Method and device for obtaining map matching truth value, electronic equipment and storage medium

Abstract

The invention discloses a method and a device for acquiring a map matching truth value, electronic equipment and a storage medium. The method comprises the following steps: acquiring GNSS data and IMU data, and generating a driving track based on a map; lane sensing data are obtained, the lane sensing data and lane lines of the map are subjected to position matching, the driving track is optimized, a target driving track is obtained, and the lane sensing data comprise vehicle pose data; and obtaining a candidate lane ID set according to the target driving track, calculating the vehicle pose positioning data and the path overhead of the candidate lane based on the time sequence, and calculating the lane track with the minimum global path overhead based on the target driving track and the path overhead. According to the method, the track matching result is optimized by matching the lane sensing data with the lane lines of the map, the road binding truth value is obtained based on the lane track with the minimum global path overhead, and the absolute precision error of the map is eliminated.

Description

Technical Field

[0001] The present invention relates to the field of map data processing, and in particular to a method, device, electronic device and storage medium for obtaining a true value of map matching. Background Art

[0002] In the prior art, map matching is the process of associating GNSS trajectory points with the road network of the map, aiming to determine the actual road and driving path of the vehicle. Due to GNSS errors and trajectory sampling rate limitations, accurate matching of trajectory points is crucial. This process includes key steps such as data collection, map data preparation, trajectory preprocessing, and map matching algorithms. Ultimately, map matching generates a sequence of road sections along which the vehicle travels, providing key information for applications such as navigation, traffic management, and intelligent driving. In the face of GNSS errors and complex road networks, it is crucial to develop more accurate matching algorithms to meet the needs of different application scenarios, such as autonomous driving and intelligent transportation systems.

[0003] However, the accuracy measurement and accuracy optimization of the map matching algorithm rely on the true value of the map matching, but such true value is very difficult to obtain and often requires a lot of manual annotation and proofreading. Summary of the invention

[0004] In response to the above technical problems, the present invention provides a method, device, electronic device and storage medium for obtaining the true value of map matching, which can automatically obtain the true value of map matching and better assist manual proofreading to initially improve the accuracy of the true value of map matching.

[0005] A first aspect of the present invention provides a method for obtaining a true value of a map match, comprising: Obtain GNSS data and IMU data and generate driving trajectories based on maps; Acquire lane perception data and positionally match the lane perception data with lane lines of the map to optimize the driving trajectory to obtain a target driving trajectory, wherein the lane perception data includes vehicle posture data; A candidate lane ID set is obtained according to the target driving trajectory, the vehicle posture positioning data and the path cost of the candidate lanes are calculated based on the time series, and the lane trajectory with the minimum global path cost is calculated based on the target driving trajectory and the path cost.

[0006] In a possible implementation, the calculating of the path cost of the vehicle posture positioning data and the candidate lanes based on the time series, and the calculating of the lane trajectory with the minimum global path cost based on the target driving trajectory include: Calculate the path cost between the vehicle posture positioning data and the corresponding candidate lane at each moment, where there is at least one candidate lane; Calculate the transfer path cost from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment; A directed acyclic graph is constructed according to the path cost and the transfer path cost, and an optimal path of the directed acyclic graph is calculated, wherein the optimal path corresponds to a lane trajectory with the minimum global path cost.

[0007] In a possible implementation, the calculating of the path cost of the vehicle posture positioning data and the candidate lanes based on the time series, and the calculating of the lane trajectory with the minimum global path cost based on the target driving trajectory include: The path cost of the vehicle posture positioning data and the corresponding candidate lane at each moment is calculated according to the following formula: CostSinglePoint({x i})=|p i -LaneCenter LID |+α|p i -LaneHeading LID |; The transfer path cost from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment is calculated according to the following formula: CostTransfer({x i →x i+1})=transible of Lane x i to Lane x i+1 ; Calculate the optimal path of the directed acyclic graph according to the following formula; Cost({x i})=CostSinglePoint({x i})+CostTransfer({x i →x i+1}); Among them, α is the conversion coefficient between angle and distance, CostSinglePoint({x i}) is the path cost of the location point, p i is the location coordinate, LaneCenter LID LaneHeading is the lane centerline coordinate, LID is the lane heading angle, CostTransfer({x i →x i+1}) is the transfer path cost, Cost({x i}) is the global path cost.

