Road construction area identification method and system

Abstract

The embodiment of the invention provides a road construction area identification method and system. The method comprises the following steps: determining a plurality of candidate objects based on road data; forming a plurality of three-point sets based on the plurality of candidate objects; constructing a directed graph based on the plurality of three-point sets; determining at least one directed path based on the directed graph, and determining an optimal path from the at least one directed path; and identifying the road construction area based on the optimal path.

Description

Technical Field

[0001] This specification relates to the field of environmental perception, and in particular to a road construction area identification method and system. Background Art

[0002] In the current field of environmental perception, it is difficult to identify the location and size of road construction areas, which may make it difficult for autonomous vehicles (e.g., autonomous cars) to travel as needed. Existing road construction area recognition technologies require a lot of prior information, consume a lot of computing power, manpower and time costs, and the accuracy of the recognition results is also not ideal.

[0003] Based on this, it is desired to propose a road construction area identification method and system. Summary of the invention

[0004] One or more embodiments of the present specification provide a method for identifying a road construction area. The method comprises: determining multiple candidate objects based on road data; forming multiple three-point sets based on the multiple candidate objects; constructing a directed graph based on the multiple three-point sets; determining at least one directed path based on the directed graph, and determining the best path from the at least one directed path; and identifying a road construction area based on the best path.

[0005] One or more embodiments of the present specification provide a road construction area recognition system. The road construction area recognition system includes: an object determination module, used to determine multiple candidate objects based on road data; a three-point set determination module, used to form multiple three-point sets based on multiple candidate objects; a graph construction module, used to construct a directed graph based on multiple three-point sets; a path determination module, used to determine at least one directed path based on the directed graph, and determine the best path from the at least one directed path; and an area determination module, used to identify the road construction area based on the best path.

[0006] One or more embodiments of the present specification provide a road construction area recognition device, including a processor, wherein the processor is used to execute a road construction area recognition method.

[0007] One or more embodiments of the present specification provide a computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a road construction area identification method. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0009] Figure 1 is a schematic diagram of an application scenario of a road construction area identification system according to some embodiments of this specification;

[0010] Figure 2 is an exemplary flow chart of a method for identifying a road construction area according to some embodiments of this specification;

[0011] Figure 3 is an exemplary flow chart of a method for determining a road construction area according to some embodiments of this specification;

[0012] Figure 4 is a schematic diagram of location feature points of candidate objects in road data according to some embodiments of this specification;

[0013] Figure 5 is a schematic diagram of a directed graph construction process according to some embodiments of this specification;

[0014] Fig. 6A is a schematic diagram of an initial area according to some embodiments of the present specification;

[0015] Figure 6B is a schematic diagram of a method for determining a road construction area according to some embodiments of this specification;

[0016] Figure 7 It is a module diagram of a road construction area identification system according to some embodiments of this specification. DETAILED DESCRIPTION

[0017] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0018] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0019] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0020] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0021] Figure 1 is a schematic diagram of an application scenario of a road construction area identification system according to some embodiments of this specification. Figure 1 As shown, an application scenario 100 of a road construction area recognition system (hereinafter referred to as application scenario 100 ) may include a processing device 110 , a network 120 , a storage device 130 , a data acquisition device 140 and a road 150 .

[0022] The processing device 110 refers to a device for processing data and / or information from at least one component of the application scenario 100 or an external data source (e.g., a cloud data center). The processing device 110 can access data or information from the storage device 130 and / or the data acquisition device 140 through the network 120. The processing device 110 can also directly connect to the storage device 130 and / or the data acquisition device 140 to access information and / or data. For example, the processing device 110 can obtain image data of the road 150 (e.g., a road surface image of the road 150) from the data acquisition device 140 and / or the storage device 130. The processing device 110 can process the acquired data and / or information. For example, the processing device 110 can determine multiple candidate objects based on the road data; form multiple three-point sets based on the multiple candidate objects; construct a directed graph based on the multiple three-point sets; determine at least one directed path based on the directed graph, and determine the best path from the at least one directed path; and identify the road construction area based on the best path. In some embodiments, the processing device 110 can be a single server or a server group. The processing device 110 can be local or remote. For more information about candidate objects, point sets, directed graphs, directed paths, and optimal paths, see Figure 2 and its related description.

[0023] The network 120 may connect the various components of the application scenario 100 and / or connect the application scenario 100 with external resources. In some embodiments, one or more components of the application scenario 100 (e.g., the processing device 110, the storage device 130, and / or the data acquisition device 140) may exchange information and / or data through the network 120.

[0024] In some embodiments, the network 120 may be any one or more of a wired network or a wireless network. In some embodiments, the network 120 may include one or more network access points. For example, the network 120 may include a wired or wireless network access point (e.g., a base station and / or a network switching point), through which one or more components of the application scenario 100 may be connected to the network 120 to exchange data and / or information.

