Lane-level map evaluation method and device and vehicle

Abstract

The invention provides a lane-level map evaluation method and device and a vehicle, and is applied to an automatic driving vehicle, and the evaluation method comprises the steps: obtaining a to-be-evaluated lane-level map; under the condition that the vehicle is manually driven, real driving data of the vehicle on the target road section are obtained, and the real driving data comprise driving parameters and a real driving path; lane data of a lane-level map on the target road section is obtained, and the lane data comprises map elements and a simulated driving path, generated by the lane-level map, of the vehicle on the target road section under the condition that the vehicle is in automatic driving; evaluating map elements based on the driving parameters, and evaluating a simulated driving path based on the real driving path; based on a result of the evaluation, it is determined whether the lane data satisfies a usage criterion for automatic driving of the vehicle. By acquiring the real driving data, the lane data of the lane-level map can be accurately and quickly evaluated, and the time cost and the labor cost in the lane-level map evaluation process are saved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a lane-level map evaluation method, device, and vehicle. Background Art

[0002] Domestic autonomous driving assistance technology is booming. Among them, high-precision lane-level map data is one of the important sensors to ensure the safety and stability of autonomous driving assistance functions in autonomous driving.

[0003] At present, lane-level maps collected and produced based on mass-produced autonomous driving vehicles are mostly non-professionally collected, so some lane scenes are missing in the stored roads. For such incomplete lane-level maps, whether they can be used for autonomous driving is currently lacking in the industry.

[0004] Application Contents

[0005] In view of this, the present application aims to propose a lane-level map evaluation method, device and vehicle to evaluate the lane-level map collected and produced based on mass-produced autonomous driving vehicles.

[0006] To achieve the above purpose, the technical solution of this application is implemented as follows:

[0007] A lane-level map evaluation method is applied to an autonomous driving vehicle, the evaluation method comprising:

[0008] Obtain the lane-level map to be evaluated;

[0009] When the vehicle is under manual driving, obtaining real driving data of the vehicle on a target road section, wherein the real driving data includes driving parameters and a real driving path;

[0010] Acquire lane data of the lane-level map on the target road section, the lane data including map elements and a simulated driving path of the vehicle on the target road section generated by the lane-level map when the vehicle is in the automatic driving state;

[0011] evaluating the map elements based on the driving parameters, and evaluating the simulated driving path based on the actual driving path;

[0012] Based on the evaluation result, it is determined whether the lane data meets the usage criteria for the vehicle's autonomous driving.

[0013] Further, the evaluating the map element based on the driving parameter includes:

[0014] Acquire the environmental information of the vehicle in real time, the environmental information at least including road signs, road markings, traffic lights, and pedestrians and other vehicles around the vehicle;

[0015] The map element is evaluated based on the driving parameter and the environmental information to determine whether the map element actually exists on the target road segment.

[0016] Furthermore, the map elements include at least a longitudinal deceleration element and a lane linear element, and the step of evaluating the map elements based on the driving parameters and the environmental information to determine whether the map elements actually exist on the target road section includes:

[0017] determining a type of the driving parameter;

[0018] In the case where the type is a longitudinal deceleration type, determining whether the longitudinal deceleration element actually exists on the target road section based on the environmental information;

[0019] In the case where the type is a lateral lane change type, determining whether the lane linear element actually exists on the target road section based on the environmental information;

[0020] The longitudinal deceleration element corresponds to information in the environmental information indicating vehicle deceleration; and the lane linear element corresponds to information in the environmental information indicating whether the vehicle changes lanes.

[0021] Further, the evaluating the simulated driving path based on the actual driving path includes:

[0022] Dividing the real driving path into a plurality of real sub-paths, and dividing the simulated driving path into a plurality of simulated sub-paths, wherein the real sub-paths correspond to the simulated sub-paths in a one-to-one manner;

[0023] Determine the lanes where each segment of the real sub-path and each segment of the simulated sub-path are located respectively;

[0024] The real lane where each real sub-path is located is compared with the simulated lane where its corresponding simulated sub-path is located to determine whether the real lane is consistent with the simulated lane.

[0025] Furthermore, the dividing the real driving path into a plurality of real sub-paths, and the dividing the simulated driving path into a plurality of simulated sub-paths, comprises:

[0026] Taking a real intersection in the real driving path as a first node, and based on the first node, determining a path between two adjacent first nodes as the real sub-path;

[0027] And taking a simulated intersection in the simulated driving path as a second node, and based on the second node, determining a path between two adjacent second nodes as the simulated sub-path.

[0028] Furthermore, the target road section is the road section that the vehicle passes through from starting to shutting down, and the step of obtaining the simulated driving path includes:

[0029] Determining the position of the vehicle when it is started as the starting point of the target section, and determining the position of the vehicle when it is turned off as the end point of the target section;

[0030] Based on the starting point and the end point, the simulated driving path is generated on the lane-level map.