[0008] In a possible implementation, obtaining a candidate lane ID set according to the target driving trajectory includes: All lanes within a preset distance in the vehicle's driving direction are traversed, the error between the vehicle position and the lane centerline is calculated, and lanes whose errors are less than a preset threshold are obtained to form the lane ID set.

[0009] In a possible implementation, the acquiring lane perception data and performing position matching of the lane perception data with a lane line of the map includes: Construct a factor graph including GNSS position and IMU pre-integration, and obtain the position and attitude through nonlinear optimization; Calculate the rotation R and translation T required to convert the lane perception data to the lane lines of the map; The rotation R and the translation T are added to the factor graph to obtain the target driving trajectory.

[0010] In a possible implementation, the calculating of the path cost of the vehicle posture positioning data and the candidate lanes based on the time series, and the calculating of the lane trajectory with the minimum global path cost based on the target driving trajectory include: Calculate a first probability that the current position point belongs to a candidate lane according to the vehicle posture positioning data and the target driving trajectory; Calculating a second probability of transferring from the previous candidate lane to the current candidate lane; A lane path with a maximum probability in the vehicle driving lane is calculated according to the target driving trajectory, and the lane path with a maximum probability corresponds to the lane trajectory with the minimum global path cost.

[0011] In a possible implementation, the method further includes: matching the lane trajectory with the smallest global path cost with the lane line on the map to obtain a road binding result, and generating a visual map matching interface for the road binding result based on geographic information software.

[0012] A second aspect of the present invention provides a device for obtaining a true value of map matching, comprising: Trajectory generation module, used to obtain GNSS data and IMU data and generate driving trajectories based on maps; An optimization module, used to obtain lane perception data and positionally match the lane perception data with lane lines of the map to optimize the driving trajectory to obtain a target driving trajectory, wherein the lane perception data includes vehicle posture data; The binding road generation module is used to obtain a candidate lane ID set according to the target driving trajectory, calculate the vehicle posture positioning data and the path cost of the candidate lane based on the time series, and calculate the lane trajectory with the minimum global path cost based on the target driving trajectory and the path cost.

[0013] For example, the path cost between the vehicle posture positioning data and the corresponding candidate lane at each moment is calculated, and there is at least one candidate lane; the transfer path cost from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment is calculated; a directed acyclic graph is constructed according to the path cost and the transfer path cost, and the optimal path of the directed acyclic graph is calculated, and the optimal path corresponds to the lane trajectory with the smallest global path cost.

[0014] For another example, a first probability that the current position point belongs to a candidate lane is calculated based on the vehicle posture positioning data and the target driving trajectory; a second probability of transferring from the previous candidate lane to the current candidate lane is calculated; and a lane path with the highest probability in the vehicle's driving lane is calculated based on the target driving trajectory, and the lane path with the highest probability corresponds to the lane trajectory with the lowest global path cost.

[0015] According to a third aspect of the present invention, there is provided an electronic device, comprising: At least one processor; and at least one memory in communication with the processor, wherein: the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method for obtaining a map matching true value as described in the first aspect of an embodiment of the present invention.

[0016] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a computer, the method for obtaining a map matching true value as described in the first aspect of the embodiment of the present invention is executed.

[0017] The present invention optimizes the trajectory matching result by matching lane perception data with the lane lines of the map, obtains the road truth value based on the lane trajectory with the minimum global path cost, and eliminates the absolute accuracy error of the map; the present invention can obtain the truth value of map matching in an automated manner, which is more efficient than giving the truth value manually and can reduce the error rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The figure is a flowchart of a method for obtaining a true value of map matching in an embodiment of the present invention.

[0019] Figure 2 Schematic diagram of global path overhead in an embodiment of the present invention.

[0020] Figure 3 This is a flow chart of the results of matching lane perception data with lane line positions on a map in an embodiment of the present invention.

[0021] Figure 4 for Figure 3 Schematic diagram of lane lines corresponding to the matching results.

