Method and system for checking high-precision map data based on path matching

Abstract

The invention discloses a high-precision map data inspection method and system based on path matching, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining the road network data of an actual road, and generating a road navigation path; mapping a shape point in the road navigation path to the acquired high-precision map data as a starting point, and acquiring a starting lane within a preset range of the starting point; matching and updating the continuous lane according to the starting lane in the forward direction to obtain a cut-off lane; the previous lane is updated through reverse backtracking of the cut-off lane; forming a lane-level navigation path of high-precision map data by taking the preorder lane meeting the preset judgment condition as a target lane; and checking based on the stop lane and the end point on the road navigation path to obtain a checking result of the high-precision map data. According to the method, the generation of the road navigation path and the topology inspection of the high-precision map data can be realized in a multi-thread and batch manner, and the inspection and detection efficiency of the high-precision map data is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and more specifically, to a method and system for verifying high-precision map data based on path matching. Background Art

[0002] High-precision maps are like a huge information container, covering lane information, such as slope, curvature, heading, marking type, road markings, road restriction information, etc., as well as information about objects around the lanes, such as traffic signs, traffic lights, lane height limits, obstacles, overhead objects, guardrails, road edge types, roadside facilities, etc. These data play a decisive role in the perception, decision-making and planning control of autonomous vehicles at level L3 and above. High-precision maps can provide super-perception distance and surrounding information, expand the boundaries of perception, provide perception redundancy, and improve perception efficiency. During use, the vehicle's decision-making system refers to the data of high-precision maps for path planning. This process requires high-precision maps to provide lane-level navigation and corresponding lane topology. However, the connection problems caused by missing data collection and data production errors have a great impact on the decision-making of the meeting end. Before the data is officially released to the vehicle end, it will be safer and more efficient to verify the map data in a simulation environment.

[0003] The reliance of autonomous vehicles on high-precision maps during operation is not only reflected in the description of routes in traditional maps, but also in the provision of lane-level "navigation" information. Users need to call the high-precision map data of the current vehicle location in real time to make path decisions and plans. Its essence is to make more detailed lane-level navigation derivations based on road-level navigation and select drivable lane-level topological data. Therefore, navigation attributes or descriptions of drivable area attributes are crucial to autonomous driving. An effective data verification solution helps to detect errors in advance and improve the reliability of autonomous driving.

[0004] The prior art discloses a method and device for verifying a high-precision map, including: determining a path start point and a path end point in a high-precision map; obtaining a first driving route from the path start point to the path end point based on a driving trajectory library, and determining the number of types of the first driving route; obtaining a second driving route from the path start point to the path end point based on road information in the high-precision map; determining the first driving route that is the same as the second driving route as a third driving route, and determining the number of types of the third driving route; verifying a high-precision map area including the path start point and the path end point based on the number of types of the first driving route and the number of types of the third driving route. The prior art is based on the trajectory that has been traveled, and is a posteriori method; and it is difficult to collect enough trajectories to verify all roads, and the detection efficiency of high-precision map information is low. Summary of the invention

[0005] In order to overcome the defect of low efficiency of the above-mentioned prior art when verifying high-precision map data, the present invention provides a method and system for verifying high-precision map data based on path matching, which can realize batch verification of high-precision map data and improve detection efficiency.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0007] The present invention provides a method for verifying high-precision map data based on path matching, comprising:

[0008] Acquiring road network data of an actual road, and generating a road navigation path based on the road network data of the actual road;

[0009] Acquire high-precision map data, map the shape point in the road navigation path onto the high-precision map data as a starting point, and acquire a starting lane within a preset range of the starting point;

[0010] Advance in the forward direction from the starting lane along the road navigation path, match and update the following lane; preset an advancement cut-off condition, and use the following lane that meets the preset advancement cut-off condition as the cut-off lane;

[0011] Backtracking from the cut-off lane along the road navigation path in reverse order to update the preceding lane; presetting a discrimination condition, taking the preceding lane that meets the preset discrimination condition as the target lane; and using the target lane to form a lane-level navigation path of high-precision map data;

[0012] An inspection is performed based on the cut-off lane and the end point on the road navigation path to obtain an inspection result of the high-precision map data.