[0025] The storage device 130 refers to a device for storing data, instructions and / or any other information. In some embodiments, the storage device 130 may store data and / or information obtained from the processing device 110, the data acquisition device 140, etc. For example, the storage device 130 may store the recognition result of the road construction area of ​​the road 150 obtained from the processing device 110. In some embodiments, the storage device 130 may include a large-capacity memory, a removable memory, etc., or any combination thereof.

[0026] The data acquisition device 140 refers to a device for acquiring an image of a captured object and / or a distance to the captured object. The captured object may include a road and objects on the road (e.g., cars, pedestrians, fences, etc. on the road). The data acquisition device 140 may include a laser radar, a planar shooting device (e.g., a black and white camera, a color camera, a scanner, etc. or any combination thereof) and / or a 3D shooting device, etc. In some embodiments, the data acquisition device 140 may be used for a road surface image of the road 150.

[0027] Road 150 refers to a road in a high-precision map or a road that can be driven by an autonomous vehicle (e.g., an autonomous car). Figure 1 As shown, there may be a road construction area 151 on the road 150 where no traffic is allowed.

[0028] Figure 2 FIG. 1 is an exemplary flow chart of a method for identifying a road construction area according to some embodiments of this specification. Figure 2 As shown, the process 200 includes the following steps: In some embodiments, the process 200 may be executed by the road construction area identification system 700 .

[0029] Step 210 , determining a plurality of candidate objects based on the road data. In some embodiments, step 210 may be performed by the object determination module 710 .

[0030] Road data refers to road surface image data of roads in high-precision maps or roads on which autonomous vehicles (e.g., autonomous vehicles) travel. Road data can be acquired based on an image data acquisition device (e.g., data acquisition device 140).

[0031] Candidate objects refer to objects used to mark road construction areas, such as signboards, water barriers, cone barrels, fences, construction vehicles, etc.

[0032] In some embodiments, the object determination module 710 may process the road data based on an image recognition model to determine candidate objects in the road data.

[0033] The image recognition model may be a machine learning model, for example, a convolutional neural network (CNN) model, etc. The input of the image recognition model may include road data, and the output may include candidate objects in the road data.

[0034] Step 220 , forming a plurality of three-point sets based on a plurality of candidate objects. In some embodiments, step 220 may be performed by a three-point set determination module 720 .

[0035] Each set of three points corresponds to three candidate objects that are adjacent in space.

[0036] A three-point set refers to an ordered set of position feature points corresponding to three candidate objects that are adjacent in space. Adjacent in space means that among the three candidate objects corresponding to the three position feature points that constitute the three-point set, there are no other candidate objects between any two candidate objects. In some embodiments, if other candidate objects between two candidate objects belong to the same clustering result as one of the two candidate objects, it can also be considered that there are no other candidate objects between the two candidate objects. For more information about clustering results, please refer to the following text. Position feature points refer to points that reflect the relative spatial positions between candidate objects in the road. For example, the geometric center point of the candidate object can be used as its corresponding position feature point to reflect the relative spatial positions between the candidate objects. Ordered means that the arrangement order of the three position feature points in a three-point set must meet the order rule. For example, taking the road as a north-south direction, the order rule can be "from south to north" (that is, the candidate object is closer to the south in space, and the corresponding position feature point is ranked higher in the three-point set). Suppose there are three candidate objects on the road, and their corresponding position feature points are a, b and c, where a is at the northernmost end, c is at the southernmost end, and b is in the middle. The corresponding three-point set is cba. The order rule can be preset. It is understandable that the direction order in the order rule may not be parallel to the road direction. For example, continuing with the above example, the road is in a north-south direction, and the order rule can also be "from east to west", etc.

[0037] Figure 4 It is a schematic diagram of location feature points of candidate objects in road data according to some embodiments of the present specification.

[0038] like Figure 4 As shown, Figure 4 The road in the figure is a two-lane road with a driving direction from south to north (the double solid line on the far left is a double yellow line, the dotted line is a lane line, and the rightmost side is a road boundary). Points a, b, c, d, e, f, g, h, i, j, k, l, and m are the position feature points corresponding to different candidate objects in the road data. For example, if the order rule is "from south to north", Figure 4 The three-point set composed of the position feature points in can include dae, aeg, bfg, fgi, etc. Figure 4 The order rules used in the examples are all "from south to north".

[0039] In some embodiments, the three-point set determination module 720 may select three position feature points that meet the preset distance condition based on the preset distance condition, and then determine the three-point set.

[0040] In some embodiments, the preset distance condition may include that the distance between the location feature points included in the three-point set is greater than a first distance threshold and less than a second distance threshold. Figure 4 As shown, for three position feature points a, e, and g, if the distance between position feature point a and position feature point e, and the distance between position feature point e and position feature point g are both greater than the first distance threshold and less than the second distance threshold, the three-point set aeg satisfies the preset distance condition, and the position feature points a, e, and g can form the three-point set aeg. For another example, for three position feature points a, d, and e, if the distance between position feature point a and position feature point d is less than the first distance threshold, the three-point set dae does not meet the preset distance condition, and the position feature points a, d, and e cannot form the three-point set. The first distance threshold and the second distance threshold can be preset, for example, the first distance threshold is preset to 1m, and the second distance threshold is preset to 8m.

