Method and apparatus for constructing road environment of autonomous driving test site

Abstract

The present invention relates to the field of autonomous driving technology, and in particular, to a method and device for constructing a road environment of an autonomous driving test site. First, obtain the road nodes and road segments included in the real road map data, as well as the subordinate relationship between the two; determine the type attributes of the road nodes and the type attributes of the road segments, and construct a classification set of typical road structures according to the type attributes of the road nodes and the type attributes of the road segments; then, based on the classification set of typical road structures, cluster the real road map data to obtain a set of basic road units included in the real road map data; finally, determine the importance index of each basic road unit, and obtain the road environment of the test site through optimal combination. The road environment of the test site constructed according to the real road map data in the present invention is closer to the real road scene, can cover road environments of multiple different scenarios, meets the test requirements of autonomous driving vehicles, and makes the test results more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and particularly to a method and device for constructing a road environment of an autonomous driving test site. Background Art

[0002] An autonomous driving vehicle is an intelligent transportation vehicle that integrates environmental perception, decision-making and planning, and control execution, representing the strategic high point of future automotive technology. With the development of a series of key technologies, autonomous driving vehicles provide a new way to solve problems such as "traffic accidents, traffic congestion, environmental pollution, and energy shortage". Scientific and perfect test verification and evaluation are crucial for improving the R & D efficiency of autonomous driving vehicles, improving relevant laws and regulations, and promoting the development of intelligent transportation.

[0003] Different from the method of the human-vehicle binary independent test of traditional vehicles, an autonomous driving vehicle is a strong coupling system of human-vehicle-road-environment. The traditional vehicle-based mileage test method has problems such as high safety risks, random test scenarios, and difficulty in repeating scenarios, so it is difficult to meet the test requirements of autonomous driving vehicles.

[0004] The closed-field test of autonomous driving vehicles has high safety, the test scenarios can be flexibly constructed according to test requirements, and the test has strong repeatability, and can realize the vehicle test of autonomous driving vehicles in typical and high-risk scenarios. Therefore, the closed-field test of autonomous driving vehicles has become an indispensable important link in the test evaluation of autonomous driving vehicles. However, the existing road environments of closed test sites for autonomous driving vehicles are mostly constructed based on standard specifications and expert experience, so there are problems of insufficient authenticity and typicality and low coverage, and it is difficult to meet the requirements of autonomous driving vehicle tests for road authenticity, typicality, and coverage. Therefore, there is an urgent need for a method for constructing a road environment of a closed test site that can meet the requirements of autonomous driving vehicle closed test sites for test road authenticity, typicality, and coverage. Summary of the Invention

[0005] The present invention provides a method and device for constructing a road environment of an autonomous driving test site to solve the technical defects of insufficient authenticity and low coverage of the road environment of the existing closed test site.

[0006] In a first aspect, the present invention provides a method for constructing a road environment of an autonomous driving test site, including:

[0007] Obtaining real road map data;

[0008] Obtaining road nodes and road segments included in the real road map data, and the subordinate relationships of the road nodes and road segments;

[0009] Determine the type attributes of the road nodes and the type attributes of the road segments, and construct a typical road structure classification set according to the type attributes of the road nodes and the type attributes of the road segments;

[0010] Based on the typical road structure classification set, cluster the real road map data to obtain the set of basic road units included in the real road map data;

[0011] Determine the importance index of each basic road unit in the set of basic road units, and combine the basic road units using an optimization combination algorithm in the order from high to low importance index to obtain the road environment of the test site.

[0012] In a second aspect, the present invention also provides a device for constructing a road environment of an autonomous driving test site, including:

[0013] An acquisition unit for acquiring real road map data;

[0014] A first processing unit for acquiring the road nodes and road segments included in the real road map data, and the subordinate relationships of the road nodes and road segments;

[0015] A second processing unit for determining the type attributes of the road nodes and the type attributes of the road segments, and constructing a typical road structure classification set according to the type attributes of the road nodes and the type attributes of the road segments;

[0016] A clustering unit for clustering the real road map data based on the typical road structure classification set to obtain the set of basic road units included in the real road map data;

[0017] A third processing unit for determining the importance index of each basic road unit in the set of basic road units, and combining the basic road units using an optimization combination algorithm in the order from high to low importance index to obtain the road environment of the test site.

[0018] In a third aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the method for constructing a road environment of an autonomous driving test site as described in any one of the above.

[0019] In a fourth aspect, the present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the method for constructing a road environment of an autonomous driving test site as described in any one of the above.

[0020] The method for constructing the road environment of an autonomous driving test site provided by the present invention first obtains the road nodes and road segments included in the real road map data, as well as the subordinate relationships between the road nodes and road segments; determines the type attributes of the road nodes and the type attributes of the road segments, and constructs a classification set of typical road structures according to the type attributes of the road nodes and the type attributes of the road segments; then based on the classification set of typical road structures, clusters the real road map data to obtain a set of basic road units included in the real road map data; finally determines the importance index of each basic road unit in the set of basic road units, and combines the basic road units using an optimization combination algorithm according to the order from high to low of the importance index to obtain the road environment of the test site. The road environment of the test site constructed according to the real road map data by the present invention is closer to the real road scene, can cover road environments of multiple different scenarios, meets the test requirements of autonomous driving vehicles, and makes the test results more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0022] Figure 1 It is a flowchart of the method for constructing the road environment of an autonomous driving test site provided by an embodiment of the present invention;