[0013] Preferably, the specific method of acquiring the road network data of the actual road and generating the road navigation path based on the road network data of the actual road includes:

[0014] Based on the public map data, the road network data of the actual roads in any area are selected as the road network dataset;

[0015] Marking road junction points in the road network dataset to form a road junction point dataset;

[0016] Determine the road grade to be tested, and extract road network data corresponding to the road grade to be tested from the road network data set as the trunk road network data set;

[0017] In the road junction point data set, the road junction points corresponding to the roads in the trunk road network data set are retained and divided into trunk road internal junction points and trunk road breakpoint junction points to form a trunk road internal junction point set and a trunk road breakpoint junction point set, and the trunk road breakpoint junction points are given direction attributes;

[0018] The set of breakpoints and junctions of the trunk roads is traversed to generate a road navigation path.

[0019] Preferably, the specific method of assigning direction attributes to the breakpoint junction of the trunk road is:

[0020] The road network data of the actual road is stored according to the actual road driving direction. When the actual road driving direction is toward the main road network dataset, the direction attribute of the main road breakpoint junction is the positive direction; when the actual road driving direction deviates from the main road network dataset, the direction attribute of the main road breakpoint junction is the reverse direction.

[0021] Preferably, the specific method of traversing the set of breakpoints and junctions of the trunk roads to generate a road navigation path is:

[0022] Select a main road breakpoint junction point from the main road breakpoint junction point set, and extend it in the road network data set in the forward and reverse directions of its direction attributes; when encountering a junction point inside the main road during the extension process, perform a route expansion;

[0023] Calculate the length of the currently expanded route and compare the length of the currently expanded route with the preset route length threshold; if the length of the currently expanded route is greater than the preset route length threshold, end the route expansion and use the currently expanded route as an extended route; otherwise, continue the route expansion;

[0024] The set of breakpoints and junctions of the trunk roads is traversed, and the above steps are repeated to obtain a plurality of extended routes, and all the extended routes are combined into a road navigation path.

[0025] Preferably, the specific method of advancing from the starting lane along the road navigation path in the forward direction and matching and updating the subsequent lane; presetting the advancing cut-off condition and taking the subsequent lane that meets the preset advancing cut-off condition as the cut-off lane is:

[0026] S101: adding a starting lane within the preset range of the starting point into a starting lane set;

[0027] S102: setting a first matching condition and a second matching condition, matching the starting lane with the road navigation path, and taking the starting lane that satisfies both the first matching condition and the second matching condition as a candidate lane, and adding it to a candidate lane set;

[0028] S103: adding the candidate lanes into the cache lane set, advancing from the starting point along the road navigation path in a forward direction, traversing the candidate lanes in the candidate lane set, and obtaining the subsequent lane of each candidate lane;

[0029] S104: taking each connecting lane as the starting lane for the next round of matching, returning to step S102, updating the connecting lanes until no connecting lanes that meet both the first matching condition and the second matching condition can be obtained; and adding all connecting lanes that meet both the first matching condition and the second matching condition into the connecting lane set;

[0030] S105: After replacing the candidate lane set with the connecting lane set, the connecting lane set is cleared and the process returns to step S103;

[0031] S106: Preset the advancement cut-off condition, determine whether the current connecting lane meets the preset advancement cut-off condition; if the current connecting lane meets the preset advancement cut-off condition, end the advancement, and use the current connecting lane that meets the preset advancement cut-off condition as the cut-off lane; otherwise, continue the advancement.

[0032] Preferably, the first matching condition, the second matching condition and the preset advancement cut-off condition are specifically:

[0033] The first matching condition is direction matching, and the difference between the lane direction of the starting lane and the road driving direction of the road navigation path is calculated. When the difference is within a preset direction difference threshold range, the corresponding starting lane meets the first matching condition; otherwise, the corresponding starting lane does not meet the first matching condition;

[0034] The second matching condition is distance matching, and the Hausdorff distance between the lane direction of the starting lane and the road navigation path is calculated. When the Hausdorff distance is within a preset distance threshold range, the corresponding starting lane meets the second matching condition; otherwise, the corresponding starting lane does not meet the second matching condition;

[0035] The preset advancement cut-off condition is: for the current connecting lane, if the connecting lane of the current connecting lane cannot be obtained or the current connecting lane has covered the end point of the road navigation path, then the current connecting lane satisfies the preset advancement cut-off condition; otherwise, the current connecting lane does not satisfy the preset advancement cut-off condition.