[0041] In some embodiments, the preset distance condition may also include that the distance between the last position feature point in the three-point set and the mirror image point of the first position feature point (hereinafter referred to as the first mirror image point) is less than a third distance threshold. The first mirror image point refers to the symmetric point of the first position feature point with the second position feature point as the symmetry center. For example, Figure 4As shown, for three position feature points a, e, and g, if the distance between the first mirror point a' and the position feature point g is less than the third distance threshold, the three-point set aeg satisfies the preset distance condition, and the position feature points a, e, and g can form the three-point set aeg. For another example, for three position feature points a, e, and k, if the distance between the first mirror point a' and the position feature point k is greater than the third distance threshold, the three-point set aek does not meet the preset distance condition, and the position feature points a, e, and k cannot form the three-point set. The third distance threshold can be preset, for example, 2.5m.

[0042] In some embodiments, the three-point set determination module 720 may cluster multiple candidate objects based on a preset algorithm to obtain at least one clustering result; and determine multiple three-point sets based on the clustering result.

[0043] The preset algorithm refers to an algorithm used to cluster candidate objects. For example, the preset algorithm may be a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm, a Euclidean clustering algorithm, a k-means clustering algorithm, or the like.

[0044] Still Figure 4 For example, after clustering multiple candidate objects in the road data, the obtained clustering results may include clustering result T (including only position feature point l), clustering result U (including only position feature point m), V (including position feature points a, b, c, d), clustering result W (including position feature points e, f), clustering result X (including only position feature point g), clustering result Y (including position feature points h, i, j), and clustering result Z (including only position feature point k).

[0045] In some embodiments, the three-point set determination module 720 may determine the three-point set based on the position feature points belonging to different clustering results. For example, the three-point set egh may be determined based on the position feature points e, g, and h (which belong to the clustering results W, X, and Y, respectively). For another example, since among the position feature points e, h, and j, h and j both belong to the clustering result Y, e, h, and j cannot form a three-point set.

[0046] In some embodiments, the three-point set determination module 720 can respectively target each clustering result (referred to as the first clustering result) and respectively include each position feature point as the first position feature point in the three-point set; based on the first clustering result, select position feature points that do not exceed the quantity threshold and meet the preset distance condition from the clustering results that meet the sequence rule (referred to as the second clustering result) as the second position feature point in the three-point set; based on the second clustering result, select position feature points that meet the preset distance condition from the clustering results that meet the sequence rule as the third position feature point in the three-point set to obtain the three-point set. It can be understood that if the number of position feature points that meet the preset distance condition in the clustering results that meet the sequence rule is greater than the quantity threshold based on the first clustering result, then the position feature points with the closest quantity threshold number of distances are selected from the position feature points that meet the preset distance condition. The quantity threshold can be preset, for example, the quantity threshold can be preset to 2. For example, still based on Figure 4 For example, assuming that the quantity threshold is preset to 1, for the clustering result V (the first clustering result), the position feature points a, b, c, and d included therein can be respectively used as the first position feature point in the three-point set; taking the clustering result V as a reference, select the position feature points that meet the preset distance condition and do not exceed the quantity threshold (1) from the clustering results that meet the order rule (the second clustering result, including the clustering result U and the clustering result W) (assuming that the position feature points e and f in the clustering result W both meet the preset distance condition, then select the position feature point f with the closest distance; Assuming that the position feature point m in the clustering result U does not meet the preset distance condition) is used as the second position feature point in the three-point set; then, based on the clustering result W, the position feature point that meets the preset distance condition is selected from the clustering results that meet the order rule (including the clustering result X and the clustering result T) (assuming that the position feature point g in the clustering result X meets the preset distance condition, and the position feature point l in the clustering result T does not meet the preset distance condition) as the third position feature point in the three-point set, and the obtained three-point set includes afg, bfg, cfg and dfg.

[0047] It can be understood that the methods for determining the three-point set used in the above embodiments can be effective at the same time, so that the determined three-point set can simultaneously meet the above preset distance condition and quantity restriction.

[0048] In some embodiments of the present specification, the above-mentioned method for determining three-point sets can avoid similar candidate objects from appearing in the same three-point set, effectively simplifying the number of similar three-point sets and enhancing the specificity of the three-point sets; by limiting the number of position feature points selected from each clustering result, the number of three-point sets is further reduced, greatly reducing the amount of calculation in subsequent processes and saving computing power costs.

[0049] Step 230 , constructing a directed graph based on multiple triple point sets. In some embodiments, step 230 may be performed by the graph construction module 730 .

[0050] A directed graph refers to a graph data structure built based on a set of three points. A directed graph may include nodes and directed connected edges, wherein each node corresponds to a set of three points, and the sets of three points may be connected based on directed connected edges.

[0051] The nodes of the directed graph may have node attributes, which may include the intra-set deviation of the three-point set corresponding to the node.

