A lane matching method and related apparatus

Abstract

The application relates to the computer technical field and provides a lane matching method and related devices to improve the accuracy of lane matching, wherein the method comprises the following steps: using real-time positioning information of a target vehicle generated by a terminal device to screen a reference lane matching a current position from a map; using noise information in the real-time positioning information to generate error information, and predicting a compensation offset of the current position based on the error information; if the compensation offset meets a set compensation condition, then the current position is corrected based on the compensation offset, and the corrected position is matched with a plurality of candidate lanes in the map to obtain a lane matching result. The accuracy of vehicle positioning is improved by correcting the current position, so that the accuracy of lane matching is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and provides a lane matching method and related apparatus. Background Technology

[0002] Lane matching is a technology that uses location information and map information to match the vehicle's location with lanes on a map, thereby determining the specific lane the vehicle is in.

[0003] In related technologies, lane-level navigation can be performed using terminal devices such as smartphones while the vehicle is in motion. When using terminal devices for lane-level navigation, the positioning information generated by the terminal device is typically used as the vehicle's positioning information. Then, the vehicle's position, as indicated by the positioning information, is matched with each lane to determine the target lane where the vehicle is located.

[0004] However, the vehicle location represented by the location information generated by the terminal may deviate from the actual vehicle location. If the location information generated by the terminal device is used as the vehicle location, the final determined target lane will be inconsistent with the actual lane, affecting the accuracy of the lane matching result and thus affecting the normal use of subsequent route planning, navigation broadcasting and other services. Summary of the Invention

[0005] This application provides a lane matching method and related apparatus to improve the accuracy of lane matching results.

[0006] On one hand, embodiments of this application provide a lane matching method, including:

[0007] The real-time positioning information of the target vehicle generated by the terminal device is obtained, and the real-time positioning information includes: the current position of the target vehicle and at least one noise information of the current position;

[0008] Obtain road network data from the map, and based on the road network data, filter out a reference lane that matches the current location from multiple candidate lanes, and based on the current location, determine the adsorption position of the target vehicle on the center line of the reference lane.

[0009] Based on the at least one piece of noise information, error information is generated, which is used to measure the accuracy of the actual offset between the current position and the adsorption position, and based on the error information, the compensation offset required to adjust the current position to the adsorption position is predicted.

[0010] If the compensation offset meets the set compensation conditions, the current position is corrected based on the compensation offset to obtain the corrected position of the target vehicle, and the target lane matching the corrected position is selected from multiple candidate lanes based on the road network data.

[0011] In one possible implementation, predicting the compensation offset required to adjust the current position to the adsorption position based on the error information includes:

[0012] Based on the actual offset, combined with the historical compensation offset predicted at the previous moment, the observation residual is obtained. The observation residual is used to characterize the difference between the actual offset and the compensation offset predicted at the previous moment.

[0013] Based on the error information, the Kalman gain is obtained;

[0014] Based on the Kalman gain, the observation residual, and the historical compensation offset, the compensation offset required to adjust the current position to the adsorption position is predicted.

[0015] In one possible implementation, determining that the compensation offset meets the set compensation conditions includes:

[0016] Based on the Kalman gain and combined with the historical prediction error from the previous moment, the current prediction error is obtained; wherein, the historical prediction error is used to measure the accuracy of the historical compensation offset, and the current prediction error is used to measure the accuracy of the compensation offset.

[0017] If the current prediction error is less than the set error threshold, then the compensation offset is determined to meet the set compensation conditions.

[0018] In one possible implementation, generating error information based on the at least one piece of noise information includes at least one of the following methods:

[0019] If the at least one piece of noise information includes location-related information, then based on the location-related information, the offset noise is evaluated, and using the offset noise, error information is generated;

[0020] If the at least one piece of noise information includes speed-related information, then based on the speed-related information, speed noise is evaluated, and error information is generated using the speed noise;

[0021] If the at least one noise information includes driving behavior information, then driving behavior noise is evaluated based on the driving behavior information, and error information is generated using the driving behavior noise.

[0022] In one possible implementation, the location-related information includes: the positioning accuracy of the terminal device; then, evaluating the offset noise based on the location-related information includes:

[0023] The lane width of the reference lane is obtained from the road network data;

[0024] Select target information that meets the set selection criteria from the lane width of the reference lane and the actual offset.

[0025] Based on the target information and the positioning accuracy, and combined with the accuracy weight set for the positioning accuracy, the offset noise is obtained.

[0026] In one possible implementation, speed-related information includes: direction of movement and speed of movement.

[0027] In one possible implementation, evaluating driving behavior noise based on the driving behavior information includes:

[0028] If the driving behavior information indicates that the target vehicle was in a candidate lane other than the reference lane at the previous moment, or that the target vehicle deviated from the road boundary line, then the driving behavior noise is set to a first preset value; otherwise, the driving behavior noise is set to a second preset value, wherein the first preset value is higher than the second preset value.

[0029] In one possible implementation, the road network data includes: the lane positions of each of the multiple candidate lanes; the step of filtering a reference lane matching the current position from the multiple candidate lanes based on the road network data includes:

[0030] Based on the lane positions of each of the multiple candidate lanes, the distance between each of the multiple candidate lanes and the current position is obtained, and based on the obtained multiple distances, a reference probability is obtained that the current position belongs to the multiple candidate lanes;

[0031] Based on the obtained multiple reference probabilities, candidate lanes that meet the set lane selection conditions are selected from the multiple candidate lanes and used as reference lanes that match the current position.

[0032] In one possible implementation, obtaining the reference probability that the current position belongs to one of the multiple candidate lanes based on the obtained multiple distances includes:

[0033] For each of the multiple candidate lanes, perform the following operations:

[0034] Based on the distance between a candidate lane and the current position, the distance factor between the target vehicle and the candidate lane is obtained;

[0035] Based on the distance factor, a reference probability is obtained that the current position belongs to one of the candidate lanes.

[0036] In one possible implementation, obtaining the reference probability that the current location belongs to one of the candidate lanes based on the distance factor includes:

[0037] Based on the driving direction of the target vehicle and the lane direction of the candidate lane, the orientation factor between the target vehicle and the candidate lane is obtained.

[0038] Based on the distance factor and the orientation factor, and combined with a preset fitting function, a reference probability is obtained that the current position belongs to one of the candidate lanes.

[0039] On one hand, embodiments of this application provide a lane matching device, including:

[0040] The data acquisition unit is used to acquire real-time positioning information of the target vehicle generated by the terminal device, the real-time positioning information including: the current position of the target vehicle and at least one noise information of the current position; and to acquire road network data of the map;

[0041] The first matching unit is used to filter out a reference lane that matches the current position from multiple candidate lanes based on the road network data, and to determine the adsorption position of the target vehicle on the lane centerline of the reference lane based on the current position.

[0042] A position compensation unit is configured to generate error information based on the at least one noise information, the error information being used to measure the accuracy of the actual offset between the current position and the adsorption position, and based on the error information, to predict the compensation offset required to adjust the current position to the adsorption position;

[0043] The second matching unit is used to correct the current position based on the compensation offset if the compensation offset meets the set compensation conditions, to obtain the corrected position of the target vehicle, and to filter out the target lane that matches the corrected position from multiple candidate lanes based on the road network data.

[0044] In one possible implementation, when predicting the compensation offset required to adjust the current position to the adsorption position based on the error information, the position compensation unit is specifically used for:

[0045] Based on the actual offset, combined with the historical compensation offset predicted at the previous moment, the observation residual is obtained. The observation residual is used to characterize the difference between the actual offset and the compensation offset predicted at the previous moment.

[0046] Based on the error information, the Kalman gain is obtained;

[0047] Based on the Kalman gain, the observation residual, and the historical compensation offset, the compensation offset required to adjust the current position to the adsorption position is predicted.

[0048] In one possible implementation, the position compensation unit is also used for:

[0049] Based on the Kalman gain and combined with the historical prediction error from the previous moment, the current prediction error is obtained; wherein, the historical prediction error is used to measure the accuracy of the historical compensation offset, and the current prediction error is used to measure the accuracy of the compensation offset.