[0022] Figure 5 A factor graph containing GNSS position and IMU pre-integration is constructed in an embodiment of the present invention.

[0023] Figure 6 This is a schematic diagram of the matching results after adding parameters to the factor graph in an embodiment of the present invention.

[0024] Figure 7 The figure is a schematic diagram of a process of calculating a lane trajectory with the minimum global path cost in an embodiment of the present invention.

[0025] Figure 8 The figure is a schematic diagram of visualizing the binding path in an embodiment of the present invention.

[0026] Fig. 9 The module diagram of a device for obtaining a map matching true value in an embodiment of the present invention.

[0027] Fig.10 FIG. 1 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0029] It should be understood that the terms "first", "second", and "third" in the claims, specifications, and drawings disclosed in the present invention are used to distinguish different objects rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present disclosure indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections. It should also be understood that the terms used in this disclosure specification are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure.

[0030] Map truth matching usually refers to the process of matching the actual collected GPS trajectory points with the road network on the map. This process is very important for trajectory preprocessing because it can correct the position deviation caused by GPS error and convert the trajectory sequence into a road segment sequence. The accuracy of the map matching algorithm will be affected by GPS error and low trajectory sampling rate. Map matching technology based on environmental features is the key to positioning technology, which can be used for the positioning of autonomous vehicles. The most popular map matching positioning algorithms include iterative closest point (ICP) and normal distribution transform (NDT).

[0031] "Road binding" is a professional term in the field of map matching, which refers to the process of matching the collected GPS track points with the road network on the map. This process can correct the position deviation caused by GPS error and convert the track sequence into a road segment sequence. In short, road binding is to bind GPS track points to the most likely road on the map, thereby improving the accuracy of the location data, which is the process of true value matching.

[0032] See also Figure 1 As shown, the present invention provides a method for obtaining a true value of map matching, which provides a true value for binding roads in map matching, so as to facilitate parameter adjustment and testing of intelligent driving modules. The above method specifically includes the following steps.

[0033] Step 100: Acquire GNSS data and IMU data and generate a driving trajectory based on a map.

[0034] GNSS data (Global Navigation Satellite System) is a type of satellite navigation system data that provides global coverage of positioning, navigation and timing (PNT) services. The GNSS system sends signals to the surface of the earth via satellites, and receivers on the ground determine their own position, speed and time by receiving these signals.

[0035] IMU (Inertial Measurement Unit) is a device used to measure the three-axis attitude angle (or angular rate) and acceleration of an object, and is usually used to obtain the heading angle of a vehicle.

[0036] During intelligent driving, a navigation trajectory is usually generated based on the destination. The vehicle travels on the navigation trajectory and uses GNSS data and IMU data to locate the vehicle's current driving trajectory, such as driving at a certain location on a certain road section at a certain point in time.

[0037] Step 200: Acquire lane perception data and positionally match the lane perception data with lane lines of the map to optimize the driving trajectory to obtain a target driving trajectory, wherein the lane perception data includes vehicle posture data.

[0038] Lane detection technology is a key component of autonomous navigation for self-driving vehicles. It can help vehicles identify road boundaries and stay on the correct driving path. Vehicles usually obtain the lane lines of the road section and the position of the specific lane lines of the vehicle through visual perception and radar perception, that is, the vehicle posture positioning data.

[0039] The lane perception data is position-matched with the lane lines of the map, the curve coefficients in the ego-vehicle coordinate system are obtained through ICP or by fitting the curve, the translation data required to convert the lane lines perceived by the vehicle to the lane lines of the map are calculated, and based on the translation data, the ego-vehicle posture is projected into the map coordinate system to optimize the driving trajectory.

[0040] Step 300: Obtain a candidate lane ID set according to the target driving trajectory, calculate the vehicle posture positioning data and the path cost of the candidate lanes based on the time series, and calculate the lane trajectory with the minimum global path cost based on the target driving trajectory and the path cost.

[0041] In this step, the geometric center line of the lane in the high-precision map is matched with the target driving trajectory to calculate the global optimal binding result. The lane ID set can be selected by calculating the drivable paths of the vehicle at multiple times.