[0036] Preferably, the preceding lane is updated by tracing back from the cut-off lane along the road navigation path in reverse; a discrimination condition is preset, and the preceding lane that meets the preset discrimination condition is used as the target lane; and the specific method of using the target lane to form a lane-level navigation path of high-precision map data is as follows:

[0037] S201: Tracing back from the cut-off lane along the road navigation path in reverse order to traverse and obtain the preceding lane of the cut-off lane;

[0038] S202: Preset a judgment condition to determine whether the preceding lane satisfies the preset judgment condition; if so, save the preceding lane as a target lane into the target lane set and the backtracking lane set, and execute step S203; otherwise, execute step S204;

[0039] S203: taking the previous lane in the backtracking lane set as the new cutoff lane, returning to step S201, and updating the previous lane;

[0040] S204: taking the starting lane in the starting lane set as the new ending lane, returning to step S201, and updating the preceding lane;

[0041] S205: Preset a backtracking cutoff condition, and determine whether the current preceding lane meets the preset backtracking cutoff condition; if the current preceding lane meets the backtracking cutoff condition, then end the backtracking and execute step S206; otherwise, continue the backtracking;

[0042] S206: All target lanes stored in the target lane set are combined into a lane-level navigation path of the high-precision map data.

[0043] Preferably, the preset discrimination condition and the preset backtracking cut-off condition are specifically:

[0044] The preset judgment condition is: check whether the preceding lane exists in the cache lane set; if so, the preceding lane satisfies the preset judgment condition; otherwise, the preceding lane does not satisfy the preset judgment condition;

[0045] The preset backtracking cutoff condition is: for the current preceding lane, if the preceding lane of the current preceding lane cannot be obtained or the current preceding lane exists in the starting lane set, then the current preceding lane satisfies the preset backtracking cutoff condition; otherwise, the current preceding lane does not satisfy the preset backtracking cutoff condition.

[0046] Preferably, the specific method for obtaining the inspection result of the high-precision map data based on the inspection of the cut-off lane and the end point on the road navigation path is:

[0047] Matching the cut-off lane with an end point on the road navigation path;

[0048] If the position of the cut-off lane matches the end position on the road navigation path, the inspection result of the high-precision map data is marked as normal;

[0049] If the position of the cut-off lane does not match the end position on the road navigation path, the inspection result of the high-precision map data is marked as an error.

[0050] The present invention also provides a high-precision map data inspection system based on path matching, which is used to implement the above inspection method, including:

[0051] A road navigation path generation module, used to obtain road network data of an actual road, and generate a road navigation path based on the road network data of the actual road;

[0052] A high-precision map data acquisition module is used to acquire high-precision map data, map the shape point in the road navigation path onto the high-precision map data as a starting point, and acquire a starting lane within a preset range of the starting point;

[0053] A following lane advancing module is used to advance from the starting lane along the road navigation path in a forward direction, match and update the following lane; preset an advancing cut-off condition, and use the following lane that meets the preset advancing cut-off condition as the cut-off lane;

[0054] A preceding lane backtracking module is used to backtrack from the cut-off lane along the road navigation path in reverse order to update the preceding lane; preset a discrimination condition, and use the preceding lane that meets the preset discrimination condition as the target lane; and use the target lane to form a lane-level navigation path of high-precision map data;

[0055] The high-precision map data verification module is used to perform a verification based on the cut-off lane and the end point on the road navigation path to obtain a verification result of the high-precision map data.

[0056] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0057] The present invention generates a road navigation path based on the road network data of the actual road, maps the shape points in the road navigation path to the acquired high-precision map data as the starting point, and uses the road navigation path to match the high-precision map data to check the topological connectivity of the high-precision map data; first, the starting lane within the preset range of the starting point is advanced in the forward direction along the road navigation path, and the subsequent lane is matched and updated until the preset advancement cut-off condition is met, and the forward cut-off lane is obtained; then, the cut-off lane is reversely traced back along the road navigation path, the preceding lane is updated, and the preceding lane that meets the preset discrimination condition is used as the target lane to form a lane-level navigation path; finally, based on the cut-off lane and the end point on the road navigation path, the inspection result of the high-precision map data is obtained, which effectively reflects the path accessibility status of the high-precision map data and the scope of data coverage. In the present invention, the generation of the road navigation path and the topological inspection of the high-precision map data can be multi-threaded and batch processed, which improves the detection efficiency of the high-precision map data; and the high-precision map data is matched in the forward and reverse directions based on the road navigation path, which improves the inspection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flow chart of a method for verifying high-precision map data based on path matching as described in Example 1.