[0050] If the current prediction error is less than the set error threshold, then the compensation offset is determined to meet the set compensation conditions.

[0051] In one possible implementation, when generating error information based on the at least one piece of noise information, the position compensation unit is specifically configured to perform at least one of the following methods:

[0052] If the at least one piece of noise information includes location-related information, then based on the location-related information, the offset noise is evaluated, and using the offset noise, error information is generated;

[0053] If the at least one piece of noise information includes speed-related information, then based on the speed-related information, speed noise is evaluated, and error information is generated using the speed noise;

[0054] If the at least one noise information includes driving behavior information, then driving behavior noise is evaluated based on the driving behavior information, and error information is generated using the driving behavior noise.

[0055] In one possible implementation, the location-related information includes: the positioning accuracy of the terminal device; then the location compensation unit is further used for:

[0056] The lane width of the reference lane is obtained from the road network data;

[0057] Select target information that meets the set selection criteria from the lane width of the reference lane and the actual offset.

[0058] Based on the target information and the positioning accuracy, and combined with the accuracy weight set for the positioning accuracy, the offset noise is obtained.

[0059] In one possible implementation, speed-related information includes: direction of movement and speed of movement.

[0060] In one possible implementation, the position compensation unit is also used for:

[0061] If the driving behavior information indicates that the target vehicle was in a candidate lane other than the reference lane at the previous moment, or that the target vehicle deviated from the road boundary line, then the driving behavior noise is set to a first preset value; otherwise, the driving behavior noise is set to a second preset value, wherein the first preset value is higher than the second preset value.

[0062] In one possible implementation, the road network data includes: the lane positions of each of the multiple candidate lanes; when filtering a reference lane that matches the current position from the multiple candidate lanes based on the road network data, the first matching unit is specifically used for:

[0063] Based on the lane positions of each of the multiple candidate lanes, the distance between each of the multiple candidate lanes and the current position is obtained, and based on the obtained multiple distances, a reference probability is obtained that the current position belongs to the multiple candidate lanes;

[0064] Based on the obtained multiple reference probabilities, candidate lanes that meet the set lane selection conditions are selected from the multiple candidate lanes and used as reference lanes that match the current position.

[0065] In one possible implementation, when obtaining the reference probability that the current position belongs to one of the multiple candidate lanes based on the obtained multiple distances, the first matching unit is specifically used for:

[0066] For each of the multiple candidate lanes, perform the following operations:

[0067] Based on the distance between a candidate lane and the current position, the distance factor between the target vehicle and the candidate lane is obtained;

[0068] Based on the distance factor, a reference probability is obtained that the current position belongs to one of the candidate lanes.

[0069] In one possible implementation, when obtaining the reference probability that the current position belongs to the candidate lane based on the distance factor, the first matching unit is specifically used for:

[0070] Based on the driving direction of the target vehicle and the lane direction of the candidate lane, the orientation factor between the target vehicle and the candidate lane is obtained.

[0071] Based on the distance factor and the orientation factor, and combined with a preset fitting function, a reference probability is obtained that the current position belongs to one of the candidate lanes.

[0072] On one hand, an electronic device is provided, including a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the above-described method.

[0073] On one hand, a computer-readable storage medium is provided, comprising a computer program that, when run on an electronic device, causes the electronic device to perform the steps of any of the methods described above.

[0074] On one hand, a computer program product is provided, the program product including a computer program stored in a computer-readable storage medium, wherein a processor of an electronic device reads from the computer-readable storage medium and executes the computer program, causing the electronic device to perform the steps of any of the methods described above.

[0075] In this embodiment of the application, during the process of lane matching for a target vehicle using a terminal device, the real-time positioning information of the target vehicle generated by the terminal device is first obtained. The real-time positioning information includes the current position of the target vehicle and at least one piece of noise information.

[0076] Next, the road network data of the map is obtained, and based on the road network data, a reference lane matching the current position is selected from multiple candidate lanes. Based on the current position, the adsorption position of the target vehicle on the center line of the reference lane is obtained.

[0077] Considering that the real-time location information generated by the terminal device may be inaccurate, and the adsorption position determined based on the real-time location information may also be inaccurate, the current position of the target vehicle provided in the real-time location information is corrected to obtain the accurate position of the target vehicle. Then, the accurate position is used to re-match the lane to improve the accuracy of lane matching.

[0078] In the process of correcting the current position, specifically, error information is generated based on at least one piece of noise information to measure the accuracy of the actual offset, which refers to the offset between the current position and the adsorption position. Then, based on the error information, the compensation offset required to adjust the current position to the adsorption position is predicted. By evaluating the error information, the error caused by noise can be quantified, thereby making the corrected vehicle position closer to the accurate position of the target vehicle, which helps to improve the accuracy of lane matching.

[0079] Correcting the current position when the set compensation conditions are met not only reduces the amount of computation and improves lane matching efficiency compared to correcting the current position at every moment, but also avoids the impact of inaccurate compensation offset on the lane matching results, thereby improving the stability of lane matching.

[0080] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0081] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0082] Figure 1 This is a schematic diagram illustrating the application scenario provided in the embodiments of this application;

[0083] Figure 2 This is a flowchart illustrating the lane matching method provided in the embodiments of this application;

[0084] Figure 3 This is a schematic diagram of the map provided in the embodiments of this application;

[0085] Figure 4 This is a schematic diagram of the candidate lanes provided in the embodiments of this application;

[0086] Figure 5 This is a flowchart illustrating the reference lane selection method provided in the embodiments of this application;

[0087] Figure 6 This is a schematic diagram of the adsorption sites provided in the embodiments of this application;

[0088] Figure 7 This is a flowchart illustrating the reference probability calculation method provided in the embodiments of this application;

[0089] Figure 8 This is a logical diagram of the reference lane selection process provided in the embodiments of this application;

[0090] Figure 9 This is a logical diagram illustrating a compensation offset calculation process provided in an embodiment of this application.

[0091] Figure 10 This is a logical diagram of the lane matching process provided in the embodiments of this application;

[0092] Figure 11 This is a logical diagram illustrating another compensation offset calculation process provided in the embodiments of this application;

[0093] Figure 12 This is a schematic diagram of the lane matching device provided in the embodiments of this application;

[0094] Figure 13 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0095] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0096] It is understood that when the embodiments of this application are applied to specific products or technologies, relevant licenses or consents need to be obtained, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0097] To facilitate understanding of the technical solutions provided in the embodiments of this application, some key terms used in the embodiments of this application will be explained below:

[0098] Vehicle Dead Reckoning (VDR) is an algorithm for position estimation based on Global Navigation Satellite System (GNSS) signals and Inertial Measurement Unit (IMU) data. IMUs include, but are not limited to, accelerometers and gyroscopes, where accelerometers detect acceleration data and gyroscopes detect angular velocity data. While GNSS signals provide relatively accurate position information in open environments, their accuracy significantly decreases in environments with signal obstruction or loss, such as tunnels and underground parking lots. In these situations, VDR uses IMU data to provide position estimation through dead reckoning, compensating for insufficient GNSS signal coverage. Although the cumulative error of VDR increases over time, periodic calibration with GNSS data can effectively improve positioning accuracy and reliability. In this embodiment, VDR is used to calculate the vehicle's positioning information based on GNSS signals and IMU data generated by a terminal device placed inside the vehicle. In this document, the positioning information calculated using VDR technology can also be referred to as a positioning point or a VDR positioning point.

[0099] Lane Matching: A technology for lane matching based on VDR positioning points and road map information. Lane matching determines the specific lane a vehicle is in by comparing the location provided by the VDR positioning points with a high-precision road map. In this embodiment, lane matching can be achieved based on GNSS signals and IMU data generated by terminal devices such as mobile phones, as well as map information, without relying on other sensors. Lane matching technology is crucial for functions such as route planning and navigation announcements.