[0052] The intra-set deviation is data reflecting the relative spatial positions of the three candidate objects corresponding to the three-point set. The larger the intra-set deviation value is, the stronger the symmetry between the first candidate object and the third candidate object about the symmetry center is, with the second candidate object as the symmetry center.

[0053] In some embodiments, the intra-set deviation may be determined based on the first distance, the second distance, the third distance, the first distance upper limit, the third distance upper limit, the intra-set angle, and the angle upper limit.

[0054] The first distance is the distance between the first position feature point and the second position feature point in the three-point set. For example, taking the three-point set as abc (the first mirror point is a') as an example, the first distance is the distance between the position feature point a and the position feature point b.

[0055] The second distance is the distance between the third position feature point in the three-point set and the first mirror image point. For example, continuing with the above example, the second distance is the distance between the position feature point c and the first mirror image point a'.

[0056] The third distance is the perpendicular distance between the first mirror point and the first straight line. The first straight line refers to the straight line where the first position feature point and the second position feature point in the three-point set are located. For example, continuing with the above example, the third distance is the perpendicular distance between the first mirror point a′ and the straight line ab (the straight line where the position feature points a and b are located).

[0057] The first distance upper limit is an upper limit value of the first distance. The first distance upper limit can be preset, for example, the first distance upper limit can be preset to 7m.

[0058] The third distance upper limit is an upper limit value of the third distance. The third distance upper limit can be preset, for example, it can be preset to be equal to the lane width of the road.

[0059] The internal angle is the angle between the first vector and the second vector. The first vector is the spatial vector from the second position feature point in the three-point set to the third position feature point. The second vector is the spatial vector from the second position feature point in the three-point set to the first mirror point. For example, continuing with the above example, the internal angle is the vector (the spatial vector from the position feature point b to the position feature point c) and the vector The angle between (the spatial vector pointing from the position feature point b to the first mirror point a′).

[0060] The angle upper limit is the upper limit value of the angle in the set. The angle upper limit can be preset, for example, it can be preset as the inverse tangent function value of the ratio of the third distance upper limit to the first distance.

[0061] In some embodiments, the intra-set deviation may be calculated by a formula. Still taking the three-point set abc as an example, the formula may be as shown in the following formula (1):

[0062] Among them, e is the deviation within the set; d ca′ is the second distance; d 12 is the upper limit of the first distance; d ab is the first distance; θ is the internal angle of the set; θ max is the upper limit of the angle, and its value can be equal to d v is the third distance; d v,max is the upper limit of the third distance; α, β, γ are weight coefficients, and their values ​​can be preset.

[0063] In some embodiments of the present specification, the above method for determining the intra-set deviation is used so that the obtained intra-set deviation can fully reflect the degree of discreteness of the candidate objects corresponding to the three-point set.

[0064] The directed connected edges of the directed graph may have edge attributes. The edge attributes may include the connected distance of the directed connected edge. The connected distance refers to data reflecting the relative deviation of two nodes connected by the directed connected edge. The larger the connected distance, the greater the relative deviation of the two nodes connected by the directed connected edge.

[0065] In some embodiments, the graph construction module 730 may determine the connection distance of the directed connected edge based on the sum of the intra-set deviations of the first three-point set and the second three-point set. The first three-point set is the starting node of the directed connected edge, and the second three-point set is the end node of the directed connected edge. For example, the sum of the intra-set deviations of the first three-point set and the second three-point set may be used as the connection distance of the corresponding directed connected edge.

[0066] In some embodiments, the graph construction module 730 may use each three-point set as a node of a directed graph; for any two three-point sets, if the last two candidate objects corresponding to the first three-point set are the same as the first two candidate objects corresponding to the second three-point set, a directed connected edge is constructed between the first three-point set node and the second three-point set node to obtain an initial directed graph (for example, for three-point set abc and three-point set bcd, the last two candidate objects corresponding to three-point set abc are candidate object b and candidate object c, and the first two candidate objects corresponding to three-point set bcd are also candidate object b and candidate object c, then a directed connected edge from three-point set abc to three-point set bcd can be constructed between three-point set abc and three-point set bcd); based on the intra-set deviation of the first three-point set and the second three-point set, the connected distance of the directed connected edge is determined; based on the connected distance, the directed connected edges that do not meet the first preset condition are filtered out to obtain a directed graph.

[0067] The first preset condition refers to that the connection distance of the directed connected edge is less than a connection distance threshold. The connection distance threshold can be preset, for example, the connection distance threshold can be preset to 1.

[0068] Figure 5 It is a schematic diagram of the directed graph construction process shown in some embodiments of this specification.

[0069] like Figure 5 As shown, the three-point set 510 is based on Figure 4 The position feature points shown are a set of three point sets obtained according to the order rule of "from south to north" and the preset quantity threshold of 2. The three-point set 510 includes three-point sets A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, and P (corresponding to the three-point sets aeg, beg, ceg, deg, afg, bfg, cfg, dfg, egh, egi, egj, fgh, fgi, fgj, gik, and gjk, respectively).