[0023] Figure 2 It is a flowchart of using a genetic algorithm for optimizing the deployment of the road environment of the test site provided by an embodiment of the present invention;

[0024] FIG. 3(a) is a schematic diagram of the structure of a highway straight road unit provided by an embodiment of the present invention;

[0025] FIG. 3(b) is a schematic diagram of the structure of a highway entrance ramp intersection road unit provided by an embodiment of the present invention;

[0026] FIG. 3(c) is a schematic diagram of the structure of a highway exit ramp intersection road unit provided by an embodiment of the present invention;

[0027] FIG. 3(d) is a schematic diagram of the structure of a highway entrance - arterial road intersection road unit (driving out of the arterial road) provided by an embodiment of the present invention;

[0028] FIG. 3(e) is a schematic diagram of the structure of a highway exit - arterial road intersection road unit (merging into the arterial road) provided by an embodiment of the present invention;

[0029] Figure 3(f) is a schematic structural diagram of a main road straight road unit provided by an embodiment of the present invention;

[0030] Figure 3(g) is a schematic structural diagram of a main road - main road T - type intersection road unit provided by an embodiment of the present invention;

[0031] Figure 3(h) is a schematic structural diagram of a main road - main road cross - intersection road unit provided by an embodiment of the present invention;

[0032] Figure 3(i) is a schematic structural diagram of a main road - main road connecting line entrance road unit provided by an embodiment of the present invention;

[0033] Figure 3(j) is a schematic structural diagram of a main road - main road connecting line exit road unit provided by an embodiment of the present invention;

[0034] Figure 3(k) is a schematic structural diagram of a main road - branch road T - type intersection road unit provided by an embodiment of the present invention;

[0035] Figure 3(l) is a schematic structural diagram of a main road - branch road cross - intersection road unit provided by an embodiment of the present invention;

[0036] Figure 3(m) is a schematic structural diagram of a main road - internal road T - type intersection road unit provided by an embodiment of the present invention;

[0037] Figure 3(n) is a schematic structural diagram of a main road - internal road cross - intersection road unit provided by an embodiment of the present invention;

[0038] Figure 3(o) is a schematic structural diagram of a branch road straight road unit provided by an embodiment of the present invention;

[0039] Figure 3(p) is a schematic structural diagram of a branch road - branch road T - type intersection road unit provided by an embodiment of the present invention;

[0040] Figure 3(q) is a schematic structural diagram of a branch road - branch road cross - intersection road unit provided by an embodiment of the present invention;

[0041] Figure 3(r) is a schematic structural diagram of a branch road - internal road T - type intersection road unit provided by an embodiment of the present invention;

[0042] Figure 3(s) is a schematic structural diagram of a branch road - internal road cross - intersection road unit provided by an embodiment of the present invention;

[0043] Figure 3(t) is a schematic structural diagram of an internal road straight road unit provided by an embodiment of the present invention;

[0044] Figure 3(u) is a schematic structural diagram of an internal road - internal road T - type intersection road unit provided by an embodiment of the present invention;

[0045] Figure 3(v) is a schematic structural diagram of an internal road - internal road cross - road unit provided by an embodiment of the present invention.

[0046] Figure 4 It is a schematic structural diagram of a test site road structure obtained by optimizing and deploying based on a genetic algorithm provided by an embodiment of the present invention;

[0047] Figure 5 It is a test site road environment map obtained by optimizing and deploying based on a genetic algorithm provided by an embodiment of the present invention;

[0048] Figure 6 It is a schematic structural diagram of a device for constructing an autonomous driving test site road environment provided by an embodiment of the present invention. Detailed implementation manners

[0049] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0050] Below, in combination with Figures 1 - 6 Describe the method and device for constructing an autonomous driving test site road environment of the present invention.

[0051] Figure 1 It is a flowchart of a method for constructing an autonomous driving test site road environment provided by an embodiment of the present invention. Please refer to Figure 1 As shown, the construction method includes:

[0052] S101. Obtain real - road map data.

[0053] In one or more embodiments, a partial area of the real - map data of a certain city can be selected as the real - road map data, or the real - road map data of a certain city can be obtained by means of remote sensing images.

[0054] For example, according to the requirements, objectives, and regional characteristics of building a test site, a partial area is selected from the real map of a certain city as the source of real - road data. In this embodiment, openstreetmap (abbreviated as OSM, Chinese is Open Map) is used as the source of real - map data. In this embodiment, based on the real - map data of City A, the real - road data for building an autonomous driving test site is established, including different types of roads such as highways, urban arterial roads, branch roads, and internal roads.

[0055] S102. Obtain the road nodes and road segments included in the real road map data, as well as the subordinate relationships between the road nodes and road segments.

[0056] In this embodiment, the way data of road segments and node data of road nodes in the openstreetmap map data of City A are read respectively to determine the positions of the road nodes and analyze which road segments the road nodes belong to. In this embodiment, for a single node, the nodes located at both ends of a road segment are the boundary points of this road segment; the nodes located between the two boundary points of a road segment are the internal points of this road segment.

[0057] If a road node belongs to two or more road segments, then this node is the connection point of the road segments. If a road node belongs to two road segments, then this node is the connection point of the two road segments; if a road node belongs to three road segments, then this node is the connection point of the three road segments; if a road node belongs to four road segments, then this node is the connection point of the four road segments.

[0058] S103. Determine the type attributes of the road nodes and the type attributes of the road segments, and construct a typical road structure classification set according to the type attributes of the road nodes and the type attributes of the road segments.