[0059] Figure 2 This is a flowchart of generating a road navigation path as described in Example 2.

[0060] Figure 3 This is a schematic diagram of the locations of different types of road junctions described in Example 2.

[0061] Figure 4 This is a schematic diagram of the matching target lane described in Example 2.

[0062] Figure 5 This is a schematic diagram of the structure of a high-precision map data inspection system based on path matching as described in Example 3. DETAILED DESCRIPTION

[0063] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0064] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0065] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0066] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0067] Example 1

[0068] This embodiment provides a method for verifying high-precision map data based on path matching, such as Figure 1 As shown, including:

[0069] S1: Acquire road network data of an actual road, and generate a road navigation path based on the road network data of the actual road;

[0070] S2: obtaining high-precision map data, mapping the shape point in the road navigation path onto the high-precision map data as a starting point, and obtaining a starting lane within a preset range of the starting point;

[0071] S3: advancing in the forward direction from the starting lane along the road navigation path, matching and updating the following lane; presetting a advancing cut-off condition, and taking the following lane that meets the preset advancing cut-off condition as the cutting-off lane;

[0072] S4: trace back from the cut-off lane along the road navigation path in reverse order to update the preceding lane; preset a discrimination condition, and use the preceding lane that meets the preset discrimination condition as the target lane; and use the target lane to form a lane-level navigation path of high-precision map data;

[0073] S5: Performing an inspection based on the cut-off lane and the end point on the road navigation path to obtain an inspection result of the high-precision map data.

[0074] In the specific implementation process, this embodiment generates a road navigation path based on the road network data of the actual road, maps the shape points in the road navigation path to the acquired high-precision map data as the starting point, and uses the road navigation path to match the high-precision map data to check the topological connectivity of the high-precision map data; first, the starting lane within the preset range of the starting point is advanced along the road navigation path, and the subsequent lane is matched and updated until the preset advancement cut-off condition is met to obtain the forward cut-off lane; then, the cut-off lane is reversely traced along the road navigation path, and the previous lane is updated, and the previous lane that meets the preset judgment condition is used as the target lane to form a lane-level navigation path; finally, based on the cut-off lane and the end point on the road navigation path, the inspection result of the high-precision map data is obtained, which effectively reflects the path accessibility status of the high-precision map data and the scope of data coverage. In this embodiment, the generation of the road navigation path and the topological inspection of the high-precision map data can be multi-threaded and batch processed, which improves the detection efficiency of the high-precision map data; and the high-precision map data is matched forward and reversely on the basis of the road navigation path, and the inspection accuracy is higher.

[0075] Example 2

[0076] This embodiment provides a method for verifying high-precision map data based on path matching, including:

[0077] Acquire road network data of actual roads, and generate a road navigation path based on the road network data of actual roads; specifically:

[0078] like Figure 2 As shown, based on the public common navigation map or the open source OSM (OpenStreetMap, Open Street Map) data, the road network data of the actual roads in any area are selected as the road network data set; in this embodiment, a location is arbitrarily selected, and the road network data of the actual roads within a radius of 100 kilometers around it are used as the road network data set;

[0079] Different road relationships have different road junctions, such as divergence and confluence points between expressways and ramps, and road intersections between ordinary intersections. The road junctions are marked in the road network dataset to form a road junction dataset;

[0080] Determine the road grade to be tested, and extract road network data corresponding to the road grade to be tested from the road network data set as the trunk road network data set;

[0081] like Figure 3 As shown, in the road junction point data set, the road junction points corresponding to the roads in the trunk road network data set are retained and divided into trunk road internal junction points and trunk road breakpoint junction points. Figure 3 In the figure, the hollow points are the junction points of the main road breakpoints, and the solid points are the internal junction points of the main road, which form the internal junction point set of the main road and the junction point set of the main road breakpoints, and the direction attribute of the junction points of the main road breakpoints is given; the road network data of the actual road is stored according to the actual road driving direction. When the actual road driving direction is toward the main road road network dataset, the direction attribute of the main road breakpoint junction point is the positive direction; when the actual road driving direction deviates from the main road road network dataset, the direction attribute of the main road breakpoint junction point is the reverse direction;

[0082] Roads have different road grades, such as expressways, ramps, urban roads, rural roads, etc. If the inspection object is an expressway, the expressway is taken as the road grade to be inspected. Then the data other than the expressway is removed from the road network dataset, and the road junctions in the removed roads are also removed from the road junction point dataset.