[0042] For example, by inputting the time series {t i}, the corresponding pose sequence {p i ,A i},p i is the position coordinate, A i is the attitude angle (generally the heading angle), which is the time series data of the vehicle positioning posture: {t i ,pose i The vehicle’s positioning posture includes position and attitude, usually expressed in two-dimensional coordinates (lat i ,lon i ,heading i ); and the geometric point coordinates and heading angle of the map lane centerline.

[0043] In some embodiments, all lanes within a preset distance in the vehicle's driving direction are traversed, the error between the vehicle position and the lane centerline is calculated, and lanes with errors less than a preset threshold are obtained to form the lane ID set.

[0044] Possible lanes are selected based on the geometric relationship between the vehicle positioning posture and the lanes on the map. At a certain moment, the vehicle position and posture are known, and the geometric position of the centerline of each lane on the map and the lane ID are also known. Traverse all lanes within a certain distance (for example, 100 meters) near the vehicle, calculate the error between the vehicle position and the lane centerline, and obtain the ID set of lanes whose error is less than a certain threshold.

[0045] The lane ID set is: x i ∈{Lane ID ,for|p i -LaneCenter LID|<N*E}; where E is the positioning error level, N is the magnification factor, and LaneCenter LID is the coordinate of the lane center line.

[0046] Output the Lane i sequence {LID i} corresponding to the time series {t ID}(or pose sequence {p i}). That is, the Lane i to be optimized at time t ID is {x i}. As Figure 2 shown, candidate lane C was found at time t1, candidate lane C was found at time t2, candidate lanes B and C were found at time t3; candidate lanes B, C, and D were found at time t4, and so on.

[0047] Based on this, multiple driving routes available to the vehicle from time t1 to time t4 can be obtained, and the lane trajectory with the minimized cost can be obtained by calculating the path cost, thereby obtaining the true value of the tied road.

[0048] The present invention optimizes the trajectory matching result by matching the lane perception data with the lane lines on the map, and obtains the true value of the tied road based on the lane trajectory with the minimum global path cost, eliminating the absolute accuracy error of the map; the present invention can obtain the true value of the map matching in an automated manner, which is more efficient than the manual method of giving the true value and can reduce the error rate.

[0049] Further, in the above step 200, the obtaining of the lane perception data and the position matching of the lane perception data with the lane lines on the map include the following steps in some embodiments, as Figure 3 shown.

[0050] Step 210: Construct a factor graph including GNSS position and IMU pre-integration, and obtain the position and pose through non-linear optimization.

[0051] Step 220: Calculate the rotation R and translation T required to convert the lane perception data to the lane lines on the map.

[0052] Step 230: Add the rotation R and the translation T to the factor graph to obtain the target driving trajectory.

[0053] As Figure 4 shown, the factor graph of GNSS position and IMU pre-integration corresponds to 6D-Pose (3D map), and the 2D map corresponds to 3D-Pose. The GNSS factor and IMU are optimization constraint terms, GNSS is the observation, 6D-Pose is the pose state, and the position and pose (6D-Pose) can be obtained through non-linear optimization.

[0054] The lane perception data establishes the lane line coordinates based on the vehicle's own vehicle coordinate system. The lane lines in the lane perception data can be projected into the coordinate system of the map based on the coordinate projection method. By comparing with the lane lines on the map, the lane lines in the lane perception data and the lane lines on the map can be obtained: rotation R and translation T. Because it is two-dimensional data, only the lateral translation data T is taken. The rotation R and the translation T are added to the factor graph, and the matching result, that is, the posture observation data, can be obtained through calculation. Specifically, Figure 5 and Figure 6 As shown, Figure 5 is a factor graph containing GNSS position and IMU pre-integration, Figure 5 After adding the rotation R and the translation T to the factor graph, it becomes Figure 6 .

[0055] like Figure 7 As shown, in the above step 300, the path cost of the vehicle posture positioning data and the candidate lanes is calculated based on the time series, and the lane trajectory with the minimum global path cost is calculated based on the target driving trajectory. In some embodiments, the following steps are included.

[0056] Step 310: Calculate the path cost between the vehicle posture positioning data and the corresponding candidate lane at each moment, where there is at least one candidate lane.