[0031] Furthermore, including a plurality of the target road sections, the evaluation method further includes:

[0032] Obtaining evaluation results corresponding to lane data of the lane-level map on a plurality of target road sections respectively;

[0033] In response to the target road section where the vehicle is currently located, the automatic driving function of the vehicle is controlled to be turned on or off based on the evaluation result of the lane data corresponding to the current target road section.

[0034] Furthermore, the evaluation method further includes:

[0035] Real-time detection of whether the autonomous driving perception system and autonomous driving positioning system of the vehicle are in working state;

[0036] When it is determined that the autonomous driving perception system and the autonomous driving positioning system are both in the working state, the lane level map is evaluated.

[0037] Compared with the prior art, the installation structure described in this application has the following advantages:

[0038] The present application provides a lane-level map evaluation method, which is applied to an autonomous driving vehicle, and the evaluation method includes: obtaining a lane-level map to be evaluated; when the vehicle is in manual driving, obtaining real driving data of the vehicle on a target road section, the real driving data including driving parameters and a real driving path; obtaining lane data of the lane-level map on the target road section, the lane data including map elements and a simulated driving path of the vehicle on the target road section generated by the lane-level map when the vehicle is in the autonomous driving; evaluating the map elements based on the driving parameters, and evaluating the simulated driving path based on the real driving path; and determining whether the lane data meets the usage standards of the autonomous driving of the vehicle based on the evaluation result.

[0039] Therefore, the present application obtains the driving parameters and the actual driving path of the vehicle on the target road section when the vehicle is manually driven, and then obtains the lane data of the lane-level map on the target road section, so as to evaluate the map elements in the lane data according to the driving parameters and determine whether the map elements actually exist on the target road section. At the same time, the simulated driving path in the lane data is evaluated according to the actual driving path, so as to determine whether the lanes in the simulated driving path are consistent with the lanes in the actual driving path, and then the lane-level map with incomplete lane scenes (map elements and lanes), that is, the lane-level map produced based on the collection of mass-produced autonomous driving vehicles, can be evaluated. In addition, the evaluation method provided by the present application does not require manual quality inspection, which greatly saves the time and labor costs in the lane-level map evaluation process.

[0040] Another object of the present application is to propose a lane-level map evaluation device to evaluate lane-level maps collected and produced based on mass-produced autonomous driving vehicles.

[0041] To achieve the above purpose, the technical solution of this application is implemented as follows:

[0042] A lane-level map evaluation device, applied to an autonomous driving vehicle, comprising:

[0043] A first acquisition unit, used to acquire a lane-level map to be evaluated;

[0044] A second acquisition unit is used to acquire real driving data of the vehicle on the target road section when the vehicle is under manual driving, wherein the real driving data includes driving parameters and a real driving path;

[0045] A third acquisition unit is used to acquire lane data of the lane-level map on the target road section, wherein the lane data includes map elements and a simulated driving path of the vehicle on the target road section generated by the lane-level map when the vehicle is in the automatic driving state;

[0046] an evaluation unit, configured to evaluate the map element based on the driving parameter, and to evaluate the simulated driving path based on the real driving path;

[0047] A determination unit is used to determine whether the lane data meets the usage criteria of the vehicle's automatic driving based on the evaluation result.

[0048] The advantages of the evaluation device and the above-mentioned evaluation method over the prior art are the same, and will not be elaborated here.

[0049] Another object of the present application is to propose a vehicle for evaluating lane-level maps collected and produced based on autonomous driving mass-produced vehicles.

[0050] To achieve the above purpose, the technical solution of this application is implemented as follows:

[0051] A vehicle comprises an evaluation module, wherein the evaluation module is used to implement the above-mentioned lane-level map evaluation method.

[0052] The advantages of the vehicle and the above-mentioned evaluation method over the prior art are the same and will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0054] Figure 1 A flowchart of the steps of a lane-level map evaluation method described in an embodiment of the present application;

[0055] Figure 2 A flowchart of a method for evaluating map elements based on driving parameters according to an embodiment of the present application;

[0056] Figure 3 A flowchart of the steps of a method for determining whether a map element actually exists on a target road section according to an embodiment of the present application;

[0057] Figure 4 A flowchart of a method for evaluating a simulated driving path based on a real driving path according to an embodiment of the present application;

[0058] Figure 5 A schematic diagram of a real driving path R1 according to an embodiment of the present application;

[0059] Figure 6 A schematic diagram of a simulated driving path R2 according to an embodiment of the present application;

[0060] Figure 7 A schematic diagram of a divided real driving path R1 according to an embodiment of the present application;

[0061] Figure 8 A schematic diagram of a divided simulated driving path R2 according to an embodiment of the present application;

[0062] Fig. 9 A schematic diagram showing a comparison between a simulated driving path and a real driving path according to an embodiment of the present application;

[0063] Fig.10A schematic diagram of a module of a lane-level map evaluation device described in an embodiment of the present application.