[0083] Select a main road breakpoint junction point from the main road breakpoint junction point set, and extend it in the road network data set in the forward and reverse directions of its direction attributes; when encountering a junction point inside the main road during the extension process, perform a route expansion;

[0084] Calculate the length of the currently expanded route, and compare the length of the currently expanded route with the preset route length threshold; if the length of the currently expanded route is greater than the preset route length threshold, end the current route expansion and use the currently expanded route as an extended route; otherwise, continue the current route expansion; in this embodiment, the preset route length threshold is 10 kilometers.

[0085] The set of breakpoints and junctions of the trunk roads is traversed, and the above steps are repeated to obtain a plurality of extended routes, and all the extended routes are combined into a road navigation path.

[0086] Acquire high-precision map data, map the shape point in the road navigation path onto the high-precision map data as a starting point, and acquire a starting lane within a preset range of the starting point;

[0087] Advance in the forward direction from the starting lane along the road navigation path, match and update the subsequent lane; preset an advancement cut-off condition, and use the subsequent lane that meets the preset advancement cut-off condition as the cut-off lane; specifically:

[0088] S101: adding a starting lane within the preset range of the starting point into a starting lane set;

[0089] S102: setting a first matching condition and a second matching condition, matching the starting lane with the road navigation path, and taking the starting lane that satisfies both the first matching condition and the second matching condition as a candidate lane, and adding it to a candidate lane set;

[0090] S103: adding the candidate lanes into the cache lane set, advancing from the starting point along the road navigation path in a forward direction, traversing the candidate lanes in the candidate lane set, and obtaining the subsequent lane of each candidate lane;

[0091] S104: taking each connecting lane as the starting lane for the next round of matching, returning to step S102, updating the connecting lanes until no connecting lanes that meet both the first matching condition and the second matching condition can be obtained; and adding all connecting lanes that meet both the first matching condition and the second matching condition into the connecting lane set;

[0092] S105: After replacing the candidate lane set with the connecting lane set, the connecting lane set is cleared and the process returns to step S103;

[0093] S106: Preset the advancement cut-off condition, determine whether the current connecting lane meets the preset advancement cut-off condition; if the current connecting lane meets the preset advancement cut-off condition, end the advancement, and use the current connecting lane that meets the preset advancement cut-off condition as the cut-off lane; otherwise, continue the advancement.

[0094] In step S105, the lane data of all the following lanes in the current round are saved in the following lane set, and the final lane saved in the candidate lane set is the final lane, and the lane data of all the following lanes from the starting point to the final lane are finally saved in the cache lane set.

[0095] The first matching condition is direction matching, and the difference between the lane direction of the starting lane and the road driving direction of the road navigation path is calculated. When the difference is within a preset direction difference threshold range, the corresponding starting lane meets the first matching condition; otherwise, the corresponding starting lane does not meet the first matching condition;

[0096] The second matching condition is distance matching, and the Hausdorff distance between the lane direction of the starting lane and the road navigation path is calculated. When the Hausdorff distance is within a preset distance threshold range, the corresponding starting lane meets the second matching condition; otherwise, the corresponding starting lane does not meet the second matching condition;

[0097] The preset advancement cut-off condition is: for the current connecting lane, if the connecting lane of the current connecting lane cannot be obtained or the current connecting lane has covered the end point of the road navigation path, then the current connecting lane satisfies the preset advancement cut-off condition; otherwise, the current connecting lane does not satisfy the preset advancement cut-off condition.