[0057] Step 320: Calculate the transfer path cost of transferring from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment.

[0058] Step 330: construct a directed acyclic graph according to the path cost and the transfer path cost, and calculate the optimal path of the directed acyclic graph, wherein the optimal path corresponds to the lane trajectory with the minimum global path cost.

[0059] Specifically, it can be calculated by the following formula. Where α is the conversion coefficient between angle and distance, CostSinglePoint({x i}) is the path cost of the location point, p i is the location coordinate, LaneCenter LID LaneHeading is the lane centerline coordinate, LID is the lane heading angle, CostTransfer({x i →x i+1}) is the transfer path cost, Cost({x i}) is the global path cost.

[0060] The path cost of the vehicle posture positioning data and the corresponding candidate lane at each moment is calculated according to the following formula: CostSinglePoint({x i})=|p i -LaneCenter LID |+α|p i -LaneHeading LID |.

[0061] That is, construct the path cost of a single point, such as Figure 2 The path cost at time t1 in the figure. For each moment, there may be multiple candidate lanes, and then the path cost is calculated for each candidate lane.

[0062] refer to Figure 2 As shown in the figure, candidate lane C is found at time t1, and CostSinglePoint(t1,x=C) can be calculated. Candidate lane C is found at time t2, and CostSinglePoint(t2,x=C) can be calculated. Candidate lane B is found at time t3, and CostSinglePoint(t3,x=B) and candidate lane C are found, and CostSinglePoint(t3,x=C) can be calculated. The path costs at other times can be obtained by analogy.

[0063] Then, for each moment, the transfer path cost from the candidate lane at the previous moment to the candidate lane at the current moment is calculated. If there are m candidate lanes at the previous moment and n candidate lanes at the current moment, there are m×n transfer path costs.

[0064] The transfer path cost from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment is calculated according to the following formula: CostTransfer({x i →x i+1})=transible of Lane x i to Lane x i+1 .

[0065] The transfer path cost corresponds to Figure 2 The middle arrows, each of which corresponds to a transfer path cost.

[0066] Finally, the path cost at each moment and the transfer path cost of each path transfer are obtained, that is, a directed acyclic graph (DAG) is obtained, and the optimal (minimum path cost) path from the starting point to the end point in the directed acyclic graph is solved. The optimal path of the directed acyclic graph is calculated according to the following formula: Cost({x i})=CostSinglePoint({x i})+CostTransfer({x i →x i+1}).

[0067] The corresponding node on the optimal path indicates the lane that the vehicle is most likely to be in at that moment. Figure 2 The optimal path in is t1, C->t2, C->t3, C->t4, C->t5, D->t6, E->t7, F->t8, F->t9, F. That is, the binding result at each moment is obtained.

[0068] In some other embodiments, in the above step 300, the path cost of the vehicle posture positioning data and the candidate lanes is calculated based on the time series, and the lane trajectory with the minimum global path cost is calculated based on the target driving trajectory, including the following alternative implementation method, that is, using the form of confidence probability accumulation instead of the form of path cost accumulation for calculation.

[0069] Step 350: Calculate a first probability that the current position point belongs to a candidate lane based on the vehicle posture positioning data and the target driving trajectory.

[0070] The probability that a vehicle is in a certain lane at a certain moment is calculated by the following formula: The lane ID set is xi∈{Lane ID,for|pi-LaneCenterLID| <N*E},N S is the normalization coefficient of a single point (the coefficient is the number of lanes), and P is the probability.

[0071] Step 360: Calculate a second probability of transferring from the previous candidate lane to the current candidate lane.

[0072] The probability of a vehicle transferring from the previous lane to the current lane is calculated by the following formula: N T is the normalized coefficient of the transfer process (the coefficient is the number of lanes transferred from the previous moment to the current moment), and P is the probability.

[0073] Step 370: Calculate the lane path with the highest probability in the vehicle driving lane according to the target driving trajectory, wherein the lane path with the highest probability corresponds to the lane trajectory with the lowest global path cost.

[0074] The total probability is calculated by the following formula: Wherein P represents the product of Formula 1 and Formula 2.