[0070] Based on the aforementioned directed graph construction method and the three-point set 510, an initial directed graph is determined: since the last two candidate objects corresponding to the three-point set A (aeg) are the same as the first two candidate objects corresponding to the three-point set I (egh) (both are e and g), a directed connected edge AI is constructed between the three-point set A and the three-point set I (referring to the directed connected edge from the three-point set A to the three-point set I, and the same applies to other directed connected edges involved in the following text); since the last two candidate objects corresponding to the three-point set J (egi) are the same as the first two candidate objects corresponding to the three-point set O (gik) (both are g and i), a directed connected edge JO is constructed between the three-point set J and the three-point set O; and so on, until all feasible directed connected edges are constructed, and an initial directed graph is obtained.

[0071] Determine the connectivity distance of the directed connected edge: The connectivity distance can be calculated by the formula. Taking the directed connected edge AI as an example, the formula can be shown as follows (2): ω AI =e A +e I (2)

[0072] Among them, ω AI is the connected distance of the directed connected edge AI, e A 、e I They are the intra-set deviations of three-point set A and three-point set I respectively. The calculation method of the connected distances of other directed connected edges is the same as that of directed connected edge AI.

[0073] like Figure 5 As shown, assuming that after calculation, the directed connected edges AJ, DK, and GM do not meet the first preset condition, the above three directed connected edges are removed from the initial directed graph to obtain a directed graph.

[0074] Step 240 , determining at least one directed path based on the directed graph, and determining the best path from the at least one directed path. In some embodiments, step 240 may be performed by the path determination module 740 .

[0075] A directed path is the shortest path between any two nodes in a directed graph. A path is a set of directed edges connecting any two nodes. For example, suppose in a directed graph (different from Figure 5 In the directed graph shown in ), there is a directed connected edge LM between node L and node M, and there is a directed connected edge MN between node M and node N, then it is considered that there is a path LM-MN between node L and node N. The shortest path refers to the path with the least number of nodes passed through among all the paths between two nodes. For example, continuing with the above example, assuming that in the directed graph, in addition to the path LM-MN (passing through node M, passing through one node in total) between node L and node N, there are also paths LO-OP-PN (passing through nodes O and P, passing through two nodes in total) and LQ-QR-RS-SN (passing through nodes Q, R, S, passing through three nodes in total), then the path LM-MN is the shortest path, which will be used as the directed path between node L and node N.

[0076] In some embodiments, the path determination module 740 may determine a directed path between any two nodes in a directed graph based on a shortest path algorithm, which may include Dijkstra's algorithm, Bellman-Ford algorithm, Floyd-Warshall algorithm, and the like.

[0077] In some embodiments, the path determination module 740 may determine an evaluation value of each directed path; and determine an optimal path based on the evaluation value.

[0078] The evaluation value refers to data reflecting the degree of discreteness between candidate objects corresponding to the related nodes of the directed path. The related nodes include the starting node, the path node and the end node of the directed path. The smaller the value of the evaluation value, the lower the degree of discreteness between candidate objects corresponding to the related nodes of the directed path.

[0079] In some embodiments, the evaluation value can be determined based on the total number of candidate objects, the number of related nodes, and the maximum intra-set deviation of the three-point sets corresponding to the related nodes. The total number of candidate objects refers to the total number of candidate objects corresponding to the three-point sets corresponding to all nodes in the directed graph. The maximum intra-set deviation refers to the maximum value of the intra-set deviations of all three-point sets corresponding to the related nodes.

[0080] In some embodiments, the evaluation value can be calculated by a formula. Still taking the three-point set abc as an example, the formula can be shown as the following formula (3): S=N·e max n (3)

[0081] Among them, S is the evaluation value, N is the total number of candidate objects, and e max is the maximum intra-set deviation, and n is the number of related nodes.

[0082] In some embodiments, the path determination module 740 may determine the directional path with the smallest evaluation value as the optimal path.

[0083] In some embodiments of the present specification, the above method for determining the best path can minimize the discreteness of the candidate objects corresponding to the acquired best path, which helps to improve the accuracy of the road construction area determined subsequently.

[0084] Step 250 , determining a road construction area based on the optimal path. In some embodiments, step 250 may be performed by the area determination module 750 .

[0085] In some embodiments, the area determination module 750 may connect the candidate objects corresponding to the best path to obtain a candidate object line, and determine the road construction area based on the candidate object line and the road boundary. For example, the area surrounded by the candidate object line and the road boundary may be determined as the road construction area.

[0086] In some embodiments, the area determination module 750 can determine multiple target objects based on the candidate objects corresponding to the nodes included in the optimal path; generate an initial area based on the outer contour of the target object; determine the boundary distance of each target object to the road boundary; in response to the existence of at least one target object whose boundary distance satisfies the second preset condition: determine the first target object and the second target object with the farthest relative distance among the multiple target objects; determine the first perpendicular line between the first target object and the road boundary and the second perpendicular line between the second target object and the road boundary; determine the road construction area based on the initial area, the road boundary, the first perpendicular line and the second perpendicular line; in response to the absence of any target object whose boundary distance satisfies the second preset condition, determine the initial area as the road construction area. For more information about the above embodiments, please refer to Figure 3 and its related description.