[0064] Reference numerals: 1. evaluation device; 101. first acquisition unit; 102. second acquisition unit; 103. third acquisition unit; 104. evaluation unit; 105. determination unit. DETAILED DESCRIPTION

[0065] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments in this application may be combined with each other.

[0066] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0067] Domestic autonomous driving assistance technology is booming. Among them, high-precision lane-level map data is one of the important sensors to ensure the safety and stability of autonomous driving assistance functions in autonomous driving. The quality of lane-level map data plays an important role in whether autonomous driving vehicles can enable autonomous driving functions. At present, traditional map vendors use program logic problem checks and a large amount of manual quality inspection to ensure map quality based on high-precision maps collected and produced by professional collection vehicles. The quality inspection process requires a lot of manual quality inspection costs, and the quality inspection cycle is long, which loses the freshness of the map. How to highly automate the evaluation of map data quality to see if it can be used for autonomous driving is a current problem in the industry.

[0068] As for lane-level maps collected and produced by mass-produced autonomous driving vehicles, since most of these maps are not collected professionally, some lane scenes are missing from the stored roads. There is currently a lack of relevant evaluation methods in the industry as to whether such incomplete lane-level maps can be used for autonomous driving.

[0069] In view of this, the present application embodiment proposes a lane-level map evaluation method, which is applied to autonomous driving vehicles to quickly and accurately evaluate lane-level maps collected and produced by autonomous driving mass-produced vehicles, thereby saving time and labor costs in the lane-level map evaluation process. Figure 1 , Figure 1 A flowchart of the steps of a lane level map evaluation method according to an embodiment of the present application is shown. Figure 1 As shown, the evaluation method includes:

[0070] Step S1, obtaining a lane-level map to be evaluated.

[0071] In the embodiment of the present application, the lane-level map to be evaluated is a lane-level map collected and produced based on mass-produced autonomous driving vehicles (the lane-level map has undergone basic program automation logic checks and has no logic problems). Because it is not collected professionally, the stored road does contain some lane scenes. Among them, the lane scene mainly includes map elements and lanes.

[0072] In specific implementation, a lane-level map evaluation module can be deployed on the vehicle side, and the evaluation module obtains the lane-level map to be evaluated and performs the evaluation. Specifically, the map middleware obtains the lane-level map collected and produced based on the mass-produced autonomous driving vehicle through the map SDK, that is, the lane-level map to be evaluated, and sends the lane-level map to the evaluation module, thereby obtaining the lane-level map to be evaluated.

[0073] Step S2, when the vehicle is under manual driving, obtaining the real driving data of the vehicle on the target road section, wherein the real driving data includes driving parameters and a real driving path.

[0074] After the evaluation module obtains the lane-level map to be evaluated, it will obtain the real driving data of the vehicle on the target road section when the vehicle is under manual driving. The target road section is the road section that the vehicle passes from starting to shutting down when the vehicle is under manual driving.

[0075] The real driving data includes driving parameters and real driving paths. The driving parameters represent the actions taken by the vehicle under manual driving on the target road section, which can be deceleration, lane change, left turn, etc. The real driving path is the driving path of the vehicle on the target road section under manual driving. The real driving path can be obtained through the automatic driving positioning system in the vehicle.

[0076] Step S3, obtaining lane data of the lane-level map on the target road section, wherein the lane data includes map elements and a simulated driving path of the vehicle on the target road section generated by the lane-level map when the vehicle is in the automatic driving state.

[0077] After obtaining the real driving data of the vehicle on the target road section, the evaluation module also obtains the lane data of the lane-level map on the target road section. The lane data includes map elements and simulated driving paths.

[0078] Among them, the map elements include at least longitudinal deceleration elements and lane linear elements. The longitudinal deceleration elements include stop lines, zebra crossings, etc. The lane linear elements include the line shape of the lane edge, such as dotted lines, solid lines, double dotted lines and double solid lines.

[0079] Step S4, evaluating the map elements based on the driving parameters, and evaluating the simulated driving path based on the actual driving path.

[0080] When the evaluation module obtains the real driving data of the vehicle on the target road section and the lane data of the lane-level map on the target road section, it will evaluate the acquired lane data. Specifically, the map elements are evaluated based on the driving parameters, and the simulated driving path is evaluated based on the real driving path.

[0081] In specific implementation, evaluating the simulated driving path is actually evaluating the lane where the simulated driving path is located. Therefore, the embodiment of the present application can evaluate the lane scene in the lane-level map, that is, the map elements and lanes.

[0082] Step S5, based on the evaluation result, determining whether the lane data meets the usage criteria for the vehicle's autonomous driving.

[0083] When the evaluation module completes the evaluation of the map elements in the lane data and the simulated driving path, it will output the corresponding evaluation results. Based on the output evaluation results, it can be determined whether the lane data meets the vehicle's automatic driving usage standards. If it does, the automatic driving function is turned on when the vehicle is driving on the target road section; if it does not, the automatic driving function is turned off when the vehicle is driving on the target road section.