[0100] Hidden Markov Model (HMM): An HMM is a statistical model used to describe stochastic processes with hidden states. An HMM typically consists of a set of hidden states, state transition probabilities, observation symbols, and emission probabilities. In an HMM, it is assumed that the system is in a hidden state at any given time. This state is not directly observable but can be inferred indirectly through observed symbols. The transitions between states are determined by the state transition probability matrix, while the probability of each state generating an observed symbol is determined by the emission probability matrix. HMMs are widely used in fields such as speech recognition, natural language processing, and bioinformatics. In this paper, the hidden state can also be referred to as a state, the state transition probabilities as transition probabilities, and the observed symbols as observation data.

[0101] Kalman Filter (KF): A recursive optimal estimation algorithm suitable for handling state estimation problems of linear dynamic systems. KF recursively estimates the system state by combining the system's dynamic model and observational data, providing the optimal estimate under conditions of noise and uncertainty. KF is widely used in navigation, control, signal processing, and other fields.

[0102] In related technologies, lane matching is usually achieved in the following way: when using a terminal device located in the vehicle for lane-level navigation, the positioning information generated by the terminal device is directly used as the vehicle's positioning information, and then the vehicle's position represented by the vehicle's positioning information is matched with each lane to determine the target lane where the vehicle is located.

[0103] However, the vehicle location represented by the location information generated by the terminal may deviate from the actual vehicle location. Therefore, using the location information generated by the terminal device as the vehicle's location information will result in the final determined target lane being inconsistent with the actual lane, affecting the accuracy of lane matching results and consequently impacting the service quality of subsequent route planning, navigation broadcasting, and other services.

[0104] In the process of lane matching for a target vehicle using a terminal device, the first step is to obtain the real-time positioning information of the target vehicle generated by the terminal device. The real-time positioning information includes the current position of the target vehicle and at least one piece of noise information.

[0105] Next, the road network data of the map is obtained, and based on the road network data, a reference lane matching the current position is selected from multiple candidate lanes. Based on the current position, the adsorption position of the target vehicle on the center line of the reference lane is obtained.

[0106] Considering that the real-time location information generated by the terminal device may be inaccurate, and the adsorption position determined based on the real-time location information may also be inaccurate, the current position of the target vehicle provided in the real-time location information is corrected to obtain the accurate position of the target vehicle. Then, the accurate position is used to re-match the lane to improve the accuracy of lane matching.

[0107] In the process of correcting the current position, specifically, error information is generated based on at least one piece of noise information to measure the accuracy of the actual offset, which refers to the offset between the current position and the adsorption position. Then, based on the error information, the compensation offset required to adjust the current position to the adsorption position is predicted. By evaluating the error information, the error caused by noise can be quantified, thereby making the corrected vehicle position closer to the accurate position of the target vehicle, thus improving the accuracy of lane matching.

[0108] The following is a brief introduction to the application scenarios to which the technical solutions of the embodiments of this application are applicable. It should be noted that the application scenarios described below are only for illustrating the embodiments of this application and are not intended to limit the scope. In specific implementation, the technical solutions provided by the embodiments of this application can be flexibly applied according to actual needs.

[0109] See Figure 1 The diagram illustrates an application scenario provided by an embodiment of this application. In this scenario, a terminal device 101 and a server 102 may be included. The terminal device 101 is located in the target vehicle. The target vehicle refers to the vehicle performing lane matching.

[0110] Terminal device 101 is a device capable of acquiring GNSS signals and IMU data, such as a mobile phone, tablet computer, laptop computer, smart vehicle device, smart wearable device, smartwatch, portable navigation device, smart glasses, etc. In some embodiments, terminal device 101 may also have the ability to perform lane matching using GNSS signals and IMU data.

[0111] In some implementations, the terminal device 101 has an application installed that requires lane matching, and the server 102 is a backend server that provides lane matching functionality. The server 102 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms, but it is not limited to these.

[0112] Terminal device 101 and server 102 can communicate directly or indirectly through one or more networks. The network can be a wired network or a wireless network. For example, the wireless network can be a mobile cellular network or a Wireless-Fidelity (WIFI) network, or other possible networks. This application embodiment does not limit this. Figure 1 The examples shown are merely illustrative; in reality, the number of terminal devices and servers is unlimited and is not specifically limited in the embodiments of this application.

[0113] In one possible implementation, the lane matching method mentioned in this application embodiment can be executed by terminal device 101 or server 102. For example, terminal device 101 obtains real-time positioning information for the target vehicle generated by the terminal device. The real-time positioning information includes: the current position of the target vehicle and at least one noise information of the current position; obtains road network data of the map, and based on the road network data, filters out a reference lane that matches the current position from multiple candidate lanes, and determines the adsorption position of the target vehicle on the lane centerline of the reference lane based on the current position; generates error information based on at least one noise information, and predicts a compensation offset based on the error information; if the compensation offset meets the set compensation conditions, the current position is corrected based on the compensation offset to obtain the corrected position of the target vehicle, and a target lane that matches the corrected position is filtered out from multiple candidate lanes based on the road network data. For example, server 102 obtains real-time positioning information for the target vehicle generated by the terminal device. The real-time positioning information includes: the current position of the target vehicle and at least one noise information of the current position; it obtains road network data from the map, and based on the road network data, selects a reference lane that matches the current position from multiple candidate lanes, and determines the adsorption position of the target vehicle on the center line of the reference lane based on the current position; it generates error information based on at least one noise information, which is used to measure the accuracy of the actual offset between the current position and the adsorption position, and predicts the compensation offset required to adjust the current position to the adsorption position based on the error information; if the compensation offset meets the set compensation conditions, it corrects the current position based on the compensation offset to obtain the corrected position of the target vehicle, and selects a target lane that matches the corrected position from multiple candidate lanes based on the road network data.

[0114] In one possible implementation, terminal device 101 collects GNSS signals and IMU data, and reports the collected GNSS signals and IMU data to server 102; server 102 receives the collected GNSS signals and IMU data, and uses the collected GNSS signals and IMU data to obtain real-time positioning information generated by the terminal device.

[0115] In one possible implementation, both server 102 and terminal device 101 may include one or more processors, memory, and interactive I / O interfaces. Furthermore, server 102 may be configured with a database for storing map information, etc. The memory of server 102 and terminal device 101 may also store program instructions required for execution in the lane matching method provided in this application embodiment. These program instructions, when executed by the processor, can be used to implement the lane matching process provided in this application embodiment.

[0116] See Figure 2The diagram shown is a flowchart of a lane matching method provided in this embodiment of the application. This method is applied to a terminal device or a server, and the specific process is as follows:

[0117] S201. Obtain real-time positioning information for the target vehicle generated by the terminal device. The real-time positioning information includes the current position of the target vehicle and M noise information of the current position.

[0118] In this embodiment, M is a positive integer, and the location information can also be called a location point.

[0119] The target vehicle refers to the vehicle that requires lane-level positioning. For example, if user A needs lane-level positioning while driving, the target vehicle could refer to the vehicle being driven by user A. Driving the target vehicle can be manual, assisted driving, or autonomous driving, but is not limited to these methods.

[0120] While the target vehicle is in motion, the user can use a terminal device to determine the lane in which the target vehicle is located, thus providing lane-level positioning. For example, while driving, user A can use their mobile phone to perform lane-level positioning of the vehicle they are driving.

[0121] The current location of the target vehicle can be its latitude and longitude information.

[0122] The terminal device may be located inside the target vehicle or may have other relative positional relationships with the target vehicle; there are no restrictions on this.

[0123] Noise information refers to information that causes an error between the current position of the target vehicle and its actual offset position on the lane centerline.

[0124] In one possible implementation, the noise information includes at least one of the following: location-related information, speed-related information, and driving behavior information, but is not limited to these.

[0125] Location-related information is used to characterize the location factors that affect the accuracy of the actual offset during the target vehicle's movement. For example, location-related information includes, but is not limited to, the positioning accuracy of the terminal device. Positioning accuracy characterizes the accuracy of the current position generated by the terminal device during lane-level positioning. Generally, the lower the positioning accuracy, the more precise the current position. The more precise the current position, the smaller the error between the target vehicle's current position and its position on the lane centerline, meaning the higher the accuracy of the actual offset.