[0075] In this embodiment, the maximum probability of the vehicle driving lane corresponds to the lane trajectory with the minimum path cost in the above embodiment.

[0076] Furthermore, if Figure 8 As shown, the method further includes: matching the lane trajectory with the smallest global path cost with the lane line on the map to obtain a road binding result, and generating a visual map matching interface for the road binding result based on geographic information software.

[0077] The lanes and track points are projected onto the plane. The lane and road binding results are colored according to the lane ID. The same ID maps to the same color. For example, using the open source software QGIS to generate Figure 7 The visual binding result is shown. If the color of the binding result is inconsistent with the color of the road (usually the nearest lane) that is manually determined, there is a problem with the generated result. Figure 8 In the figure, the lane line with lane ID 9219 is the road binding result. It can be seen from the figure that the lane 9219 is bound as the driving route at multiple times.

[0078] like Fig. 9 As shown, the present invention also provides a device for obtaining a true value of map matching, comprising: a trajectory generation module 91, an optimization module 92, and a binding route generation module 93.

[0079] The trajectory generation module 91 is used to obtain GNSS data and IMU data and generate a driving trajectory based on a map; the optimization module 92 is used to obtain lane perception data and positionally match the lane perception data with the lane lines of the map to optimize the driving trajectory to obtain a target driving trajectory, wherein the lane perception data includes vehicle posture data; the binding road generation module 93 is used to obtain a candidate lane ID set according to the target driving trajectory, calculate the path cost of the vehicle posture positioning data and the candidate lanes based on a time series, and calculate the lane trajectory with the minimum global path cost based on the target driving trajectory and the path cost.

[0080] For example, the path cost between the vehicle posture positioning data and the corresponding candidate lane at each moment is calculated, and there is at least one candidate lane; the transfer path cost from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment is calculated; a directed acyclic graph is constructed according to the path cost and the transfer path cost, and the optimal path of the directed acyclic graph is calculated, and the optimal path corresponds to the lane trajectory with the smallest global path cost.

[0081] For another example, a first probability that the current position point belongs to a candidate lane is calculated based on the vehicle posture positioning data and the target driving trajectory; a second probability of transferring from the previous candidate lane to the current candidate lane is calculated; and a lane path with the highest probability in the vehicle's driving lane is calculated based on the target driving trajectory, and the lane path with the highest probability corresponds to the lane trajectory with the lowest global path cost.

[0082] For the description of the device for obtaining the true value of map matching, reference may be made to the method for obtaining the true value of map matching, which will not be described in detail.

[0083] like Fig.10 As shown, an embodiment of the present invention further discloses an electronic device, including: At least one processor; and at least one memory in communication with the processor, wherein: the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method for obtaining the true value of map matching.

[0084] The embodiment of the present invention also discloses a readable storage medium.

[0085] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for obtaining a true value of map matching described in the above embodiment.

[0086] It can be understood that computer-readable storage media may include: any entity or device capable of carrying a computer program, recording media, USB flash drives, mobile hard disks, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc. A computer program includes computer program code. The computer program code may be in source code form, object code form, executable file, or some intermediate form, etc. A computer-readable storage medium may include: any entity or device capable of carrying a computer program code, recording media, USB flash drives, mobile hard disks, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc.

[0087] In certain embodiments of the present invention, the electronic device may include a controller or a processor, and the controller is a single-chip microcomputer chip that integrates a processor, a memory, a communication module, etc. The processor may refer to the processor contained in the controller. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.

[0088] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0089] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0090] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for obtaining a true map matching value, characterized in that: include: Obtain GNSS data and IMU data and generate driving trajectories based on maps; Acquire lane perception data and positionally match the lane perception data with lane lines of the map to optimize the driving trajectory to obtain a target driving trajectory, wherein the lane perception data includes vehicle posture data; A candidate lane ID set is obtained according to the target driving trajectory, the vehicle posture positioning data and the path cost of the candidate lanes are calculated based on the time series, and the lane trajectory with the minimum global path cost is calculated based on the target driving trajectory and the path cost.