[0087] In some embodiments of the present specification, the above-mentioned road construction area identification method can be used to accurately identify the construction area in the road; the above-mentioned road construction area identification method can be implemented intelligently based on processing equipment, thereby reducing computing power, manpower and time costs.

[0088] Figure 3 FIG. 1 is an exemplary flow chart of a method for determining a road construction area according to some embodiments of this specification. Figure 3 As shown, the process 300 includes the following steps. In some embodiments, the process 300 may be executed by the area determination module 750 .

[0089] Step 310: Determine multiple target objects based on the candidate objects corresponding to the related nodes of the optimal path. For more information about related nodes, see Figure 2 and its related description.

[0090] The target object refers to a candidate object determined from the candidate objects and used to determine the road construction area.

[0091] In some embodiments, the region determination module 750 may determine the candidate object corresponding to the relevant node of the optimal path as the target object. Figure 5 For example, assuming that the best path is AK-KP, the relevant nodes include nodes A, K, and P, and the corresponding three-point sets include aeg, egj, and gjk. The corresponding candidate objects include candidate object a (that is, the candidate object corresponding to position feature point a, and the same applies to other candidate objects described later), e, g, j, and k. These candidate objects are determined as target objects.

[0092] In some embodiments, the region determination module 750 may determine multiple target objects based on candidate objects corresponding to relevant nodes of the optimal path and clustering results.

[0093] The region determination module 750 may first determine the candidate objects corresponding to the related nodes of the best path as the target objects, and then determine other candidate objects that belong to the same clustering result as the candidate objects corresponding to the related nodes as the target objects. Figure 4 For example, candidate object a and candidate objects b, c, and d belong to cluster result V, candidate object e and candidate object f belong to cluster result W, and candidate object j and candidate objects h and i belong to cluster result Y, then candidate objects b, c, d, f, h, and i are also determined as target objects.

[0094] In some embodiments of the present specification, through the above method for determining target objects, clustering results are effectively used to expand the number of target objects and improve the accuracy of the road construction area results determined subsequently.

[0095] Step 320: Generate an initial region based on the outer contour of the target object.

[0096] The initial area refers to the area enclosed by the outer contour of the target object.

[0097] In some embodiments, the region determination module 750 may connect the outer contours of the target object, and use the region enclosed by the outer contours and the lines connecting the outer contours as the initial region.

[0098] Fig. 6A is a schematic diagram of an initial area according to some embodiments of the present specification.

[0099] like Fig. 6A As shown, the target objects include target object 611, target object 612, target object 613, target object 614 and target object 615, and area 610 is the area enclosed by the outer contour of the target object and the line connecting the outer contour. The area determination module 750 can use area 610 as the initial area.

[0100] Step 330 , determining the boundary distance from each target object to the road boundary.

[0101] The road boundary refers to the boundary between the road and its adjacent areas (for example, non-motorized vehicle lanes and green belts adjacent to the road).

[0102] The boundary distance refers to the vertical distance from the location feature point of the target object to the road boundary. For example, the boundary distance can be 2m.

[0103] In some embodiments, the area determination module 750 may process the road data based on a distance recognition model to determine the boundary distance from each target object to the road boundary.

[0104] The distance recognition model may be a machine learning model. The input of the image recognition model may include road data, and the output may include the boundary distance from each target object in the road data to the road boundary.

[0105] In some embodiments, the area determination module 750 can determine the boundary distance of each target object to the road boundary based on a distance sensor (e.g., radar) configured in an autonomous driving vehicle (e.g., an autonomous driving car).

[0106] Step 340, in response to the existence of at least one target object whose boundary distance satisfies the second preset condition: determining a first target object and a second target object with the farthest relative distance among multiple target objects; determining a first perpendicular line between the first target object and the road boundary and a second perpendicular line between the second target object and the road boundary; determining a road construction area based on the initial area, the road boundary, the first perpendicular line and the second perpendicular line.

[0107] The second preset condition refers to that the boundary distance is less than a boundary distance threshold. The boundary distance threshold can be preset, for example, 2m.

[0108] The first target object and the second target object refer to two target objects with the farthest relative distance among the multiple target objects.

[0109] In some embodiments, the area determination module 750 can determine the relative distance between any two target objects among multiple target objects based on a distance sensor (e.g., a radar) configured in an autonomous driving vehicle (e.g., an autonomous driving car), and determine the two target objects with the largest relative distance as the first target object and the second target object, respectively (regardless of order).

[0110] The first perpendicular line refers to a line segment that is perpendicular to the road boundary and tangent to the outer contour of the first target object, connecting the perpendicular point and the tangent point.

[0111] The second perpendicular line refers to a line segment which is perpendicular to the road boundary and tangent to the outer contour of the first target object, and connects the perpendicular point and the tangent point.

[0112] In some embodiments, the area determination module 750 may process the road data in which the first target object and the second target object have been determined based on a vertical line determination model to determine the first vertical line and the second vertical line.