[0126] Speed-related information is used to characterize positional factors that affect the accuracy of the actual offset during the target vehicle's movement. For example, speed-related information includes, but is not limited to, the target vehicle's speed and direction of movement. Generally, when the target vehicle is traveling along the lane centerline, a slower speed will result in a larger error in the actual offset between the current position and the adsorption position. Conversely, a faster speed will result in a smaller error. Of course, if the target vehicle's speed is too high or it deviates from the lane centerline, it may also lead to errors in the actual offset.

[0127] Driving behavior information is used to characterize driving behaviors that affect the accuracy of the actual offset during the target vehicle's journey. For example, driving behavior information includes, but is not limited to, at least one of the following: lane change and road departure situations at adjacent time points. Lane change at adjacent time points characterizes whether the target vehicle changed lanes at those adjacent time points, and road departure situation characterizes whether the current position exceeds the road boundary line, or whether the distance between the current position and the road boundary line exceeds a set distance threshold. Both lane changes and lane departure can lead to errors in the actual offset between the current position and the adsorption position.

[0128] In one possible implementation, the road boundary line can be either a road edge line or a lane boundary line. Generally speaking, a road can contain multiple lanes. The road edge line is the outermost line of the road, used to mark the edge of the road, while the lane boundary line is the line used to separate multiple lanes on the same road.

[0129] In one possible implementation, the terminal device collects GNSS signals and IMU data at a set collection interval, and sends the collected GNSS signals and IMU data to the server so that the server can obtain the current positioning information based on the received GNSS signals and IMU data.

[0130] Taking a mobile phone as an example, please refer to... Figure 3As shown, the terminal device collects GNSS signals and IMU data at times t1, t2, and t3, and sends the collected GNSS signals and IMU data to the server. Based on the GNSS signals and IMU data at time t1, the server infers the current position at time t1, denoted as A1. Based on A1, and combined with error information such as positioning accuracy, movement speed, and movement direction at time t1, the server obtains the positioning information at time t1. Similarly, based on the GNSS signals and IMU data at time t2, the server infers the current position at time t2, denoted as A2. Based on A2, and combined with error information such as positioning accuracy, movement speed, and movement direction at time t2, the server obtains the positioning information at time t2. Finally, based on the GNSS signals and IMU data at time t3, the server infers the current position at time t3, denoted as A3. Based on A3, and combined with error information such as positioning accuracy, movement speed, and movement direction at time t3, the server obtains the positioning information at time t3.

[0131] S202. Obtain the road network data of the map, and based on the road network data, filter out the reference lane that matches the current position from multiple candidate lanes, and obtain the adsorption position of the target vehicle on the center line of the reference lane based on the current position.

[0132] A map refers to an electronic map, also known as a digital map, which is a map stored and viewed digitally using computer technology. Electronic maps can include those with varying degrees of precision; for example, the electronic maps in this application include, but are not limited to, high-definition (HD) maps. HD maps refer to map data that provides lane-level information, primarily for autonomous and assisted driving, with precision down to the centimeter level.

[0133] Road network data includes map data used for lane matching, such as lane data, which includes the location of each lane, lane width, lane centerline, lane boundary line, and lane type. In addition, road network data may also include map data such as road data and traffic sign data. Road network data includes road nodes, road segments, and road shapes, while traffic signs are various road signs, such as speed limit signs, stop signs, and no-left-turn signs.

[0134] Candidate lanes are lanes on the map that match the current location of the target vehicle.

[0135] In one possible implementation, candidate lanes can be determined in the following way:

[0136] Determine the map location corresponding to the current position of the target vehicle;

[0137] Within a preset range using the map location as a reference point, query the lanes on the map whose distance from the map location meets the preset distance conditions;

[0138] The queried lanes are used as candidate lanes.

[0139] The preset range for using the map location as a reference point can be a buffer zone centered on the map location and with a certain value as the radius.

[0140] The preset distance condition can be a preset distance threshold, for example, the preset distance condition is a distance of less than 50m.

[0141] In some cases, there may be multiple lanes that meet the preset distance conditions. Therefore, a certain number of these lanes with the closest distance to the map location can be selected as candidate lanes.

[0142] When using a terminal device for lane-level positioning, the current location information of the target vehicle can be obtained through the positioning function of the terminal device. Based on the map position of the current position, the lanes queried within a preset range with the map position as a reference point are selected as candidate lanes.

[0143] For example, determine the current location's position on the HD map, and then query the three lanes closest to the current location within a 10-meter radius buffer zone centered on that location as candidate lanes. (See also...) Figure 4 As shown, the candidate lanes include: lane 1 (left lane), lane 2 (middle lane) and lane 3 (right lane).

[0144] By combining high-precision map data, candidate lanes that match the current location can be quickly and accurately searched, thus providing more accurate data for subsequent lane matching.

[0145] The process of selecting a reference lane that matches the current position from multiple candidate lanes based on road network data can be called the first lane matching. The matching process in S203 can be called the second lane matching. The difference between the first and second lane matching is that the first lane matching is based on the current position, while the second lane matching is based on the compensation position. This article will only use the first lane matching as an example to explain the lane matching process.

[0146] In one possible implementation, the road network data includes: the width information and lane position of each of the multiple candidate lanes, see [reference]. Figure 5 As shown, based on road network data, reference lanes matching the current location are selected from multiple candidate lanes, including:

[0147] S202-1. Based on the lane positions of multiple candidate lanes, obtain the distances between each candidate lane and the current position, and based on the obtained distances, obtain the reference probability that the current position belongs to multiple candidate lanes.

[0148] The reference probability that the current position belongs to a candidate lane can also be called the reference probability corresponding to that candidate lane.

[0149] The distance between the current position and the candidate lane refers to the distance between the current position and the center line of the candidate lane.

[0150] The lane centerline is the line connecting the midpoints of the two side boundary lines of the lane. For example, if the lane width is 3.5 meters, the lane centerline is 1.75 meters from each side boundary line. It should be noted that, normally, the lane centerline is centered, meaning it is located in the middle of the lane. However, in some special cases, the lane design may not be symmetrical, causing the lane centerline to be off-center. For example, at curves, to improve driving safety and comfort, the lane centerline may be offset towards the outside of the curve.

[0151] The width information of a candidate lane can be either the actual width of the candidate lane or half of its actual width; there is no limitation in this regard. For example, a 6-meter-wide lane and a 3-meter-wide lane correspond to lane width values ​​of 6 and 3, respectively.

[0152] Reference probabilities are used to characterize the probability that a target vehicle is in a candidate lane. For example, if a candidate lane includes three lanes: lane 1, lane 2, and lane 3, the corresponding reference probability can be determined for each of the three lanes. Suppose that the reference probabilities for lanes 1, 2, and 3 are 0.4, 0.5, and 0.8, respectively. This means that the probability of the target vehicle being in lane 1 is 40%, the probability of the target vehicle being in lane 2 is 50%, and the probability of the target vehicle being in lane 3 is 80%.

[0153] Since the calculation process for the reference probability of each candidate lane is the same, the following explanation uses the reference probability of candidate lane i as an example. Candidate lane i can be any one of the candidate lanes.

[0154] Assume the current time is time t, and the real-time location information is the location information at time t. Using... This represents the reference probability corresponding to candidate lane i. This indicates that the current position belongs to candidate lane i.

[0155] In one possible implementation, It can be calculated in the following way:

[0156] First, based on the distance between candidate lane i and the current position, the distance factor between the target vehicle and candidate lane i is obtained;

[0157] Secondly, based on the distance factor, the reference probability of the current position belonging to candidate lane i is obtained.

[0158] The distance factor refers to data reflecting the impact of distance factors on the reference probability when a target vehicle is traveling. Because different candidate lanes are at different distances from the current location, the distance factor value differs when the target object is traveling in different candidate lanes.

[0159] For example, based on the distance between the current position and the center line of lane 1, the distance factor between the target vehicle and lane 1 is obtained. Similarly, based on the distance between the current position and the center line of lane 2, the distance factor between the target vehicle and lane 2 is obtained. Based on the distance between the current position and the center line of lane 3, the distance factor between the target vehicle and lane 3 is obtained. The larger the value of the distance factor, the greater its influence on the reference probability.