[0059] In one or more embodiments, the type attributes of the road nodes include: boundary points of straight road structures, center points of T-shaped intersection road structures, boundary points of T-shaped intersection road structures, center points of cross intersection road structures, boundary points of cross intersection road structures, center points of main road ramp intersection road structures, and boundary points of main road ramp intersection road structures;

[0060] The type attributes of the road segments include: expressways, expressway connectors, main roads, main road connectors, branch roads, and internal roads;

[0061] The basic road structure classification set includes straight roads, T-shaped intersection roads, cross intersection roads, and main road-ramp intersection roads.

[0062] In one or more embodiments, for each road segment, according to the type attribute of the road segment and the type attribute of the road node of the road segment, determine the typical road structure composed of the road segment and the corresponding road node;

[0063] Construct the typical road structure classification set according to multiple different typical road structures.

[0064] In this embodiment, if a road node is the connection point of four road segments, it indicates that the selected map contains a crossroads road structure. If a road node is the connection point of three road segments, it indicates that the selected map contains a three-way intersection. The three-way intersection can be further divided into a T-shaped intersection road structure and a main road-ramp intersection road structure. If a road segment does not belong to the crossroads road structure, the T-shaped intersection road structure or the main road-ramp intersection road structure, then this road segment is a straight road structure. Therefore, in this embodiment, the basic road structure classification set includes the straight road structure, the T-shaped intersection road structure, the crossroads road structure and the main road-ramp intersection road structure.

[0065] Analyze the attributes of road segments and road nodes. In this embodiment, according to the selected map data, the road segment way is classified into expressway, expressway connection line, main road, main road connection line, branch road, and internal road according to the type attribute.

[0066] The road nodes are classified into internal points of straight road segments, boundary points of straight road structures, center points of T-shaped intersection road structures, boundary points of T-shaped intersection road structures, center points of crossroads road structures, boundary points of crossroads road structures, center points of main road-ramp intersection road structures, and boundary points of main road-ramp intersection road structures according to the type attribute.

[0067] Then, according to the type attributes of road nodes and road segments, a typical road structure classification set is established, specifically including: The typical road structure set includes 3 basic road structures of expressways, namely expressway straight road structure, expressway (main road)-expressway connection line (ramp) intersection road structure, and arterial road (main road)-expressway connection line (ramp) intersection road structure; 8 basic road structures of arterial roads, namely arterial road straight road structure, arterial road (main road)-arterial road connection line (ramp) intersection road structure, arterial road-arterial road T-shaped intersection road structure, arterial road-arterial road crossroads road structure, arterial road-branch road T-shaped intersection road structure, arterial road-branch road crossroads road structure, arterial road-internal road T-shaped intersection road structure, arterial road-internal road crossroads road structure; 5 basic road structures of branch roads, namely branch road straight road structure, branch road-branch road T-shaped intersection road structure, branch road-branch road crossroads road structure, branch road-internal road T-shaped intersection road structure, branch road-internal road crossroads road structure; 3 basic road structures of internal roads, namely internal road straight road structure, internal road-internal road T-shaped intersection road structure, and internal road-internal road crossroads road structure.

[0068] S104. Based on the typical road structure classification set, cluster the real road map data to obtain the set of basic road units included in the real road map data.

[0069] In one or more embodiments, according to the typical road structures included in the typical road structure classification set, clustering is performed on the real road map data to obtain multiple basic road units included in the real road map data;

[0070] A basic road unit set is obtained based on the multiple basic road units. That is, the basic road unit set includes multiple basic road units.

[0071] In this embodiment, according to the designed basic road structure, the roads within the selected area of City A are traversed, Kmeans clustering analysis is performed on the real roads, the road segment lengths and the included angles between each road segment in each basic road unit are extracted, and the road segment length with the largest proportion and the included angles between each road segment in the clustering result are selected as typical parameters to form 22 typical basic road units. Please refer to Figures 3(a) - 3(v) As shown, there are 5 basic road units for expressways: expressway straight road unit, expressway entrance ramp intersection road unit, expressway exit ramp intersection road unit, expressway entrance - arterial road intersection road unit, expressway exit - arterial road intersection road unit; 9 basic road units for arterial roads: arterial road straight road unit, arterial road - arterial road T - intersection road unit, arterial road - arterial road cross - intersection road unit, arterial road - arterial road connection line entrance road unit, arterial road - arterial road connection line exit road unit, arterial road - branch road T - intersection road unit, arterial road - branch road cross - intersection road unit, arterial road - internal road T - intersection road unit, arterial road - internal road cross - intersection road unit; 5 basic road units for branch roads: branch road straight road unit, branch road - branch road T - intersection road unit, branch road - branch road cross - intersection road unit, branch road - internal road T - intersection road unit, branch road - internal road cross - intersection road unit; 3 basic road units for internal roads: internal road straight road unit, internal road - internal road T - intersection road unit, internal road - internal road cross - intersection road unit.

[0072] In one or more embodiments, after obtaining the basic road unit set, the length of each road segment in the obtained basic road unit set is also revised to obtain standard basic road units. For example, it specifically includes:

[0073] (1) For the expressway straight road unit, the road length is revised according to the test process. In the test process, with 80 km / h as the maximum speed of the test section, according to the test vehicle accelerating from a standstill to 80 km / h at 1.85 m / s 2 and then traveling at a constant speed of 80 km / h for 10 s, and finally decelerating to 0 km / h at - 1.85 m / s 2 the road length is selected as 500 m.

