t is the high-latitude data point set of the tth dense point, and n is the number of dense points.

7. The method for fast road plan layout based on unsupervised learning as claimed in claim 1, characterized in that: The method for determining whether the design parameters of the intersection and the corresponding road module meet the specification; if not, optimizing and updating the design parameters of the intersection and the corresponding road module includes: The radius of the circular curve module of the intersection is calculated according to the limit value, and the length of the transition curve is calculated and included in the optimized plane design parameters.

8. A system for fast road plane routing based on unsupervised learning, the system being used to execute the method in claims 1 to 7, characterized in that: include: A road wiring basic parameter acquisition unit is used to acquire basic parameters of road plane wiring; Planning unit for planning road routing areas and prospective lines; Input unit, including a mouse, for information input; Preprocessing unit, used for intensive data processing and feature value extraction; Computational control center, used for clustering and calculation of plane design parameters; Display unit, used for displaying data and wiring operations; Storage unit, used for storing data.

9. A storage medium having computer-readable instructions stored thereon, characterized in that: The computer-readable instructions are executed by one or more processors to implement the method for fast road plane routing based on unsupervised learning as described in any one of claims 1 to 7.

Images