[0113] The vertical line determination model may be a machine learning model. The input of the vertical line determination model may include road data in which the first target object and the second target object have been determined, and the output may include the first vertical line and the second vertical line.

[0114] In some embodiments, the area determination module 750 may merge the initial area, the road boundary, the area enclosed by the first vertical line and the second vertical line, and the initial area, and determine the merged result as the road construction area.

[0115] Figure 6B It is a schematic diagram of a method for determining a road construction area according to some embodiments of this specification.

[0116] like Figure 6B As shown, since the relative distance between target object 611 and target object 615 is the farthest among all target objects, target object 611 and target object 615 are respectively determined as the first target object and the second target object; based on this, a first perpendicular line 671 between the first target object and the road boundary and a second perpendicular line 681 between the second target object and the road boundary are obtained; and straight line 661 is the road boundary. The area enclosed by the initial area, the road boundary, the first perpendicular line and the second perpendicular line is merged with the initial area, and the merged result (area 620) is the road construction area.

[0117] Step 350 : In response to the absence of any target object whose boundary distance satisfies the second preset condition, the initial area is determined as a road construction area.

[0118] In some embodiments of the present specification, the above-mentioned method for determining the road construction area can cope with the determination of the road construction area under different situations (different distances from the target object to the road boundary), so that the size of the determined road construction area is appropriate, thereby enhancing the adaptability of the determined road construction area.

[0119] It should be noted that the above description of process 300 is only for example and illustration, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 300 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification. For example, the boundary distance threshold is set to 2.5m.

[0120] Figure 7 is a module diagram of a road construction area identification system according to some embodiments of this specification. Figure 7 As shown, the road construction area recognition system 700 may include an object determination module 710 , a three-point set determination module 720 , a graph construction module 730 , a path determination module 740 and an area determination module 750 .

[0121] The object determination module 710 is used to determine multiple candidate objects based on the road data. For more information about the function of the object determination module 710, see Figure 2 and its related description.

[0122] The three-point set determination module 720 is used to form multiple three-point sets based on multiple candidate objects. For more information about the function of the three-point set determination module 720, see Figure 2 and its related description.

[0123] The graph construction module 730 is used to construct a directed graph based on multiple three-point sets. For more information about the function of the graph construction module 730, see Figure 2 and its related description.

[0124] The path determination module 740 is used to determine at least one directed path based on the directed graph, and determine the best path from the at least one directed path. For more information about the function of the path determination module 740, see Figure 2 and its related description.

[0125] The area determination module 750 is used to identify road construction areas based on the optimal path.

[0126] In some embodiments, the area determination module 750 is also used to determine multiple target objects based on the candidate objects corresponding to the nodes included in the optimal path; generate an initial area based on the outer contour of the target object; determine the boundary distance of each target object to the road boundary; in response to the existence of at least one target object whose boundary distance satisfies the second preset condition: determine the first target object and the second target object with the farthest relative distance among the multiple target objects; determine the first perpendicular line between the first target object and the road boundary and the second perpendicular line between the second target object and the road boundary; determine the road construction area based on the initial area, the road boundary, the first perpendicular line and the second perpendicular line; in response to the absence of any target object whose boundary distance satisfies the second preset condition, determine the initial area as the road construction area. For more information about the functions of the area determination module 750, see Figure 2 and / or Figure 3 and its related description.

[0127] It should be noted that the above description of the road construction area identification system and its modules is only for convenience of description and does not limit the present specification to the scope of the embodiments. It is understandable that, after understanding the principle of the system, those skilled in the art may arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from the principle. In some embodiments, Figure 7 The object determination module 710, the three-point set determination module 720, the graph construction module 730, the path determination module 740 and the area determination module 750 disclosed in the specification can be different modules in a system, or a module can realize the functions of two or more of the above modules. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the protection scope of this specification.

[0128] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0129] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0130] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0131] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0132] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0133] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0134] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A road construction area identification method, characterized in that: The method comprises: Acquire multiple candidate objects based on road data; Based on the multiple candidate objects, a plurality of three-point sets are formed, each of the three-point sets corresponds to three candidate objects that are adjacent in spatial position; constructing a directed graph based on the multiple triple point sets; Determining at least one directed path based on the directed graph, and determining an optimal path from the at least one directed path; A road construction area is identified based on the optimal path.

2. The method according to claim 1, characterized in that The forming a plurality of three-point sets based on the plurality of candidate objects comprises: Based on a preset algorithm, pre-clustering the multiple candidate objects to obtain at least one clustering result; Based on the clustering result, the plurality of three-point sets are determined.

3. The method according to claim 1, characterized in that The constructing a directed graph based on the multiple three-point sets comprises: Taking each of the three point sets as a node of the directed graph; For any two three-point sets, if the second two candidate objects corresponding to the first three-point set are the same as the first two candidate objects corresponding to the second three-point set, a directed connected edge is constructed between the nodes of the first three-point set and the nodes of the second three-point set to obtain an initial directed graph; Determining the connected distance of the directed connected edge based on the intra-set deviation of the first three-point set and the second three-point set; the intra-set deviation is data reflecting the relative spatial positions of the three candidate objects corresponding to the three-point set; Based on the connectivity distance, the directed connectivity edges that do not satisfy a first preset condition are filtered out to obtain the directed graph.