[0074] (2) For the highway entrance ramp intersection road unit (merging into the highway), with 80 km / h as the maximum speed of the test section, according to the specification requirements, a 180 m speed change lane and a 70 m transition section are selected; the main line length before entering the confluence area is based on the road length required for the test vehicle to accelerate from a standstill to 80 km / h at 1.85 m / s 2 and is set to 135 m; the ramp length before the confluence area should meet the distance of 60 m for the clear sight triangle. Considering safety factors, a ramp length of 90 m is selected.

[0075] (3) For the highway exit ramp intersection road unit (exiting the highway), with 80 km / h as the maximum speed of the test section, according to the specification requirements, a 110 m speed change lane and an 80 m transition section are selected; the main line length before entering the transition section is set to 110 m according to the safety stopping sight distance requirements; the ramp length is set to 40 m according to the safety stopping sight distance requirements at a speed of 40 km / h; the main line length after diverging is set to 230 m considering the transition section, speed change lane and ramp length.

[0076] (4) For the highway exit ramp - arterial road intersection road unit (merging into the arterial road), with 60 km / h as the maximum speed of the test section, according to the specification requirements, the main line before the arterial road enters the confluence area meets the safety intersection stopping sight distance requirements and is set to 115 m; the ramp length is the same as that of the highway and is set to 90 m; the road length after merging meets the stopping sight distance requirements and is set to 75 m.

[0077] (5) For the highway entrance - arterial road intersection road unit (exiting the arterial road), with 60 km / h as the maximum speed of the test section, the arterial road length meets the stopping sight distance requirements and is set to 115 m; the ramp length is increased by a 30 m safety distance on the basis of meeting the 40 km / h stopping sight distance and is set to 70 m; the arterial road length after the ramp meets the stopping sight distance and is set to 75 m.

[0078] (6) For the arterial road straight road unit, with 60 km / h as the maximum speed of the test section, according to the test process of the test vehicle accelerating from a standstill to 60 km / h at 1.85 m / s 2 then traveling at a constant speed of 60 km / h for 10 s, and finally decelerating to 0 km / h at -1.85 m / s 2 , a road length of 320 m is selected.

[0079] (7) For the arterial road - arterial road T - intersection road unit, with 60 km / h as the maximum speed of each direction test section, according to the test vehicle accelerating from a standstill to 60 km / h at 1.85 m / s 2Accelerate to 60 km / h and then enter the intersection area at a constant speed of 60 km / h. The test process selects a road length of 110 m. At the same time, it meets the requirement of the safe stopping sight distance of 75 m in the intersection area.

[0080] (8) For the arterial - arterial cross - road unit, with 60 km / h as the maximum speed of each test section in all directions, according to the test vehicle starting from rest at 1.85 m / s 2 Accelerate to 60 km / h and then enter the intersection area at a constant speed of 60 km / h. The test process selects a road length of 110 m. At the same time, it meets the requirement of the safe stopping sight distance of 75 m in the intersection area.

[0081] (9) For the entrance road unit of the arterial - arterial connection road (merging into the arterial road), with 60 km / h as the maximum speed of the test section, the length of the main line before the confluence area meets the requirement of the safe cross - stopping sight distance and is set to 115 m; the length of the ramp is the same as that of the expressway and is set to 90 m; the length of the main line after the confluence area meets the requirement of the stopping sight distance and is set to 75 m.

[0082] (10) For the exit road unit of the arterial - arterial connection road (driving out of the arterial road), with 60 km / h as the maximum speed of the test section, the length of the arterial road meets the requirement of the stopping sight distance and is set to 115 m; the length of the ramp is increased by 30 m of safety distance on the basis of meeting the 40 km / h stopping sight distance and is set to 70 m; the length of the arterial road after the ramp meets the requirement of the stopping sight distance and is set to 75 m.

[0083] (11) For the arterial - branch T - shaped cross - road unit, with 60 km / h as the maximum speed of the arterial road and 40 km / h as the maximum speed of the branch road. According to the test vehicle starting from rest at 1.85 m / s 2 Accelerate to the maximum speed and then enter the intersection area at a constant speed of the maximum speed. The selected test road length of the arterial road is 110 m, and the test road length of the branch road is 60 m. At the same time, it meets the requirement of the safe stopping sight distance in the intersection area.

[0084] (12) For the arterial - branch cross - road unit, with 60 km / h as the maximum speed of the arterial road and 40 km / h as the maximum speed of the branch road. According to the test vehicle starting from rest at 1.85 m / s 2 Accelerate to the maximum speed and then enter the intersection area at a constant speed of the maximum speed. The selected test road length of the arterial road is 110 m, and the test road length of the branch road is 60 m. At the same time, it meets the requirement of the safe stopping sight distance in the intersection area.

[0085] (13) For the arterial - internal road T - shaped cross - road unit, with 60 km / h as the maximum speed of the arterial road and 20 km / h as the maximum speed of the internal road. According to the test vehicle starting from rest at 1.85 m / s 2The test process of accelerating to the maximum speed and then entering the intersection area at a constant speed at the maximum speed. The test road length of the arterial road is selected as 110 m, and the test road length of the internal road is 22 m. At the same time, the safety stopping sight distance requirements of the intersection area are met.

[0086] (14) For the arterial road - internal road cross - road unit, the maximum speed of the arterial road is 60 km / h, and the maximum speed of the internal road is 20 km / h. According to the test vehicle starting from rest and accelerating at 1.85 m / s 2 The test process of accelerating to the maximum speed and then entering the intersection area at a constant speed at the maximum speed. The test road length of the arterial road is selected as 110 m, and the test road length of the internal road is 22 m. At the same time, the safety stopping sight distance requirements of the intersection area are met.