[0084] Therefore, in the embodiment of the present application, the driving parameters and the actual driving path of the vehicle on the target road section are obtained by the vehicle being manually driven, and then the lane data of the lane-level map on the target road section are obtained, so as to evaluate the map elements in the lane data according to the driving parameters and determine whether the map elements actually exist on the target road section. At the same time, the simulated driving path in the lane data is evaluated according to the actual driving path, so as to determine whether the lanes in the simulated driving path are consistent with the lanes in the actual driving path, and then the lane-level map with incomplete lane scenes (map elements and lanes), that is, the lane-level map produced based on the collection of mass-produced autonomous driving vehicles, can be evaluated. In addition, the evaluation method provided by the present application does not require manual quality inspection, which greatly saves the time and labor costs in the lane-level map evaluation process.

[0085] In an optional embodiment, the specific steps of evaluating map elements based on driving parameters can refer to Figure 2 , Figure 2 A flowchart of a method for evaluating map elements based on driving parameters according to an embodiment of the present application is shown. Figure 2 As shown, the steps are as follows:

[0086] Step S411, acquiring in real time the environmental information of the vehicle, wherein the environmental information at least includes road signs, road markings, traffic lights, and pedestrians and other vehicles around the vehicle.

[0087] In the implementation of this application, when evaluating map elements, the evaluation module also needs to obtain the environmental information of the vehicle, specifically, the map element evaluation submodule in the evaluation module obtains it. In the specific implementation, the environmental information can be obtained in real time through the vehicle's autonomous driving perception system, and after the acquisition is completed, the autonomous driving perception system sends it to the map element evaluation submodule.

[0088] The environmental information includes at least road signs, road markings, traffic lights, and pedestrians and other vehicles around the vehicle.

[0089] Step S412: evaluating the map element based on the driving parameter and the environmental information to determine whether the map element actually exists on the target road section.

[0090] After obtaining the vehicle's environmental information, the map element evaluation submodule evaluates the map elements based on the driving parameters and environmental information to determine whether the map elements actually exist on the target road section. Specifically, it determines whether the stop line, zebra crossing, dotted line, solid line, double dotted line or double solid line actually exist on the target road section.

[0091] The specific steps of evaluating the map element based on the driving parameters and the environmental information and determining whether the map element actually exists on the target road section can be referred to in Figure 3 , Figure 3 A flowchart of the method steps for determining whether a map element actually exists on a target road segment according to an embodiment of the present application is shown. Figure 3 As shown, the steps are as follows:

[0092] Step S4121, determining the type of the driving parameter.

[0093] In an embodiment of the present application, the types of driving parameters include at least a longitudinal deceleration type and a lateral lane change type, wherein the longitudinal deceleration type characterizes the type of action made by the vehicle on the target section under manual driving, specifically, the type of longitudinal deceleration of the vehicle; the lateral lane change type characterizes the type of action made by the vehicle on the target section under manual driving, specifically, the type of lateral lane change of the vehicle.

[0094] Step S4122: When the type is a longitudinal deceleration type, based on the environmental information, determine whether the longitudinal deceleration element actually exists on the target road section.

[0095] In the embodiment of the present application, the longitudinal deceleration element corresponds to information in the environmental information indicating vehicle deceleration, specifically, a stop line and a zebra crossing, etc.

[0096] When the map element evaluation submodule determines that the type of driving parameter is a longitudinal deceleration type, that is, when the vehicle is decelerating longitudinally, it will determine whether the longitudinal deceleration element actually exists on the target road section based on the environmental information. For example, the map element evaluation submodule excludes the possibility that the vehicle is longitudinally decelerating due to the deceleration of the vehicle in front based on the environmental information sent by the autonomous driving perception system, and determines that a stop line actually exists on the target road section. Next, the map element evaluation submodule determines whether the longitudinal deceleration element is a stop line. If it is a stop line, it is determined that the longitudinal deceleration element actually exists on the target road section, that is, the longitudinal deceleration element of the lane-level map on the target road section is correctly made; if it is not a stop line, it is determined that the longitudinal deceleration element does not exist on the target road section, that is, the longitudinal deceleration element of the lane-level map on the target road section is made incorrectly.

[0097] Step S4123, when the lane change type is a lateral lane change type, based on the environmental information, determining whether the lane linear element actually exists on the target road section.

[0098] In an embodiment of the present application, the lane linear element corresponds to information in the environmental information indicating whether the vehicle changes lanes, specifically, a dotted line, a solid line, a double dotted line, a double solid line, etc.

[0099] When the map element evaluation submodule determines that the type of driving parameters is a lateral lane change type, that is, when the vehicle changes lanes laterally, it will determine whether the lane linear elements actually exist on the target road section based on environmental information. Specifically, it will determine whether the dotted line, solid line, double dotted line and double solid line actually exist on the target road section.