[0098] Reversely trace back from the cut-off lane along the road navigation path to update the preceding lane; preset a discrimination condition, and use the preceding lane that meets the preset discrimination condition as the target lane; and use the target lane to form a lane-level navigation path of high-precision map data; specifically:

[0099] S201: Tracing back from the cut-off lane along the road navigation path in reverse order to traverse and obtain the preceding lane of the cut-off lane;

[0100] S202: Preset a judgment condition to determine whether the preceding lane satisfies the preset judgment condition; if so, save the preceding lane as a target lane into the target lane set and the backtracking lane set, and execute step S203; otherwise, execute step S204;

[0101] S203: taking the previous lane in the backtracking lane set as the new cutoff lane, returning to step S201, and updating the previous lane;

[0102] S204: taking the starting lane in the starting lane set as the new ending lane, returning to step S201, and updating the preceding lane;

[0103] S205: Preset a backtracking cutoff condition, and determine whether the current preceding lane meets the preset backtracking cutoff condition; if the current preceding lane meets the backtracking cutoff condition, then end the backtracking and execute step S206; otherwise, continue the backtracking;

[0104] S206: All target lanes stored in the target lane set are combined into a lane-level navigation path of the high-precision map data.

[0105] The target lane set ultimately saves all target lanes, which are lane-level navigation paths matched according to the road navigation paths on the high-precision map data. Figure 4 The figure shows a schematic diagram of matching the target lane. The direction of the road navigation path is from the lower right corner to the upper left corner. The lanes to be connected are selected in sequence along the direction of the arrow. When the lanes are connected to the breakpoint position, i.e. the end lane, the lanes are backtracked. The goal of the backtracking is to exclude the forks that are not on the road navigation path. The data selected in the rectangular box is the lane data that can be excluded during the backtracking, and the target lane is finally obtained.

[0106] The preset judgment condition is: check whether the preceding lane exists in the cache lane set; if so, the preceding lane satisfies the preset judgment condition; otherwise, the preceding lane does not satisfy the preset judgment condition;

[0107] The preset backtracking cutoff condition is: for the current preceding lane, if the preceding lane of the current preceding lane cannot be obtained or the current preceding lane exists in the starting lane set, then the current preceding lane satisfies the preset backtracking cutoff condition; otherwise, the current preceding lane does not satisfy the preset backtracking cutoff condition.

[0108] Performing a check based on the cut-off lane and an end point on the road navigation path, matching the cut-off lane and the end point on the road navigation path;

[0109] If the position of the cut-off lane matches the end position on the road navigation path, the inspection result of the high-precision map data is marked as normal;

[0110] If the position of the cut-off lane does not match the end position on the road navigation path, check whether the navigation point after the cut-off lane exists in the high-precision map data. If so, the road connection relationship in the high-precision map data is interrupted, and an error report is required for manual confirmation.

[0111] Example 3

[0112] This embodiment provides a high-precision map data inspection system based on path matching, which is used to implement the inspection method described in Embodiment 1 or 2. Figure 5 As shown, including:

[0113] A road navigation path generation module, used to obtain road network data of an actual road, and generate a road navigation path based on the road network data of the actual road;

[0114] A high-precision map data acquisition module is used to acquire high-precision map data, map the shape point in the road navigation path onto the high-precision map data as a starting point, and acquire a starting lane within a preset range of the starting point;

[0115] A following lane advancing module is used to advance from the starting lane along the road navigation path in a forward direction, match and update the following lane; preset an advancing cut-off condition, and use the following lane that meets the preset advancing cut-off condition as the cut-off lane;

[0116] A preceding lane backtracking module is used to backtrack from the cut-off lane along the road navigation path in reverse order to update the preceding lane; preset a discrimination condition, and use the preceding lane that meets the preset discrimination condition as the target lane; and use the target lane to form a lane-level navigation path of high-precision map data;

[0117] The high-precision map data verification module is used to perform a verification based on the cut-off lane and the end point on the road navigation path to obtain a verification result of the high-precision map data.

[0118] The same or similar reference numerals correspond to the same or similar components;

[0119] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0120] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for verifying high-precision map data based on path matching, characterized in that: include: Acquiring road network data of an actual road, and generating a road navigation path based on the road network data of the actual road; Acquire high-precision map data, map the shape point in the road navigation path onto the high-precision map data as a starting point, and acquire a starting lane within a preset range of the starting point; Advancing in the forward direction along the road navigation path from the starting lane, matching and updating the subsequent lane; A preset advancing cut-off condition is set, and a connecting lane that meets the preset advancing cut-off condition is used as a cut-off lane; Backtracking from the cut-off lane along the road navigation path in reverse order to update the preceding lane; Presets A judgment condition is used to determine the preceding lane that meets the preset judgment condition as the target lane; Using the target lane to form a lane-level navigation path of high-precision map data; An inspection is performed based on the cut-off lane and the end point on the road navigation path to obtain an inspection result of the high-precision map data.