2. The method for obtaining a true map matching value according to claim 1, characterized in that: The method of calculating the path cost of the vehicle posture positioning data and the candidate lanes based on the time series, and calculating the lane trajectory with the minimum global path cost based on the target driving trajectory, includes: Calculate the path cost between the vehicle posture positioning data and the corresponding candidate lane at each moment, where there is at least one candidate lane; Calculate the transfer path cost from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment; A directed acyclic graph is constructed according to the path cost and the transfer path cost, and an optimal path of the directed acyclic graph is calculated, wherein the optimal path corresponds to a lane trajectory with the minimum global path cost.

3. The method for obtaining a true map matching value according to claim 2, characterized in that: The method of calculating the path cost of the vehicle posture positioning data and the candidate lanes based on the time series, and calculating the lane trajectory with the minimum global path cost based on the target driving trajectory, includes: The path cost of the vehicle posture positioning data and the corresponding candidate lane at each moment is calculated according to the following formula: CostSinglePoint({x i }) = |p i - LaneCenter LID | + α |p i - LaneHeading LID |; The transfer path cost from the candidate lane where the vehicle was at the previous moment to the candidate lane where the vehicle is at the current moment is calculated according to the following formula: CostTransfer({x i →x i+1 }) = transible of Lane x i to Lane x i+1 ; Calculate the optimal path of the directed acyclic graph according to the following formula; Cost({x i }) = CostSinglePoint({x i }) + CostTransfer({x i →x i+1 }); Among them, α is the conversion coefficient between angle and distance, CostSinglePoint({x i }) is the path cost of the location point, p i is the location coordinate, LaneCenter LID LaneHeading is the lane centerline coordinate, LID is the lane heading angle, CostTransfer({x i →x i+1 }) is the transfer path cost, Cost({x i }) is the global path cost.

4. The method for obtaining a true map matching value according to claim 2, characterized in that: The step of obtaining a candidate lane ID set according to the target driving trajectory includes: All lanes within a preset distance in the vehicle's driving direction are traversed, the error between the vehicle position and the lane centerline is calculated, and lanes whose errors are less than a preset threshold are obtained to form the lane ID set.

5. The method for obtaining a true map matching value according to claim 2, characterized in that: The acquiring lane perception data and performing position matching of the lane perception data with the lane line of the map includes: Construct a factor graph including GNSS position and IMU pre-integration, and obtain the position and attitude through nonlinear optimization; Calculate the rotation R and translation T required to convert the lane perception data to the lane lines of the map; The rotation R and the translation T are added to the factor graph to obtain the target driving trajectory.

6. The method for obtaining a true map matching value according to claim 1, characterized in that: The method of calculating the path cost of the vehicle posture positioning data and the candidate lanes based on the time series, and calculating the lane trajectory with the minimum global path cost based on the target driving trajectory, includes: Calculate a first probability that the current position point belongs to a candidate lane according to the vehicle posture positioning data and the target driving trajectory; Calculating a second probability of transferring from the previous candidate lane to the current candidate lane; A lane path with a maximum probability in the vehicle driving lane is calculated according to the target driving trajectory, and the lane path with a maximum probability corresponds to the lane trajectory with the minimum global path cost.

7. The method for obtaining a true map matching value according to claim 1, characterized in that: The method further includes: matching the lane trajectory with the smallest global path cost with the lane line on the map to obtain a road binding result, and generating a visual map matching interface for the road binding result based on geographic information software.

8. A device for obtaining a true value of map matching, characterized in that: include: Trajectory generation module, used to obtain GNSS data and IMU data and generate driving trajectories based on maps; An optimization module, used to obtain lane perception data and positionally match the lane perception data with lane lines of the map to optimize the driving trajectory to obtain a target driving trajectory, wherein the lane perception data includes vehicle posture data; The binding road generation module is used to obtain a candidate lane ID set according to the target driving trajectory, calculate the vehicle posture positioning data and the path cost of the candidate lane based on the time series, and calculate the lane trajectory with the minimum global path cost based on the target driving trajectory and the path cost.

9. An electronic device, characterized in that: include: at least one processor; and at least one memory communicatively connected to the processor, wherein: the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the method for obtaining a map matching true value as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a computer, the method for obtaining a map matching true value as claimed in any one of claims 1 to 7 is executed.

Images