[0100] Therefore, the embodiment of the present application can determine whether the map elements (longitudinal deceleration elements and lane linear elements) of the lane-level map on the target road section are correct through the map element evaluation submodule. If correct, it indicates that the lane-level map has a high confidence level.

[0101] In an optional implementation, the specific steps of evaluating the simulated driving path based on the actual driving path can refer to Figure 4 , Figure 4 A flowchart of a method for evaluating a simulated driving path based on a real driving path according to an embodiment of the present application is shown. Figure 4 As shown, the steps are as follows:

[0102] Step S421 , dividing the real driving path into a plurality of real sub-paths, and dividing the simulated driving path into a plurality of simulated sub-paths, wherein the real sub-paths correspond to the simulated sub-paths one by one.

[0103] When the evaluation module obtains the real driving path and the simulated driving path, the path evaluation submodule in the evaluation module obtains the real driving path and the simulated driving path, and the path evaluation submodule divides the real driving path and the simulated driving path respectively. Specifically, the real driving path is divided into multiple real sub-paths according to the preset division standard, and the simulated driving path is divided into multiple simulated sub-paths. The preset division standard is as follows:

[0104] Taking a real intersection in the real driving path as a first node, and based on the first node, determining a path between two adjacent first nodes as a real sub-path;

[0105] And taking the simulated intersection in the simulated driving path as the second node, and based on the second node, determining the path between two adjacent second nodes as the simulated sub-path.

[0106] Therefore, the embodiment of the present application divides the real driving path into multiple real sub-paths and divides the simulated driving path into multiple simulated sub-paths in a manner from intersection to intersection.

[0107] Step S422, respectively determine the lanes where each segment of the real sub-path and each segment of the simulated sub-path are located.

[0108] The path evaluation submodule divides the real driving path and the simulated driving path into multiple real sub-paths and multiple simulated sub-paths, and then determines the lane where each real sub-path and each simulated sub-path are located.

[0109] Step S423, comparing the real lane where each real sub-path is located with the simulated lane where its corresponding simulated sub-path is located to determine whether the real lane is consistent with the simulated lane.

[0110] After the path evaluation submodule determines the lanes where each real subpath and each simulated subpath are located, it compares the real lane where each real subpath is located with the simulated lane where its corresponding simulated subpath is located, so as to determine whether the real lane is consistent with the simulated lane. If the real lane is consistent with the simulated lane, it indicates that the simulated subpath corresponding to the simulated lane can be used for the vehicle's automatic driving, that is, when the vehicle is on the road section corresponding to the simulated subpath, the vehicle's automatic driving function can be turned on.

[0111] In the specific implementation, there are multiple lanes in the target section, so there are multiple solutions for the simulated driving path generated by the lane-level map, each of which corresponds to a different lane. Correspondingly, there are also multiple solutions in the simulated sub-path obtained by division, and each solution corresponds to a different lane. For example, the target section is a straight section without an intersection, and the straight section consists of a first straight lane, a second straight lane, and a third straight lane. Then the solutions for the simulated driving path generated by the lane-level map include at most 3, which are a solution including the first straight lane, a solution including the second straight lane, and a solution including the third straight lane. Correspondingly, the solutions for the simulated sub-path obtained by division include at most 3, which are a solution including the first straight lane, a solution including the second straight lane, and a solution including the third straight lane. If the real lane where the real sub-path is located is the first straight lane, and the solution provided by the simulated sub-path includes the solution of the first straight lane, then it is considered that the real lane where the real sub-path is located is consistent with the simulated lane where the simulated sub-path is located. If the real lane where the real subpath is located is the first through lane, and the solutions provided by the simulated subpath only include solutions for the third through lane, then it is considered that the real lane where the real subpath is located is inconsistent with the simulated lane where its simulated subpath is located.

[0112] In an optional implementation, the target road section is a road section that the vehicle passes through from startup to shutdown, and the step of obtaining the simulated driving path includes:

[0113] Determining the position of the vehicle when it is started as the starting point of the target section, and determining the position of the vehicle when it is turned off as the end point of the target section;

[0114] Based on the starting point and the end point, the simulated driving path is generated on the lane-level map.

[0115] When the vehicle is under manual driving, the vehicle's automatic driving positioning system is turned on. The automatic driving positioning system records the position of the vehicle when it is started and the position of the vehicle when it is shut down, and sends these two positions to the lane-level map. The lane-level map determines the position of the vehicle when it is started as the starting point of the target section, and the position of the vehicle when it is shut down as the end point of the target section, and then automatically generates a simulated driving path based on the starting point and the end point.

[0116] In the following, in conjunction with a specific embodiment, an exemplary description is given of how the path evaluation submodule evaluates the simulated driving path:

[0117] The vehicle travels along a target section under manual driving, where the starting point of the target section is starting point A and the end point is end point B. After the vehicle has traveled along the target section, that is, after the engine is turned off, the path evaluation submodule evaluates the simulated driving path.