[0087] (15) For the branch - straight road unit, the maximum speed of the test section is 40 km / h. According to the test vehicle starting from rest and accelerating at 1.85 m / s 2 To 40 km / h, then driving at a constant speed of 40 km / h for 10 s, and finally decelerating to 0 km / h. The selected road length is 190 m.

[0088] (16) For the branch - branch T - intersection road unit, the maximum speed of each direction of the test section is 40 km / h. According to the test vehicle starting from rest and accelerating at 1.85 m / s 2 Accelerating to 40 km / h, then entering the intersection area at a constant speed of 40 km / h. The selected road length is 60 m. At the same time, the safety stopping sight distance requirements of the intersection area are met.

[0089] (17) For the branch - branch cross - road unit, the maximum speed of each direction of the test section is 40 km / h. According to the test vehicle starting from rest and accelerating at 1.85 m / s 2 Accelerating to 40 km / h, then entering the intersection area at a constant speed of 40 km / h. The selected road length is 60 m. At the same time, the safety stopping sight distance requirements of the intersection area are met.

[0090] (18) For the branch - internal road T - intersection road unit, the maximum speed of the branch road is 40 km / h, and the maximum speed of the internal road is 20 km / h. According to the test vehicle starting from rest and accelerating at 1.85 m / s 2 Accelerating to the maximum speed, then entering the intersection area at a constant speed at the maximum speed. The selected test road length of the branch road is 60 m, and the test road length of the internal road is 22 m. At the same time, the safety stopping sight distance requirements of the intersection area are met.

[0091] (19) For the branch - internal road cross - road unit, the maximum speed of the branch road is 40 km / h, and the maximum speed of the internal road is 20 km / h. According to the test vehicle starting from rest and accelerating at 1.85 m / s2 Accelerate to the maximum speed, and then enter the intersection area at a constant speed of the maximum speed. The length of the test road for the branch road is selected as 60 m, and the length of the test road for the internal road is 22 m. At the same time, the safety stopping sight distance requirements for the intersection area are met.

[0092] (20) For the straight road unit of the internal road, with 20 km / h as the maximum speed of the test section, according to the test vehicle starting from a standstill at 1.85 m / s 2 Accelerate to 20 km / h, then drive at a constant speed of 20 km / h for 10 s, and finally decelerate to 0 km / h. The length of the selected road is 85 m.

[0093] (21) For the internal road - internal road T - shaped intersection road unit, with 20 km / h as the maximum speed of the internal road. According to the test vehicle starting from a standstill at 1.85 m / s 2 Accelerate to the maximum speed, and then enter the intersection area at a constant speed of the maximum speed. The length of the selected test road is 22 m. At the same time, the safety stopping sight distance requirements for the intersection area are met.

[0094] (22) For the internal road - internal road cross - shaped intersection road unit, with 20 km / h as the maximum speed of the internal road. According to the test vehicle starting from a standstill at 1.85 m / s 2 Accelerate to the maximum speed, and then enter the intersection area at a constant speed of the maximum speed. The length of the selected test road is 22 m. At the same time, the safety stopping sight distance requirements for the intersection area are met.

[0095] Since the road width is relatively small compared to the road length and has little impact on road deployment. Therefore, the road width is not considered in the study of road deployment for the time being.

[0096] The revised results are shown in Table 1:

[0097] Table 1

[0098]

[0099]

[0100]

[0101] S105. Determine the importance index of each basic road unit in the basic road unit set. According to the order from high to low of the importance index, use the optimization combination algorithm to combine the basic road units to obtain the road environment of the test site.

[0102] Optionally, the optimization combination algorithm can adopt the genetic algorithm.

[0103] In one or more embodiments, an initial individual set of basic road units is obtained in the order from high to low of the importance index;

[0104] Based on the size of the test site, the initial individual set of basic road units is optimized and deployed by using an optimization combination algorithm to obtain the road environment of the test site.

[0105] In this embodiment, the greater the length of the basic road unit, the greater the impact on the site. Therefore, for different types of basic road units, the importance index from high to low is: highway, arterial road, branch road, internal road.

[0106] Meanwhile, the importance index of the basic road unit is determined according to the number of path conflicts existing in different basic road units. The more path conflicts may exist, the higher the importance of the basic road unit. The number of test scenarios that different basic road units in the basic road unit set can support: there are no less than 30 path conflicts in the crossroads, no less than 10 path conflicts in the T-junction, 4 path conflicts in the main road ramp merging road, 3 path conflicts in the main road ramp exiting road, and 3 path conflicts in the straight road. Therefore, in the embodiment, for the same type of basic road unit, the importance index from high to low is: crossroads, T-junction, main road - ramp merging road, main road - ramp exiting road, straight road.

[0107] For the 22 basic road units obtained by clustering, their importance index from high to low is: highway entrance ramp crossroad unit (merging into the highway), highway exit - arterial road crossroad unit (merging into the arterial road), highway straight road unit, highway exit ramp crossroad unit (exiting the highway), highway entrance - arterial road crossroad unit (exiting the arterial road), arterial road - arterial road crossroads unit, arterial road - arterial road T-junction unit, arterial road - arterial road connecting line entrance road unit (merging into the arterial road), arterial road straight road unit, arterial road - arterial road connecting line exit road unit (exiting the arterial road), arterial road - branch road crossroads unit, arterial road - branch road T-junction unit, arterial road - internal road crossroads unit, arterial road - internal road T-junction unit, branch road - branch road crossroads unit, branch road - branch road T-junction unit, branch road straight road unit, branch road - internal road crossroads unit, branch road - internal road T-junction unit, internal road - internal road crossroads unit, internal road - internal road T-junction unit, internal road straight road unit.