4. The method according to claim 3, characterized in that The in-set deviation is determined based on the first distance, the second distance, the third distance, the first distance upper limit, the third distance upper limit, the in-set angle and the angle upper limit; the first distance is the distance between the first position feature point and the second position feature point in the three-point set, the second distance is the distance between the third position feature point and the first mirror point in the three-point set, the third distance is the perpendicular distance between the first mirror point and the first straight line, and the in-set angle refers to the angle between the first vector and the second vector; the first vector is a spatial vector pointing from the second position feature point in the three-point set to the third position feature point, and the second vector is a spatial vector pointing from the second position feature point in the three-point set to the first mirror point.

5. The method according to claim 3, characterized in that The determining the connected distance of the directed connected edge based on the intra-set deviation of the first three-point set and the second three-point set comprises: Based on the sum of the intra-set deviations of the first three-point set and the second three-point set, a connectivity distance of the directed connected edge is determined.

6. The method according to claim 1, characterized in that Determining the best path from the at least one directed path comprises: For each of the directed paths, determining an evaluation value of the directed path; Based on the evaluation value, an optimal path is determined.

7. The method according to claim 1, characterized in that The identifying of the road construction area based on the optimal path comprises: Determine a plurality of target objects based on the candidate objects corresponding to the nodes included in the optimal path; Based on the outer contour of the target object, generating an initial area; Determining a boundary distance from each of the target objects to a road boundary; In response to at least one boundary distance of the target object satisfying a second preset condition: Determine a first target object and a second target object that are farthest in relative distance from each other among the plurality of target objects; Determine a first perpendicular line between the first target object and the road boundary and a second perpendicular line between the second target object and the road boundary; Determining the road construction area based on the initial area, the road boundary, the first vertical line and the second vertical line; In response to the fact that there is no boundary distance of any of the target objects that meets the second preset condition, the initial area is determined as the road construction area.

8. The method according to claim 2 or 7, characterized in that The determining of a plurality of target objects based on the candidate objects corresponding to the nodes included in the optimal path further comprises: Based on the candidate objects corresponding to the nodes included in the optimal path and the clustering result, a plurality of target objects are determined.

9. A road construction area identification system, characterized in that: The system comprises: An object determination module, for obtaining a plurality of candidate objects based on road data; A three-point set determination module, configured to form a plurality of three-point sets based on the plurality of candidate objects, each of the three-point sets corresponding to three candidate objects adjacent in spatial position; A graph construction module, configured to construct a directed graph based on the plurality of triple point sets; A path determination module, configured to determine at least one directed path based on the directed graph, and determine an optimal path from the at least one directed path; An area determination module is used to identify a road construction area based on the optimal path.

10. The system according to claim 9, characterized in that The three-point set determination module is further used for: Based on a preset algorithm, pre-clustering the multiple candidate objects to obtain at least one clustering result; Based on the clustering result, the plurality of three-point sets are determined.

11. The system according to claim 9, characterized in that The graph construction module is further used to: Taking each of the three point sets as a node of the directed graph; For any two three-point sets, if the second two candidate objects corresponding to the first three-point set are the same as the first two candidate objects corresponding to the second three-point set, a directed connected edge is constructed between the nodes of the first three-point set and the nodes of the second three-point set to obtain an initial directed graph; Determining the connected distance of the directed connected edge based on the intra-set deviation of the first three-point set and the second three-point set; the intra-set deviation is data reflecting the relative spatial positions of the three candidate objects corresponding to the three-point set; Based on the connectivity distance, the directed connectivity edges that do not satisfy a first preset condition are filtered out to obtain the directed graph.

12. The system according to claim 9, characterized in that The path determination module is further used for: For each of the directed paths, determining an evaluation value of the directed path; Based on the evaluation value, an optimal path is determined.

13. The system of claim 9, wherein: The area determination module is further used for: Determine a plurality of target objects based on the candidate objects corresponding to the nodes included in the optimal path; Based on the outer contour of the target object, generating an initial area; Determining a boundary distance from each of the target objects to a road boundary; In response to at least one boundary distance of the target object satisfying a second preset condition: Determine a first target object and a second target object that are farthest in relative distance from each other among the plurality of target objects; Determine a first perpendicular line between the first target object and the road boundary and a second perpendicular line between the second target object and the road boundary; Determining the road construction area based on the initial area, the road boundary, the first vertical line and the second vertical line; In response to the fact that there is no boundary distance of any of the target objects that meets the second preset condition, the initial area is determined as the road construction area.

14. A road construction area recognition device, comprising a storage medium and a processor, wherein the storage medium stores computer instructions, and the processor is used to execute at least part of the computer instructions to implement the method according to any one of claims 1 to 8.

15. A storage medium for storing computer instructions, when at least a part of the computer instructions is executed by a processor, implementing the method according to any one of claims 1 to 8.

Images