[0118] First, the path evaluation submodule obtains the actual driving path R1 of the vehicle on the target road section through the automatic driving positioning system, that is, the driving path of the vehicle from starting point A to starting point B. Figure 5 , Figure 5 A schematic diagram of a real driving path R1 described in an embodiment of the present application is shown. Next, the path evaluation submodule obtains a simulated driving path R2 automatically generated by the lane-level map. The simulated driving path is specifically referred to as Figure 6 , Figure 6 A schematic diagram of a simulated driving path R2 according to an embodiment of the present application is shown. Next, according to a preset division method, the real driving path R1 is divided into multiple real sub-paths, and the simulated driving path R2 is divided into multiple simulated sub-paths.

[0119] The actual driving path R1 after division refers to Figure 7 , Figure 7 A schematic diagram of a divided real driving path R1 according to an embodiment of the present application is shown, as shown in FIG. Figure 7 As shown, there are two intersections in the real driving path R1. These two intersections are taken as first nodes. Based on the first nodes, the paths between two adjacent first nodes are determined as real sub-paths L1, L2, L3, L4 and L5. Figure 7 As shown, the actual lane where each actual sub-path is located can be directly determined.

[0120] The simulated driving path R2 after division refers to Figure 8 , Figure 8 A schematic diagram of a divided simulated driving path R2 according to an embodiment of the present application is shown, as shown in FIG. Figure 8 As shown, there are two intersections in the simulated driving path R2, and these two intersections are taken as second nodes. Based on the second nodes, the paths between two adjacent second nodes are determined as simulated sub-paths L11, L22, L33, L44 and L55. Figure 8 As shown, the simulated lane where each simulated sub-path is located can be directly determined. For example, the simulated sub-path L11 includes three lanes.

[0121] Next, refer to Fig. 9 , Fig. 9 A schematic diagram showing a comparison between a simulated driving path and a real driving path according to an embodiment of the present application is shown. Fig. 9As shown, the simulated driving path is compared with the real driving path. Specifically, the real lane where each real sub-path is located is compared with the simulated lane where its corresponding simulated sub-path is located. Thus, it can be determined that the real lane where the real sub-path L1 is located is consistent with the simulated lane where the simulated sub-path L11 is located, the real lane where the real sub-path L2 is located is inconsistent with the simulated lane where the simulated sub-path L22 is located, the real lane where the real sub-path L3 is located is consistent with the simulated lane where the simulated sub-path L33 is located, the real lane where the real sub-path L4 is located is inconsistent with the simulated lane where the simulated sub-path L44 is located, and the real lane where the real sub-path L5 is located is consistent with the simulated lane where the simulated sub-path L55 is located.

[0122] When the map element evaluation submodule and the path evaluation submodule complete their evaluations, the corresponding evaluation results will be output and sent to the lane-level map storage. In this way, when the vehicle is configured with the lane-level map and is on the target road section, the vehicle's autonomous driving function can be turned on or off based on the stored evaluation results.

[0123] The evaluation results are stored in the evaluation result storage medium of the lane-level map, and the evaluation result storage medium includes a map element evaluation result storage medium and a simulated driving path evaluation result storage medium. The map element evaluation result storage medium can be represented by the following Table 1:

[0124] Table 1. Map element evaluation results

[0125]

[0126] The simulated driving path evaluation result storage medium may be represented by the following simulated driving path evaluation result table 2:

[0127] Table 2. Simulated driving path evaluation results

[0128]

[0129] As shown in Table 1 and Table 2, for the map elements whose Check_result value is 1 in Table 1, their confidence is set to high; for the map elements whose Check_result value is not 1 in Table 1, their confidence is set to low. For the simulation sub-paths whose Check_result value is 1 in Table 2, it is confirmed that the simulation sub-path can be used for autonomous driving, that is, when the vehicle is on the road section corresponding to the simulation sub-path, the autonomous driving function will be turned on; for the simulation sub-paths whose Check_result value is not 1 in Table 2, the autonomous driving is downgraded, that is, when the vehicle is on the road section corresponding to the simulation sub-path, the autonomous driving function will be turned off, thereby improving the safety and reliability of autonomous driving.

[0130] In an optional implementation, including a plurality of the target road sections, the evaluation method further includes:

[0131] Obtaining evaluation results corresponding to lane data of the lane-level map on a plurality of target road sections respectively;

[0132] In response to the target road section where the vehicle is currently located, the automatic driving function of the vehicle is controlled to be turned on or off based on the evaluation result of the lane data corresponding to the current target road section.

[0133] In an embodiment of the present application, evaluation results corresponding to lane data of the lane-level map on multiple target sections have been stored in the lane-level map. When the vehicle is in any target section, the evaluation result corresponding to the lane data of the lane-level map on the target section can be directly obtained. According to the evaluation result, the automatic driving function of the vehicle on the target section is controlled to be turned on or off.