[0108] In one or more embodiments, a genetic algorithm is used to optimize the deployment of the initial individual set of basic road units to obtain the road environment of the test site, specifically including:

[0109] Use the genetic algorithm to perform the first optimization on the initial individual set of basic road units to generate multiple groups of initial basic road units similar to those obtained by clustering the basic road unit set for further optimization and combination;

[0110] Perform the second optimization connection on the initial basic road units to combine the basic road units to obtain the optimal road network;

[0111] Lay out the optimal road network in the test site to obtain the road environment of the test site.

[0112] For example, determine the geometric parameters of the preset test site. In this embodiment, the preset autonomous driving test site is a rectangular site with a length of 1600 meters and a width of 600 meters, and it is required to include a highway test area, an urban main road test area, an urban branch road test area, and an internal road test area.

[0113] In this embodiment, the genetic algorithm is used as the optimization deployment algorithm for the road environment of the test site. The convergence of the genetic algorithm is beneficial to obtaining the optimization deployment result. In the order of the importance of the basic road units, the basic road units are combined and connected in sequence. Through the iterative combination of the genetic algorithm, the optimal result is selected as the road environment within the site. Adjust the optimization result to make the road environment within the site more suitable for actual construction without changing the road form.

[0114] Exemplarily, Figure 2 is the flowchart of the optimization deployment of the road environment of the test site using the genetic algorithm provided by the embodiment of the present invention. As Figure 2 shown, in this embodiment, the genetic algorithm is used for two iterative optimizations. In the first iterative optimization, a basic road unit set that conforms to the standard basic road units obtained through clustering and revision is generated, including 22 independent basic road units, satisfying the section lengths and the mutual positional relationships between sections similar to those of the standard basic road units obtained by clustering. In the second iterative optimization, the optimal combined road network that meets the constraint conditions is generated. The road network is formed by combining and connecting 22 basic road units.

[0115] In the embodiment, the constraint conditions for the optimization deployment using the genetic algorithm are:

[0116] (1) To ensure that the road network conforms to the typical road units obtained by clustering, the deviation of the connection line length between the endpoints of each basic road unit in the optimization from the connection line length between the corresponding endpoints of the standard basic road unit does not exceed 10%;

[0117] (2) The basic road units and the connecting lines forming the road network cannot overlap or cross.

[0118] (3) The area occupied by the optimized generated road network cannot exceed the preset site range.

[0119] The genetic algorithm is used to optimize the deployment of the road environment of the test site, including the following steps:

[0120] Step S301: In implementation, initialize the optimization parameters, including but not limited to the number of individuals N1 included in the population in the first optimization iteration, the number of individuals N2 included in the population in the second optimization iteration, the crossover rate α1 in the first optimization iteration, the mutation rate β1 in the first optimization iteration, the number of times n1 of the first optimization iteration, the crossover rate α2 in the second optimization iteration, the mutation rate β2 in the second optimization iteration, and the number of times n2 of the second optimization iteration.

[0121] Step S302: In this embodiment, prepare for the first optimization and generate an initial set of basic road units S1(0) contains N1 individuals. Each individual P i generates 22 basic road units during initialization as the 22 chromosomes of the individuals in the genetic algorithm. Each basic road unit is described by its center point coordinates, endpoint coordinates, and the expressions of each road section in the unit. In the first optimization, the center point coordinates of each basic road unit are all (0, 0).

[0122] Step S303: In this embodiment, verify whether the number of iterations of the first optimization reaches the preset number of iterations n1. If the preset number of iterations n1 is reached, select the optimal N2 individuals from the set of basic road units S1(n1) obtained from the last round of iteration and output them as the optimal set of basic road units; if the preset number of iterations n1 is not reached, perform optimization.

[0123] Step S304: In this embodiment, randomly cross the individuals in the set of basic road unit individuals S1(t) according to the crossover rate α1. During the crossover, two individuals exchange their respective partial chromosomes, that is, two random individuals P i and P j randomly exchange some basic road units between them to generate new individuals. The newly generated individuals are merged with the original set of basic road unit individuals S1(t) into a set of basic road units S′1(t).

[0124] Step S305: In this embodiment, individuals in the basic road unit set S′1(t) are randomly mutated according to the mutation rate β1. During the mutation, the segment lengths and relative positional relationships between segments of the basic road units represented by some chromosomes of the individual to be mutated are randomly changed. The newly generated individuals are merged with the original basic road unit set S′1(t) to form the basic road unit set S″1(t).

[0125] Step S306: Calculate the fitness Fitness of all individuals in the basic road unit set S″1(t) after crossover and mutation u . In the embodiment, the fitness Fitness of a single basic road unit e is defined as the sum of the percentages of the errors between the lengths d of the connections between pairwise different endpoints in the basic road unit represented by each chromosome ij and the lengths d of the connections between the corresponding endpoints in the standard basic road unit ij . The fitness Fitness of an individual in the basic road unit set u is the sum of the fitness Fitness of all basic road units therein e . The smaller the Fitness e , the closer the single basic road unit is to the standard basic road unit obtained by clustering. The smaller the Fitness u , the closer the individual is to the 22 standard basic road units obtained by clustering.