[0160] Since the distance factor of the target vehicle is determined based on the distance between the current location and the candidate lane, different distance factors are assigned to different candidate lanes under different distance values. This allows the reference probability to reflect the differences in distance between multiple candidate lanes, thereby distinguishing each candidate lane based on the reference probability and improving the matching accuracy of the reference lane.

[0161] In one possible implementation, the distance factor between the target vehicle and the candidate lane i is obtained based on the distance between the candidate lane i and the current position, including:

[0162] Based on the distance between candidate lane i and the current position, and combined with a preset distance coefficient, a distance factor is determined.

[0163] The distance coefficient is used to scale the distance between candidate lane i and the current position. For example, the distance coefficient can be determined based on the error of the GNSS signal, such as using the standard deviation of the GNSS signal error as the distance coefficient.

[0164] Assuming, using This indicates that at time t, the target vehicle is traveling in candidate lane i. This indicates that at time t, the target vehicle belongs to candidate lane i. This represents the event at time t, where, given that the target vehicle is traveling in candidate lane i, the target vehicle is assigned to candidate lane i. Represents the distance factor.

[0165] For example, The result is obtained using formula (1):

[0166]

[0167] Where, d i m represents the distance between candidate lane i and the current position; d This represents the distance coefficient, for example, m. d The value range is 2 to 4; exp() represents the natural exponential function, that is, the exponential function with the natural constant (e) as the base.

[0168] In one possible implementation, a Gaussian function can also be used to express the distance factor.

[0169] In one possible implementation, the distance factor can be determined directly using the distance between candidate lane i and the current position. In this case, m in formula (1) can be considered as... d The value is set to 1.

[0170] In some cases, to further improve the accuracy of the reference probability, the reference probability can be determined by combining the distance factor with the orientation factor. Specifically, based on the target vehicle's driving direction and the lane direction of candidate lane i, the orientation factor between the target vehicle and candidate lane i is obtained. Then, based on the distance factor and the orientation factor, and combined with a preset fitting function, the reference probability that the current position belongs to candidate lane i is obtained.

[0171] Orientation factor refers to data that reflects the influence of the direction of a target vehicle during its journey on the reference probability.

[0172] For example, the direction of movement is represented by a heading angle, which is the angle between the direction of travel and a reference direction, typically due north. Lane direction is also represented by a lane direction angle.

[0173] Generally speaking, the direction of movement of the target vehicle is consistent with the direction of the lane. The smaller the angle between the direction of movement and the direction of the lane, the greater the probability that the target vehicle belongs to the target vehicle. Therefore, calculating the reference probability by taking into account the direction factor helps to better distinguish each candidate lane and further improve the matching accuracy of the reference lane.

[0174] In one possible implementation, based on the target vehicle's driving direction and combined with the lane direction of candidate lane i, the orientation factor between the target vehicle and candidate lane i is obtained, including:

[0175] Obtain the directional difference between the target vehicle's direction of movement and the lane direction of candidate lane i;

[0176] Based on the orientation difference, the orientation factor between the target vehicle and candidate lane i is obtained.

[0177] For example, the direction of movement is represented by the heading angle, which is the angle between the direction of travel and a reference direction, typically due north. The lane direction is represented by the lane direction angle, which is the angle between the tangent direction of the lane at a point (e.g., the center point on the lane centerline closest to the current position) and the reference direction. The difference between the heading angle and the lane direction angle can be used as the direction difference between the target vehicle's direction of movement and the lane direction of candidate lane i.

[0178] Assuming, using Represents the orientation factor, for example, The result is obtained by calculation using formula (2):

[0179]

[0180] Among them, a i This represents the directional difference between the direction of movement of the target vehicle and the lane direction of candidate lane i.

[0181] In one possible implementation, see [link to relevant documentation]. Figure 7 As shown, based on the distance factor and orientation factor, combined with a preset fitting function, the reference probability of the current position belonging to candidate lane i is obtained, including the following steps:

[0182] S701. Based on the azimuth factor and the distance factor, the emission probability is obtained.

[0183] The launch probability is used to characterize the probability that, at time t, given that the target vehicle is attracted to candidate lane i, the target vehicle belongs to candidate lane i.

[0184] As an example, the emission probability at time t can be obtained by combining the orientation factor and distance factor with an adjustment factor, where the adjustment factor is set according to the lane type.

[0185] Assuming, using Let represent the emission probability at time t, for example. The result is obtained by calculation using formula (3):

[0186]

[0187] Where m0 represents the adjustment factor, the value of which can be set according to the lane type. For example, for normal driving lanes, the value of m0 is set to 1, and for non-normal driving lanes such as shoulders and green belts, the value of m0 is set to 0.

[0188] Taking both distance and direction factors into account in the calculation of launch probability can improve matching accuracy.

[0189] S702, Obtain the transfer probability of the target vehicle.

[0190] The transition probabilities include the probability that the target vehicle will move from one of the multiple candidate lanes to candidate lane i between time t-1 and time t. Taking candidate lane i as lane 1 as an example, the multiple candidate lanes include lane 1, lane 2, and lane 3. The transition probabilities include the probability that the target vehicle will move from lane 1 to lane 1 between time t-1 and time t; the probability that the target vehicle will move from lane 2 to lane 1 between time t-1 and time t; and the probability that the target vehicle will move from lane 3 to lane 1 between time t-1 and time t.

[0191] Assuming, using Let represent the probability of the target vehicle moving from candidate lane j to candidate lane i from time t-1 to time t. Candidate lane j can be any one of multiple candidate lanes.

[0192] As an example, when candidate lane j and candidate lane i belong to the same lane, it means that the target vehicle has not changed lanes. Therefore, The value can be set to 1.

[0193] When candidate lane j and candidate lane i belong to different lanes, it indicates that the target vehicle has changed lanes. In this case, the distance between candidate lane j and candidate lane i is used to determine the lane.

[0194] In one possible implementation, the distance between candidate lane j and candidate lane i can be represented by the spacing between the center lines of candidate lane j and candidate lane i.

[0195] In one possible implementation, determining the transition probability based on the distance between candidate lane j and candidate lane i includes: determining the transition probability based on the distance between the center lines of candidate lane j and candidate lane i, combined with a preset distance coefficient.

[0196] For example, The result is obtained using formula (4):

[0197]

[0198] In formula (4), μ represents the transmission coefficient, which is usually a set parameter. For example, when i = j, μ can be set to 1, and when i ≠ j, μ can be set to m. t D ij , where D ij Let m be the distance between the center lines of lane i and lane j. t This represents the distance coefficient, for example, m. tThe value range is 1 to 20.

[0199] S703. Based on the launch probability and the transfer probability, obtain the reference probability that the current position belongs to candidate lane i.

[0200] In one possible implementation, after obtaining the reference probability that the current position belongs to candidate lane i based on the emission probability and the transfer probability, the reference probability can be normalized to obtain the normalized probability. Then, using the normalized probability, candidate lanes whose reference probabilities meet the set lane selection conditions are selected from multiple candidate lanes as reference lanes that match the current position.

[0201] In one possible implementation, the reference probability of the current position belonging to candidate lane i can be obtained by combining the launch probability and the transition probability with the reference probability of the current position belonging to multiple candidate lanes at time t-1. For example, based on the transition probability and the reference probability of the current position belonging to multiple candidate lanes at time t-1, the maximum probability is obtained, and based on the maximum probability, combined with the launch probability, the reference probability of the current position belonging to candidate lane i is obtained.

[0202] For example, the normalized probability is calculated using formula (5), and the reference probability is calculated using formula (6):

[0203]

[0204]

[0205] Where i and j represent candidate lanes, Let represent the normalized probability that the target vehicle belongs to candidate lane i at time t. This represents the reference probability that the target vehicle belongs to candidate lane i at time t.

[0206] The reference probability consists of two parts: and in, For the probability of launch, Characterization: The maximum probability of candidate lane i when transitioning from time t-1 to time t. Let be the normalized probability that the current position belongs to candidate lane j. Let be the probability of moving from candidate lane j at time t-1 to candidate lane i at time t.