[0134] In an optional implementation, the evaluation method further includes:

[0135] Real-time detection of whether the autonomous driving perception system and autonomous driving positioning system of the vehicle are in working state;

[0136] When it is determined that the autonomous driving perception system and the autonomous driving positioning system are both in the working state, the lane level map is evaluated.

[0137] In an embodiment of the present application, when the map element evaluation submodule evaluates map elements, the autonomous driving perception system is required to obtain the environmental information of the vehicle, and when the path evaluation submodule evaluates the simulated driving path, the autonomous driving positioning system is required to obtain the actual driving path and the starting point and end point of the target section. Therefore, before evaluating the lane-level map, it is necessary to ensure that the autonomous driving perception system and the autonomous driving positioning system are both in working condition.

[0138] Based on the same application concept, the present application embodiment also proposes a lane level map evaluation device, referring to Fig.10 , Fig.10 A schematic diagram of a module of a lane-level map evaluation device according to an embodiment of the present application is shown. Fig.10 As shown, the evaluation device 1 comprises:

[0139] The first acquisition unit 101 is used to acquire a lane-level map to be evaluated.

[0140] In a specific implementation, the map middleware obtains the lane-level map collected and produced based on the mass-produced autonomous driving vehicle through the map SDK, that is, the lane-level map to be evaluated, and sends the lane-level map to the first acquisition unit 101 .

[0141] The second acquisition unit 102 is used to acquire the real driving data of the vehicle on the target road section when the vehicle is under manual driving, and the real driving data includes driving parameters and a real driving path.

[0142] After the first acquisition unit 101 acquires the lane-level map to be evaluated, the second acquisition unit 102 acquires the real driving data of the vehicle on the target road section when the vehicle is manually driven. The target road section is the road section that the vehicle passes through from start to shutdown when the vehicle is manually driven.

[0143] The real driving data includes driving parameters and real driving paths. The driving parameters represent the actions taken by the vehicle under manual driving on the target road section, which can be deceleration, lane change, left turn, etc. The real driving path is the driving path of the vehicle on the target road section under manual driving. The real driving path can be obtained through the automatic driving positioning system in the vehicle.

[0144] The third acquisition unit 103 is used to acquire lane data of the lane-level map on the target road section, wherein the lane data includes map elements and a simulated driving path of the vehicle on the target road section generated by the lane-level map when the vehicle is in the automatic driving state.

[0145] After the second acquisition unit 102 acquires the real driving data of the vehicle on the target road section, the third acquisition unit 103 acquires the lane data of the lane-level map on the target road section, wherein the lane data includes map elements and simulated driving paths.

[0146] Among them, the map elements include at least longitudinal deceleration elements and lane linear elements. The longitudinal deceleration elements include stop lines, zebra crossings, etc. The lane linear elements include the line shape of the lane edge, such as dotted lines, solid lines, double dotted lines and double solid lines.

[0147] The evaluation unit 104 is configured to evaluate the map element based on the driving parameter, and to evaluate the simulated driving path based on the actual driving path.

[0148] After the second acquisition unit 102 acquires the real driving data of the vehicle on the target road section, and the third acquisition unit 103 acquires the lane data of the lane-level map on the target road section, the evaluation unit 104 evaluates the acquired lane data. Specifically, the map elements are evaluated based on the driving parameters, and the simulated driving path is evaluated based on the real driving path.

[0149] The determination unit 105 is used to determine whether the lane data meets the usage standard of the vehicle automatic driving based on the evaluation result.

[0150] After the evaluation unit 104 completes the evaluation of the map elements in the lane data and the simulated driving path, the corresponding evaluation result will be output. Based on the output evaluation result, the determination unit 105 can determine whether the lane data meets the use standard of the vehicle automatic driving. If it does, the automatic driving function is turned on when the vehicle is driving on the target road section; if it does not, the automatic driving function is turned off when the vehicle is driving on the target road section.

[0151] Based on the same application concept, an embodiment of the present application further proposes a vehicle, which includes an evaluation module, and the evaluation module is used to implement the above-mentioned lane-level map evaluation method.

[0152] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0153] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0154] It should also be noted that, in this article, the orientation or position relationship indicated by the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations, nor can they be understood as indicating or implying relative importance. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. Without more restrictions, an element defined by the phrase "comprising a ..." does not exclude the existence of other identical elements in the process, method, article or terminal device comprising the element.

[0155] The technical solution provided by the present application is described in detail above. The principle and implementation method of the present application are described in this article using specific examples. The description of the above embodiments is only used to help understand the present application, and the content of this specification should not be understood as limiting the present application. At the same time, for those of ordinary skill in the art, according to the present application, there will be different forms of changes in the specific implementation methods and application scopes. It is not necessary and impossible to list all the implementation methods here, and the obvious changes or modifications derived therefrom are still within the scope of protection of the present application.