[0126] Among them,

[0127]

[0128] Fitness u = ∑ Fitness e

[0129] Step S307: In this embodiment, select the N1 individuals with the smallest fitness value Fitness in the basic road unit set S″1(t) after crossover and mutation to form a new generation of basic road unit set S1(t + 1). u

[0130] Step S308: This round of iteration is completed, go to step S303 to determine whether the iteration is completed.

[0131] Step S309: The first iteration optimization is completed. In the embodiment, from the basic road unit set S1(n1) obtained in the last round of iteration, select the optimal N2 individuals and output them as the optimal basic road unit set S2(0) for the second iteration optimization.

[0132] Step S310: In this embodiment, the 22 basic road units in each individual in the optimal basic road unit set are combined in sequence according to the obtained importance index to form an initial road network. During the combination process, the endpoints of the same type in the basic road units can be connected, while the endpoints of different types cannot be connected. For example, the endpoints of the arterial road can be connected to the endpoints of the arterial road; the endpoints of the highway connection line cannot be connected to the endpoints of the branch road. In it, the 22 basic road units are combined to form an initial road network. During the combination process, the endpoints of the same type in the basic road units can be connected, while the endpoints of different types cannot be connected. For example, the endpoints of the arterial road can be connected to the endpoints of the arterial road; the endpoints of the highway connection line cannot be connected to the endpoints of the branch road.

[0133] Each individual Q in the optimal basic road unit set i represents a road network obtained by a combination. Each chromosome q in the individual i represents a basic road unit in the road network. The individuals in these basic road unit sets are interconnected to form a road network S2(0).

[0134] Step S311: In the embodiment, verify whether the number of iterations of the second optimization reaches the preset number of iterations n2. If the preset number of iterations n2 is reached, select the optimal individual from the road network set S2(n2) obtained in the last round of iteration and output it as the optimal road network; if the preset number of iterations n2 is not reached, perform the optimization.

[0135] Step S312: In the embodiment, randomly pair-wise cross the individuals in the road network set according to the crossover rate α2. During the crossover, the two individuals exchange their respective partial chromosomes, that is, randomly exchange some basic road units between two random individuals Q i and Q j . After the crossover, the basic road units represented by the chromosomes are recombined into a road network to form new road network individuals. The newly generated individuals are merged with the original road network set S2(t) into a road network set S′2(t).

[0136] Step S313: In the embodiment, randomly mutate the individuals in the road network set S′2(t) according to the mutation rate β2. During the mutation, randomly change the road segment lengths and the relative positional relationships between road segments of the basic road units represented by the partial chromosomes of the individual to be mutated. After the mutation, the basic road units represented by the chromosomes are recombined into a road network to form new road network individuals. The newly generated individuals are merged with the original road network set S′2(t) into a road network set S′2′(t).

[0137] Step S314: Calculate the fitness of all individuals in the road network set after crossover and mutation. In this embodiment, the fitness Fitness D of the individuals in the road network set is defined as the sum of the lengths of the connection lines between the basic road units in the combined road network. The fitness Fitness DThe smaller it is, the more compact the deployment of basic road units in the road network is, which can reduce the area occupied by the road network.

[0138] At the same time, in order to meet the optimization constraints, the fitness value of the individual in the road network set should reflect whether there is overlap or intersection between the basic road units and their connections in the road network. If there is overlap or intersection, the fitness value is D The value is superimposed with a larger value of 1000. The smaller the fitness value of the road network, the more compact the combined road network is, and there is no overlap of basic road units and road intersections.

[0139] Step S315: In the embodiment, select the fitness value Fitness in the road network set S″2(t) after crossover and mutation D The smallest N2 individuals form the new generation road network set S2(t+1).

[0140] Step S316: This round of iteration is completed, and the process goes to step S311 to determine whether the iteration is completed.

[0141] Step S317: The second iteration optimization is completed. In this embodiment, Fitness is selected from the road network set obtained from the last iteration. D The smallest road network is the optimal individual output as the road network. The result can be manually adjusted according to the actual test site to achieve the construction of the closed test site road environment.

[0142] Figure 4 A schematic diagram of a test site road structure obtained based on genetic algorithm optimization deployment provided in an embodiment of the present invention; Figure 5 The test site road environment map obtained based on the genetic algorithm optimization deployment provided by the embodiment of the present invention. Please refer to Figure 4 and Figure 5 As shown, it can be seen that the test site road environment map obtained according to the method of this embodiment is close to the actual road structure. The test site road environment map includes a variety of basic road units, which is conducive to the testing of autonomous driving vehicles.

[0143] The following is a description of the autonomous driving test site road environment construction device provided by the present invention. The autonomous driving test site road environment construction device described below and the autonomous driving test site road environment construction method described above can be referenced to each other.

[0144] Figure 6 For a schematic diagram of the structure of the road environment construction device for the autonomous driving test site provided by the embodiment of the present invention, please refer to Figure 6 As shown, the device 60 includes:

[0145] An acquisition unit 601, configured to acquire real road map data;

[0146] A first processing unit 602, configured to acquire road nodes and road segments included in the real road map data, as well as the subordinate relationships between the road nodes and the road segments;

[0147] A second processing unit 603, configured to determine the type attributes of the road nodes and the type attributes of the road segments, and construct a typical road structure classification set according to the type attributes of the road nodes and the type attributes of the road segments;

[0148] A clustering unit 604, configured to cluster the real road map data based on the typical road structure classification set to obtain a set of basic road units included in the real road map data;

[0149] A third processing unit 605, configured to determine the importance index of each basic road unit in the set of basic road units, and combine the basic road units by using an optimization combination algorithm according to the order from high to low of the importance index to obtain a road environment of a test site.