2. The method for verifying high-precision map data based on path matching according to claim 1 is characterized in that: The specific method of obtaining the road network data of the actual road and generating the road navigation path based on the road network data of the actual road includes: Based on the public map data, the road network data of the actual roads in any area are selected as the road network dataset; Marking road junction points in the road network dataset to form a road junction point dataset; Determine the road grade to be tested, and extract road network data corresponding to the road grade to be tested from the road network data set as the trunk road network data set; In the road junction point data set, the road junction points corresponding to the roads in the trunk road network data set are retained and divided into trunk road internal junction points and trunk road breakpoint junction points to form a trunk road internal junction point set and a trunk road breakpoint junction point set, and the trunk road breakpoint junction points are given direction attributes; The set of breakpoints and junctions of the trunk roads is traversed to generate a road navigation path.

3. The method for verifying high-precision map data based on path matching according to claim 2 is characterized in that: The specific method of assigning the direction attribute of the breakpoint junction point of the trunk road is: The road network data of the actual road is stored according to the actual road driving direction. When the actual road driving direction is toward the main road network dataset, the direction attribute of the main road breakpoint junction is the positive direction; when the actual road driving direction deviates from the main road network dataset, the direction attribute of the main road breakpoint junction is the reverse direction.

4. The method for verifying high-precision map data based on path matching according to claim 2 is characterized in that: The specific method of traversing the set of breakpoints and junctions of the trunk roads to generate a road navigation path is as follows: Selecting a main road breakpoint junction point from the main road breakpoint junction point set, and extending it in the road network data set in the forward direction and the reverse direction of its direction attribute respectively; During the extension process, when encountering the internal junction of the main road, the route is expanded; Calculate the length of the currently expanded route and compare the length of the currently expanded route with the preset route length threshold; if the length of the currently expanded route is greater than the preset route length threshold, end the route expansion and use the currently expanded route as an extended route; otherwise, continue the route expansion; The set of breakpoints and junctions of the trunk roads is traversed, and the above steps are repeated to obtain a plurality of extended routes, and all the extended routes are combined into a road navigation path.

5. The method for verifying high-precision map data based on path matching according to claim 3 or 4, characterized in that: Advancing in the forward direction along the road navigation path from the starting lane, matching and updating the subsequent lane; The specific method of presetting the advancement cut-off condition and using the connecting lane that meets the preset advancement cut-off condition as the cut-off lane is as follows: S101: adding a starting lane within the preset range of the starting point into a starting lane set; S102: setting a first matching condition and a second matching condition, matching the starting lane with the road navigation path, and taking the starting lane that satisfies both the first matching condition and the second matching condition as a candidate lane, and adding it to a candidate lane set; S103: adding the candidate lanes into the cache lane set, advancing from the starting point along the road navigation path in a forward direction, traversing the candidate lanes in the candidate lane set, and obtaining the subsequent lane of each candidate lane; S104: taking each connecting lane as the starting lane for the next round of matching, returning to step S102, updating the connecting lanes until no connecting lanes that meet both the first matching condition and the second matching condition can be obtained; and adding all connecting lanes that meet both the first matching condition and the second matching condition into the connecting lane set; S105: After replacing the candidate lane set with the connecting lane set, the connecting lane set is cleared and the process returns to step S103; S106: Preset the advance cut-off condition, and determine whether the current connecting lane meets the preset advance cut-off condition; If the current connecting lane meets the preset advancing cut-off condition, the advancing is terminated, and the current connecting lane that meets the preset advancing cut-off condition is used as the cut-off lane; Otherwise, keep moving forward.