[0207] In the above implementation, by introducing emission probability and transition probability, since emission probability and transition probability describe the relationship between observed data and states and the transition relationship between states respectively, HMM can model complex dynamic systems, capture dependencies in time series data, and handle the uncertainty and noise of observations, thereby improving the predictive performance of the model.

[0208] Using the above The calculation process can obtain the reference probability of the current position belonging to each candidate lane.

[0209] S202-2. Based on the obtained multiple reference probabilities, select the candidate lanes whose corresponding reference probabilities meet the set lane selection conditions from multiple candidate lanes, and use them as reference lanes to match the current position.

[0210] Setting lane selection criteria refers to the conditions used to filter candidate lanes. For example, setting lane selection criteria could be: selecting the highest probability value among multiple reference probabilities.

[0211] For example, see Figure 8 As shown, the reference probabilities of lanes 1, 2, and 3 are 0.6, 0.7, and 0.9, respectively. Therefore, lane 3 with the highest reference probability is selected from lanes 1, 2, and 3 as the reference lane to match the current position.

[0212] In one possible implementation, determining the adsorption position of the target vehicle on the lane centerline of the reference lane based on the current position includes:

[0213] Obtain the distances between the current position and each center point on the center line of the reference lane;

[0214] From the center points along the lane centerline, select the center points whose distance from the current position meets the preset adsorption conditions;

[0215] The selected center point is used as the attachment point of the target vehicle on the center line of the lane.

[0216] The preset adsorption conditions are used to filter adsorption locations. For example, setting lane selection conditions could be: selecting the center point with the smallest distance from the current position among all center points.

[0217] In one possible implementation, if the center line of the reference lane is a straight line, then on the map, a perpendicular line is drawn from the current position to the center line of the reference lane, and the intersection of the perpendicular line and the center line is taken as the snap-in position.

[0218] Taking candidate lane i as lane 2 as an example, see [link / reference] Figure 6 As shown, the dots represent the current location, the dashed lines represent the lane centerlines, and the distance between the current location and the lane centerline of lane 1 is the distance between candidate lane 1 and the current location. In one possible implementation, the target vehicle's current location is a geographic location. If the geographic coordinate system and the map coordinate system are different, coordinate system transformation can be used to convert the geographic location and map data to the same coordinate system, thus enabling lane matching within the same coordinate system. For example, the current location can be transformed from the geographic coordinate system to the map coordinate system to obtain its map location.

[0219] Accordingly, based on the lane positions of each of the multiple candidate lanes, the distances between each of the multiple candidate lanes and the current position are obtained, including: based on the lane positions of each of the multiple candidate lanes, the distances between each of the multiple candidate lanes and the map position are obtained, and based on the obtained multiple distances, the reference probability of the map position belonging to the multiple candidate lanes is obtained, and then based on the obtained multiple reference probabilities, the candidate lanes whose corresponding reference probabilities meet the set lane selection conditions are selected from the multiple candidate lanes as reference lanes that match the current position.

[0220] Based on the current location, the target vehicle's attachment position on the center line of the reference lane is obtained, including: obtaining the distance between each center point on the center line of the reference lane and the map position; selecting center points from the center points on the center line whose distance from the map position meets the preset attachment conditions; and using the selected center points as the target vehicle's attachment position on the center line of the lane.

[0221] In one possible implementation, where the geographic coordinate system and the map coordinate system are different coordinate systems, if the lane centerline is a straight line, a perpendicular line can be drawn from the map location to the lane centerline of the reference lane, and the intersection of the perpendicular line and the lane centerline can be used as the snap-in location.

[0222] In one possible implementation, if the geographic coordinate system and the map coordinate system are the same coordinate system, the distance between each of the multiple candidate lanes and the current position can be obtained directly using the current position, based on the lane positions of each of the multiple candidate lanes, and the adsorption position of the target vehicle on the lane centerline of the reference lane can be obtained based on the current position.

[0223] S203. Based on the noise information of item M, generate error information. The error information is used to measure the accuracy of the actual offset between the current position and the adsorption position. Based on the error information, predict the compensation offset required to adjust the current position to the adsorption position.

[0224] The compensation offset is the offset after adjusting the actual offset based on the error information.

[0225] For ease of description, R is used. t X represents the error information at time t. t d represents the compensation offset at time t. t This represents the actual offset at time t.

[0226] One possible implementation involves evaluating the error information based on M noise terms, including at least one of the following methods:

[0227] Method 1: If the M noise information items include location-related information, then evaluate the offset noise based on the location-related information, and use the offset noise to obtain R. t .

[0228] Offset is used to characterize the degree to which the positional factors of the target vehicle during its movement affect the accuracy of the actual offset.

[0229] In one possible implementation, location-related information includes: the positioning accuracy of the terminal device, and then, from the lane width of the reference lane and d... t In the process, target information that meets the set selection criteria is selected; based on the target information and positioning accuracy, and combined with the accuracy weight set for the positioning accuracy, offset noise is obtained.

[0230] The selection criteria can be set as follows: from the lane width of the reference lane, d t Among the absolute values, the maximum value is selected as the target information.

[0231] Assuming we use r d The offset noise is represented by the following formula (7):

[0232]

[0233] Where, d l The reference lane width is indicated by h, and the precision radius is indicated by m. h This represents the precision scaling factor, or precision weight, m. h The value of m can be set according to the actual performance of the accuracy estimation model, for example, m h The value of can be between 0 and 1.

[0234] Method 2: If the noise information in item M includes speed-related information, then the speed noise is evaluated based on the speed-related information, and the error information is obtained using the speed noise.

[0235] In one possible implementation, if speed-related information includes the target vehicle's direction of movement and speed, then the speed noise is evaluated based on the target vehicle's direction of movement and speed.

[0236] Speed ​​noise is used to characterize the degree to which the speed factor of the target vehicle during its movement affects the accuracy of the actual offset.

[0237] As an example, the direction difference between the moving direction and the lane direction can be obtained based on the target vehicle's moving direction and the lane direction of the reference lane. Then, based on the direction difference and the moving speed, the speed noise can be evaluated.

[0238] Using r s To represent velocity noise, for example, velocity noise can be calculated using the following formula (8):

[0239]

[0240] Where v is the speed, a is the difference between the direction angle of the matching lane and the heading of the positioning point, and m v This represents the speed scaling factor, which can range from 1 to 10.

[0241] Method 3: If the noise information in item M includes driving behavior information, then the driving behavior noise is evaluated based on the driving behavior information, and the error information is obtained using the driving behavior noise.

[0242] Driving behavior noise is used to characterize the impact of the target vehicle's driving behavior on the accuracy of the actual offset. Driving behavior includes, but is not limited to, lane changes and leaving the road.

[0243] In one possible implementation, if the driving behavior information indicates that the target vehicle is located in a candidate lane other than the reference lane at time t-1, or that the terminal device deviates from the road boundary line, then the driving behavior noise is set to a first preset value; otherwise, the driving behavior noise is set to a second preset value, where the first preset value is higher than the second preset value.

[0244] In one possible implementation, it can be determined whether the target vehicle is located in a candidate lane other than the reference lane at time t-1 based on the lane change information of adjacent moments before and after the driving behavior information.

[0245] In one possible implementation, it can be determined whether the terminal device has deviated from the road boundary line based on the road departure information in the driving behavior information. Deviation of the terminal device from the road boundary line can mean either that the current position exceeds the road boundary line, or that the distance between the current position and the road boundary line exceeds a set distance threshold.

[0246] Assuming we use rc Represents velocity noise, r c This represents driving behavior noise, typically 1, but when a lane change occurs or the distance between the current position and the road boundary line exceeds a set distance threshold, r... c This value can be set to a relatively large value, for example, a value within the range of 2 to 10. In this way, when the terminal device leaves the road or changes lanes, adjusting the value of the driving behavior noise affects the predicted compensation offset, thereby reducing the impact of the terminal device leaving the road or changing lanes on the lateral deviation between the current position and the actual position of the vehicle, thus improving lane matching accuracy. Lateral refers to the direction perpendicular to the lane direction; for example, the direction perpendicular to the lane centerline is lateral.