[0150] In one or more embodiments, the device further includes a user interaction module, which is configured for a user to interact with a device for constructing a road environment of an autonomous driving test site based on real map data. The user can select a real map range in the real map module, modify the parameters of the basic road units, select a subset of typical basic road units and an optimization algorithm, and customize the optimization control parameters through a user interface. The user interface receives the real-time operation status feedback and the display of the final optimization result fed back by the optimization operation module.

[0151] This embodiment also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements any one of the methods for constructing a road environment of an autonomous driving test site.

[0152] This embodiment also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements any one of the above methods for constructing a road environment of an autonomous driving test site.

[0153] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0154] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing a road environment of an autonomous driving test site, characterized in that, Including: Obtaining real road map data; Obtaining road nodes and road segments included in the real road map data, as well as the subordinate relationships of the road nodes and road segments; Determining the type attributes of the road nodes and the type attributes of the road segments, and constructing a typical road structure classification set according to the type attributes of the road nodes and the type attributes of the road segments; Based on the typical road structure classification set, clustering the real road map data to obtain a basic road unit set included in the real road map data; Determining the importance index of each basic road unit in the basic road unit set, and combining the basic road units by using an optimization combination algorithm according to the order from high to low of the importance index to obtain a test site road environment; Among them, the step of combining the basic road units by using an optimization combination algorithm according to the order from high to low of the importance index to obtain a test site road environment includes: Obtaining an initial individual set of the basic road unit set according to the order from high to low of the importance index; based on the size of the test site, using a genetic algorithm to perform the first optimization on the individuals of the initial basic road unit set, generating multiple groups of initial basic road units similar to the basic road unit set obtained by clustering, performing the second optimization on the initial basic road units, optimizing and connecting the initial basic road units to obtain an optimal road network, and laying out the optimal road network in the test site to obtain the test site road environment; Among them, the step of determining the importance index of each basic road unit in the basic road unit set includes: For each basic road unit, determining the importance index of the basic road unit according to the number of path conflicts existing in the basic road unit, where the more path conflicts exist, the higher the importance index of the basic road unit.

2. The method for constructing a road environment of an autonomous driving test site according to claim 1, wherein, The type attributes of the road nodes include: straight road structure boundary points, T-shaped intersection road structure center points, T-shaped intersection road structure boundary points, cross intersection road structure center points, cross intersection road structure boundary points, main road ramp intersection road structure center points, and main road ramp intersection road structure boundary points; The type attributes of the road segments include: expressways, expressway connecting lines, main roads, main road connecting lines, branch roads, and internal roads; The typical road structure classification set includes straight roads, T-shaped intersection roads, cross intersection roads, and main road-ramp intersection roads.

3. The method for constructing a road environment of an autonomous driving test site according to claim 2, wherein The step of constructing a typical road structure classification set according to the type attributes of the road nodes and the type attributes of the road segments includes: For each road segment, determining the typical road structure formed by the road segment and the corresponding road node according to the type attribute of the road segment and the type attribute of the road node of the road segment; Constructing the typical road structure classification set according to multiple different typical road structures.

4. The method for constructing a road environment of an autonomous driving test site according to claim 1, wherein The step of clustering the real road map data based on the typical road structure classification set to obtain a basic road unit set included in the real road map data includes: Cluster the real road map data according to the typical road structures included in the typical road structure classification set to obtain multiple basic road units included in the real road map data; Obtain the basic road unit set according to the multiple basic road units.

5. An automatic driving test site road environment construction device, characterized in that, It includes: An acquisition unit for acquiring real road map data; A first processing unit for acquiring road nodes and road segments included in the real road map data, as well as the subordinate relationships of the road nodes and road segments; A second processing unit for determining the type attributes of the road nodes and the type attributes of the road segments, and constructing a typical road structure classification set according to the type attributes of the road nodes and the type attributes of the road segments; A clustering unit for clustering the real road map data based on the typical road structure classification set to obtain a set of basic road units included in the real road map data; A third processing unit for determining the importance index of each basic road unit in the basic road unit set, and combining the basic road units using an optimization combination algorithm according to the order from high to low of the importance index to obtain a test site road environment; Among them, the combining the basic road units using an optimization combination algorithm according to the order from high to low of the importance index to obtain a test site road environment includes: Obtaining an individual set of an initial basic road unit set according to the order from high to low of the importance index; based on the size of the test site, using a genetic algorithm to perform a first optimization on the individuals of the initial basic road unit set, generating multiple groups of initial basic road units similar to the basic road unit set obtained by clustering, performing a second optimization on the initial basic road units, optimizing the connection of the initial basic road units to obtain an optimal road network, and laying out the optimal road network in the test site to obtain the test site road environment; Among them, the determining the importance index of each basic road unit in the basic road unit set includes: For each of the basic road units, determining the importance index of the basic road unit according to the number of path conflicts existing in the basic road unit, where the more path conflicts exist, the higher the importance index of the basic road unit.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method for constructing an autonomous driving test site road environment according to any one of claims 1 to 4.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for constructing an autonomous driving test site road environment according to any one of claims 1 to 4.

Images