6. The method for verifying high-precision map data based on path matching according to claim 5 is characterized in that: The first matching condition, the second matching condition and the preset advancement cut-off condition are specifically: The first matching condition is direction matching, and the difference between the lane direction of the starting lane and the road driving direction of the road navigation path is calculated. When the difference is within a preset direction difference threshold range, the corresponding starting lane meets the first matching condition; Otherwise, the corresponding starting lane does not meet the first matching condition; The second matching condition is distance matching, and the Hausdorff distance between the lane direction of the starting lane and the road navigation path is calculated. When the Hausdorff distance is within a preset distance threshold range, the corresponding starting lane meets the second matching condition; Otherwise, the corresponding starting lane does not meet the second matching condition; The preset advancement cut-off condition is: for the current connecting lane, if the connecting lane of the current connecting lane cannot be obtained or the current connecting lane has covered the end point of the road navigation path, then the current connecting lane meets the preset advancement cut-off condition; Otherwise, the current connecting lane does not meet the preset advancement cut-off condition.

7. The method for verifying high-precision map data based on path matching according to claim 6 is characterized in that: Backtracking from the cut-off lane along the road navigation path in reverse order to update the preceding lane; Preset a judgment condition, and take the preceding lane that meets the preset judgment condition as the target lane; The specific method of using the target lane to form a lane-level navigation path of high-precision map data is as follows: S201: Tracing back from the cut-off lane along the road navigation path in reverse order to traverse and obtain the preceding lane of the cut-off lane; S202: Preset a judgment condition, and determine whether the preceding lane satisfies the preset judgment condition; If satisfied, the preceding lane is saved as a target lane in the target lane set and the backtracking lane set, and step S203 is executed; Otherwise, execute step S204; S203: taking the previous lane in the backtracking lane set as the new cutoff lane, returning to step S201, and updating the previous lane; S204: taking the starting lane in the starting lane set as the new ending lane, returning to step S201, and updating the preceding lane; S205: Preset a backtracking cutoff condition, and determine whether the current preceding lane meets the preset backtracking cutoff condition; if the current preceding lane meets the backtracking cutoff condition, then end the backtracking and execute step S206; otherwise, continue the backtracking; S206: All target lanes stored in the target lane set are combined into a lane-level navigation path of the high-precision map data.

8. The method for verifying high-precision map data based on path matching according to claim 7 is characterized in that: The preset discrimination condition and the preset backtracking cut-off condition are specifically: The preset judgment condition is: checking whether the preceding lane exists in the cache lane set; If so, the preceding lane satisfies the preset judgment condition; Otherwise, the preceding lane does not satisfy the preset judgment condition; The preset backtracking cutoff condition is: for the current preceding lane, if the preceding lane of the current preceding lane cannot be obtained or the current preceding lane exists in the starting lane set, then the current preceding lane satisfies the preset backtracking cutoff condition; otherwise, the current preceding lane does not satisfy the preset backtracking cutoff condition.

9. The method for verifying high-precision map data based on path matching according to claim 8, characterized in that: The specific method for obtaining the inspection result of the high-precision map data based on the inspection of the cut-off lane and the end point on the road navigation path is: Matching the cut-off lane with an end point on the road navigation path; If the position of the cut-off lane matches the end position on the road navigation path, the inspection result of the high-precision map data is marked as normal; If the position of the cut-off lane does not match the end position on the road navigation path, the inspection result of the high-precision map data is marked as an error.

10. A high-precision map data inspection system based on path matching, used to implement the inspection method according to any one of claims 1 to 9, characterized in that: include: A road navigation path generation module, used to obtain road network data of an actual road, and generate a road navigation path based on the road network data of the actual road; A high-precision map data acquisition module is used to acquire high-precision map data, map the shape point in the road navigation path onto the high-precision map data as a starting point, and acquire a starting lane within a preset range of the starting point; A subsequent lane advancing module, used to advance from the starting lane along the road navigation path in a forward direction, and match and update the subsequent lane; A preset advancing cut-off condition is set, and a connecting lane that meets the preset advancing cut-off condition is used as a cut-off lane; A previous lane backtracking module, used to backtrack from the cut-off lane along the road navigation path in reverse order to update the previous lane; Presets A judgment condition is used to determine the preceding lane that meets the preset judgment condition as the target lane; Using the target lane to form a lane-level navigation path of high-precision map data; The high-precision map data verification module is used to perform a verification based on the cut-off lane and the end point on the road navigation path to obtain a verification result of the high-precision map data.

Images