[0247] In one implementation, if the M noise information items include any number of the following: location-related information, speed-related information, and lane-related information, then the corresponding noise can be calculated based on the M noise information items, and the calculated noise can be combined to obtain error information.

[0248] For example, the M noise information items include: location-related information, speed-related information, and lane-related information. Then, based on the location-related information, offset noise is evaluated; based on the speed-related information, speed noise is evaluated; based on the lane-related information, driving behavior noise is evaluated; finally, error information is obtained using offset noise, speed noise, and driving behavior noise.

[0249] For example, R t The following formula (9) is used to calculate:

[0250] R t =r c r d r s (9)

[0251] One possible implementation is based on R. t Predict X t This includes the following steps:

[0252] S203-1, Based on d t Combined with X t-1 , obtain Y t .

[0253] Among them, X t-1 Y represents the compensation offset at time t-1. t This represents the observation residual at time t, which is used to characterize d. t With X t-1 The differences between them.

[0254] For example, Y t =d t-X t-1 .

[0255] S203-2, According to R t , obtain K t|t-1 K t|t-1 The gain coefficient used to characterize time t, also known as the Kalman gain, can be considered as Y. t The weight.

[0256] In one possible implementation, the historical prediction error at time t-1 and the preset initial prediction error can be combined with R. t The historical prediction error at time t-1 is used to measure the accuracy of the compensation offset predicted at time t-1.

[0257] For example, obtaining K t|t-1 The result is obtained using formula (10):

[0258]

[0259] Among them, Q t-1 Q represents the historical prediction error at time t-1, and Q0 represents the initial prediction error. In this embodiment, Q can also be understood as the variance of X.

[0260] S203-3, according to K t|t-1 Y t X t-1 Predict X t .

[0261] For example, X t The following formula (11) is used to calculate:

[0262] X t =X t-1 +K t|t-1 (d t -X t-1 (11)

[0263] Predicting X using Kalman gain t Kalman gain can be used to balance the weights of the predicted compensation noise and the actual offset, reducing prediction error and improving prediction accuracy, thereby improving the accuracy of compensation and thus improving the accuracy of lane matching.

[0264] In one possible implementation, if the geographic coordinate system and the map coordinate system are different coordinate systems, the error information is used to measure the accuracy of the actual offset between the map position and the snap-in position, and the offset compensation is used to adjust the map position to the snap-in position. Accordingly, the current position used to evaluate the error information is always the map position, and the current position used to predict the offset compensation is also always the map position.

[0265] In one possible implementation, if the geographic coordinate system and the map coordinate system are the same coordinate system, the current location can be directly used to evaluate error information, and the current location can also be directly used to predict compensation offset.

[0266] S204. If the compensation offset meets the set compensation conditions, the current position is corrected based on the compensation offset to obtain the corrected position of the target vehicle. Based on the road network data, the target lane that matches the corrected position is selected from multiple candidate lanes.

[0267] The corrected position of the target vehicle refers to the position after adjusting the current position based on the compensation offset. If the map coordinate system and the geographic coordinate system are different, then adjusting the current position based on the compensation offset means adjusting the current position on the map based on the compensation offset. If the map coordinate system and the geographic coordinate system are the same, then adjusting the current position based on the compensation offset means adjusting the current position based on the compensation offset.

[0268] The target lane is the lane where the target vehicle will be located in the final output. The target lane may be the same as or different from the reference lane.

[0269] In one possible implementation, if the geographic coordinate system and the map coordinate system are different coordinate systems, the current position is corrected based on the compensation offset to obtain the corrected position of the target vehicle. This includes: correcting the map position based on the compensation offset to obtain the corrected position of the target vehicle, where the map position is the location of the current position on the map.

[0270] In one possible implementation, if the geographic coordinate system and the map coordinate system are the same coordinate system, the current position is directly corrected based on the compensation offset to obtain the corrected position of the target vehicle.

[0271] In one possible implementation, after predicting X... t Then, based on K t|t-1 Combined with Q t-1 , obtain Q t And in Q t <m q When, determine X t It meets the established compensation conditions. Among them, Q t Q represents the current prediction error. t Used to measure X t The accuracy of Q t-1 Used to measure X t-1 The accuracy of m qThis represents the set error threshold. For example, m q The value range is: less than 1 square meter (m 2 For example, Q t = (1-K) t|t-1 (Q) t-1 +Q0).

[0272] See Figure 9 As shown, in obtaining R t and d t Subsequently, on the one hand, based on R t Q t-1 Q0, obtain K t|t-1 and based on d t With X t-1 , obtain Y t Then, according to Y t and X t-1 Can predict X t On the other hand, according to K t|t-1 and Q t-1 , obtain Q t Thus, when Q t Converging to less than the threshold m q When X reaches convergence, the algorithm is considered to have converged. t It can compensate to the current position. Compared to compensating at every moment, compensating at convergence can not only reduce the amount of calculation and improve vehicle matching efficiency, but also improve the accuracy of compensation, thereby improving the accuracy of lane matching.

[0273] One possible implementation involves selecting a target lane that matches the corrected position from multiple candidate lanes based on road network data, including:

[0274] Based on the lane positions of multiple candidate lanes, the distances between each candidate lane and the corrected position are obtained, and based on the obtained distances, a reference probability is obtained that the corrected position belongs to multiple candidate lanes.

[0275] Based on the obtained multiple reference probabilities, candidate lanes that meet the set lane selection conditions are selected from multiple candidate lanes and used as target lanes to match the corrected position.

[0276] Since the process of selecting the target lane that matches the corrected position from multiple candidate lanes based on road network data is similar to the process of selecting the target lane that matches the current position from multiple candidate lanes based on road network data, it will not be repeated here. Please refer to the above for details.

[0277] In some implementations, the target vehicle's position on the center line of the target lane can be obtained based on the corrected position, and then the target vehicle can be displayed in the navigation interface of the terminal device according to the position of adsorption.

[0278] In some implementations, if the compensation offset does not meet the set compensation conditions, the reference lane is used as the lane where the target vehicle is located.

[0279] The following section will explain this application using a specific process as an example.

[0280] Suppose a user is inside the target vehicle and is using a tablet for navigation, during which lane matching is required.

[0281] See Figure 10 As shown, at the first moment, the tablet computer collects GNSS signals and IMU data and reports the collected data to the server. Based on the GNSS signals and IMU data at the first moment, the server infers the target vehicle's location information at that moment, denoted as Location Information 1. Location Information 1 includes: current position 1, and noise information such as movement speed, movement direction, and positioning accuracy. The server performs the first lane matching based on the map and current position 1 to obtain a reference lane, and determines the target vehicle's adsorption position 1 on the center line of the reference lane based on current position 1. Furthermore, based on the noise information at the first moment, it evaluates R1, and based on R1, combined with current position 1 and adsorption position 1, predicts X1. Further, the server bases K... 1|0 Combining Q0, we obtain Q1. Assume Q1 > m. q Therefore, if X1 does not meet the set compensation conditions, the lane where the target vehicle is located is designated as the reference lane. Furthermore, the server can notify the tablet computer that the target vehicle is in the reference lane, allowing the tablet computer to display the target vehicle's location on the navigation interface.

[0282] At the second moment, the tablet computer collects GNSS signals and IMU data, and reports the collected data to the server. Based on the GNSS signals and IMU data at the second moment, the server infers the target vehicle's location information at that moment, denoted as Location Information 2. Location Information 2 includes: current position 2, and noise information such as movement speed, movement direction, and positioning accuracy. The server performs the first lane matching based on the map and current position 2 to obtain a reference lane, and based on current position 2, determines the target vehicle's attachment position 2 on the center line of the reference lane; and, based on the noise information at the second moment, evaluates R2, and based on R2, combined with current position 2 and attachment position 2, predicts X2. Further, the server, based on K... 2|1 Combining Q1, we obtain Q2. Assume Q2... <mq Therefore, since X2 meets the set compensation conditions, the current position 2 is corrected based on X2 to obtain the corrected position at the second time step. Then, the server performs a second lane matching based on the road network data and the corrected position to obtain the target lane. Furthermore, the server can notify the tablet computer that the target vehicle is in the target lane, allowing the tablet computer to display the target vehicle's location on the navigation interface.