Claims

1. A lane-level map evaluation method, characterized in that: Applied to autonomous driving vehicles, the evaluation method includes: Obtain the lane-level map to be evaluated; When the vehicle is under manual driving, obtaining real driving data of the vehicle on a target road section, wherein the real driving data includes driving parameters and a real driving path; Acquire lane data of the lane-level map on the target road section, the lane data including map elements and a simulated driving path of the vehicle on the target road section generated by the lane-level map when the vehicle is in the automatic driving state; evaluating the map elements based on the driving parameters, and evaluating the simulated driving path based on the actual driving path; Based on the evaluation result, it is determined whether the lane data meets the usage criteria for the vehicle's autonomous driving.

2. The lane level map evaluation method according to claim 1, characterized in that: The evaluating the map element based on the driving parameter comprises: Acquire the environmental information of the vehicle in real time, the environmental information at least including road signs, road markings, traffic lights, and pedestrians and other vehicles around the vehicle; The map element is evaluated based on the driving parameter and the environmental information to determine whether the map element actually exists on the target road segment.

3. The lane level map evaluation method according to claim 2, characterized in that: The map elements at least include a longitudinal deceleration element and a lane linear element, and the step of evaluating the map elements based on the driving parameters and the environmental information to determine whether the map elements actually exist on the target road section includes: determining a type of the driving parameter; In the case where the type is a longitudinal deceleration type, determining whether the longitudinal deceleration element actually exists on the target road section based on the environmental information; In the case where the type is a lateral lane change type, determining whether the lane linear element actually exists on the target road section based on the environmental information; The longitudinal deceleration element corresponds to information in the environmental information indicating vehicle deceleration; and the lane linear element corresponds to information in the environmental information indicating whether the vehicle changes lanes.

4. The lane level map evaluation method according to claim 1, characterized in that: The evaluating the simulated driving path based on the actual driving path comprises: Dividing the real driving path into a plurality of real sub-paths, and dividing the simulated driving path into a plurality of simulated sub-paths, wherein the real sub-paths correspond to the simulated sub-paths in a one-to-one manner; Determine the lanes where each segment of the real sub-path and each segment of the simulated sub-path are located respectively; The real lane where each real sub-path is located is compared with the simulated lane where its corresponding simulated sub-path is located to determine whether the real lane is consistent with the simulated lane.

5. The lane level map evaluation method according to claim 3, characterized in that: The step of dividing the real driving path into a plurality of real sub-paths and dividing the simulated driving path into a plurality of simulated sub-paths comprises: Taking a real intersection in the real driving path as a first node, and based on the first node, determining a path between two adjacent first nodes as the real sub-path; And taking a simulated intersection in the simulated driving path as a second node, and based on the second node, determining a path between two adjacent second nodes as the simulated sub-path.

6. The lane level map evaluation method according to claim 1, characterized in that: The target road section is the road section that the vehicle passes through from starting to shutting down. The step of obtaining the simulated driving path includes: Determining the position of the vehicle when it is started as the starting point of the target section, and determining the position of the vehicle when it is turned off as the end point of the target section; Based on the starting point and the end point, the simulated driving path is generated on the lane-level map.

7. The lane level map evaluation method according to claim 1, characterized in that: Including a plurality of the target road sections, the evaluation method further comprises: Obtaining evaluation results corresponding to lane data of the lane-level map on a plurality of target road sections respectively; In response to the target road section where the vehicle is currently located, the automatic driving function of the vehicle is controlled to be turned on or off based on the evaluation result of the lane data corresponding to the current target road section.

8. The lane level map evaluation method according to claim 1, characterized in that: The evaluation method also includes: Real-time detection of whether the autonomous driving perception system and autonomous driving positioning system of the vehicle are in working state; When it is determined that the autonomous driving perception system and the autonomous driving positioning system are both in the working state, the lane level map is evaluated.

9. A lane-level map evaluation device, characterized in that: Applied to an autonomous driving vehicle, the evaluation device comprises: A first acquisition unit, used to acquire a lane-level map to be evaluated; A second acquisition unit is used to acquire real driving data of the vehicle on the target road section when the vehicle is under manual driving, wherein the real driving data includes driving parameters and a real driving path; A third acquisition unit is used to acquire lane data of the lane-level map on the target road section, wherein the lane data includes map elements and a simulated driving path of the vehicle on the target road section generated by the lane-level map when the vehicle is in the automatic driving state; an evaluation unit, configured to evaluate the map element based on the driving parameter, and to evaluate the simulated driving path based on the real driving path; A determination unit is used to determine whether the lane data meets the usage criteria of the vehicle's automatic driving based on the evaluation result.

10. A vehicle, characterized in that: The vehicle includes an evaluation module, and the evaluation module is used to implement the lane-level map evaluation method as described in any one of claims 1-8.

Images