[0283] Similarly, for each moment, the server can use the lane matching method provided in this application embodiment to determine the lane where the target vehicle is located. By dynamically compensating the current position of the target vehicle, the lateral deviation caused by factors such as the placement position of the terminal device, the positioning accuracy of the GNSS signal, and the quality of the positioning information is corrected, thereby effectively correcting the current position to near the center position of the target vehicle, and comprehensively improving the accuracy and stability of lane matching.

[0284] In some implementations, X t The prediction process can be understood as a state estimation process based on KF.

[0285] I. Parameter Description

[0286] The state estimation process based on KF mainly involves state-related parameters and observation-related parameters.

[0287] (1) State-related parameters

[0288] The state variable, i.e. the predicted compensation offset, is denoted by X;

[0289] The variance of the state variable, i.e. the variance of X, is denoted by Q. In the embodiments of this application, the variance of the state variable can also be called the prediction error.

[0290] The state transition matrix, denoted by F, is given by the state equation: X t|t-1 =F t X t-1 In this embodiment, since the state variables are one-dimensional data, F is the identity matrix I, which can be simplified to F = 1. That is, the state equation can be expressed as: X t|t-1 =X t-1 ;

[0291] Q0 represents the initial prediction error.

[0292] (2) Observation of relevant parameters

[0293] The observed quantity, i.e., the actual offset between the current position and the adsorption position, is denoted by d;

[0294] The variance of the observations (also known as error information), i.e. the variance of d, is denoted by R;

[0295] The observation matrix is ​​represented by H, and the observation equation is: d = HX. In this embodiment, since the state variables are one-dimensional data, H is the identity matrix I, which can be simplified to H = 1.

[0296] The observation residuals are represented by Y.

[0297] Kalman gain, denoted by K, Figure 11 K in t Equivalent to K in the above text t|t-1 .

[0298] II. KF-based state estimation process

[0299] The state estimation process based on KF can include two parts: state transition and measurement update.

[0300] See Figure 11 As shown, X is calculated using formula (12) during the state transition. t|t-1 Q is calculated using formula (13). t|t-1 Formulas (12) and (13) are shown below:

[0301] X t|t-1 =F t X t-1 (12)

[0302]

[0303] In the measurement update, the calculated Q is used t|t-1 and X t|t-1 Predict X t and Q t Specifically, according to d t and X t|t-1 Y is obtained using formula (14) t According to R t and Q t|t-1 K is obtained using formula (15). t Then, X is calculated using formulas (16) and (17) respectively. t and Q t Formulas (14) to (17) are shown below:

[0304] Y t =d t -H t X t|t-1 (14)

[0305]

[0306] X t =X t|t-1 +Kt Y t (16)

[0307] Q t =(IK t H t )Q t|t-1 (17)

[0308] When the state variables are one-dimensional data, both F and H can be simplified to 1, and the above formulas (12) to (17) can be transformed into the above formulas (7) to (11). In practical applications, if the state variables are multi-dimensional data, then the above formulas (12) to (13) can also be used directly for calculation.

[0309] Based on the same inventive concept, embodiments of this application provide a lane matching device. For example... Figure 12 As shown, this is a structural schematic diagram of the lane matching device 1200, which may include:

[0310] The data acquisition unit 1201 is used to acquire real-time positioning information of the target vehicle generated by the terminal device. The real-time positioning information includes: the current position of the target vehicle and at least one noise information of the current position; and to acquire road network data of the map.

[0311] The first matching unit 1202 is used to filter out a reference lane that matches the current position from multiple candidate lanes based on the road network data, and to determine the adsorption position of the target vehicle on the lane centerline of the reference lane based on the current position.

[0312] The position compensation unit 1203 is used to generate error information based on the at least one noise information, the error information being used to measure the accuracy of the actual offset between the current position and the adsorption position, and based on the error information, to predict the compensation offset required to adjust the current position to the adsorption position;

[0313] The second matching unit 1204 is used to correct the current position based on the compensation offset if the compensation offset meets the set compensation conditions, to obtain the corrected position of the target vehicle, and to filter out the target lane that matches the corrected position from multiple candidate lanes based on the road network data.

[0314] In one possible implementation, when predicting the compensation offset required to adjust the current position to the adsorption position based on the error information, the position compensation unit 1203 is specifically used for:

[0315] Based on the actual offset, combined with the historical compensation offset predicted at the previous moment, the observation residual is obtained. The observation residual is used to characterize the difference between the actual offset and the compensation offset predicted at the previous moment.

[0316] Based on the error information, the Kalman gain is obtained;

[0317] Based on the Kalman gain, the observation residual, and the historical compensation offset, the compensation offset required to adjust the current position to the adsorption position is predicted.

[0318] In one possible implementation, the position compensation unit 1203 is further used for:

[0319] Based on the Kalman gain and combined with the historical prediction error from the previous moment, the current prediction error is obtained; wherein, the historical prediction error is used to measure the accuracy of the historical compensation offset, and the current prediction error is used to measure the accuracy of the compensation offset.

[0320] If the current prediction error is less than the set error threshold, then the compensation offset is determined to meet the set compensation conditions.

[0321] In one possible implementation, when generating error information based on the at least one piece of noise information, the position compensation unit 1203 is specifically configured to perform at least one of the following methods:

[0322] If the at least one piece of noise information includes location-related information, then based on the location-related information, the offset noise is evaluated, and using the offset noise, error information is generated;

[0323] If the at least one piece of noise information includes speed-related information, then based on the speed-related information, speed noise is evaluated, and error information is generated using the speed noise;

[0324] If the at least one noise information includes driving behavior information, then driving behavior noise is evaluated based on the driving behavior information, and error information is generated using the driving behavior noise.

[0325] In one possible implementation, the location-related information includes: the positioning accuracy of the terminal device; then the location compensation unit 1203 is further used for:

[0326] The lane width of the reference lane is obtained from the road network data;

[0327] Select target information that meets the set selection criteria from the lane width of the reference lane and the actual offset.

[0328] Based on the target information and the positioning accuracy, and combined with the accuracy weight set for the positioning accuracy, the offset noise is obtained.

[0329] In one possible implementation, speed-related information includes: direction of movement and speed of movement.

[0330] In one possible implementation, the position compensation unit 1203 is further used for:

[0331] If the driving behavior information indicates that the target vehicle was in a candidate lane other than the reference lane at the previous moment, or that the target vehicle deviated from the road boundary line, then the driving behavior noise is set to a first preset value; otherwise, the driving behavior noise is set to a second preset value, wherein the first preset value is higher than the second preset value.

[0332] In one possible implementation, the road network data includes: the lane positions of each of the multiple candidate lanes; when filtering a reference lane matching the current position from the multiple candidate lanes based on the road network data, the first matching unit 1202 is specifically used for:

[0333] Based on the lane positions of each of the multiple candidate lanes, the distance between each of the multiple candidate lanes and the current position is obtained, and based on the obtained multiple distances, a reference probability is obtained that the current position belongs to the multiple candidate lanes;

[0334] Based on the obtained multiple reference probabilities, candidate lanes that meet the set lane selection conditions are selected from the multiple candidate lanes and used as reference lanes that match the current position.

[0335] In one possible implementation, when obtaining the reference probability that the current position belongs to one of the multiple candidate lanes based on the obtained multiple distances, the first matching unit 1202 is specifically used for:

[0336] For each of the multiple candidate lanes, perform the following operations:

[0337] Based on the distance between a candidate lane and the current position, the distance factor between the target vehicle and the candidate lane is obtained;

[0338] Based on the distance factor, a reference probability is obtained that the current position belongs to one of the candidate lanes.

[0339] In one possible implementation, when obtaining the reference probability that the current position belongs to the candidate lane based on the distance factor, the first matching unit 1202 is specifically used for:

[0340] Based on the driving direction of the target vehicle and the lane direction of the candidate lane, the orientation factor between the target vehicle and the candidate lane is obtained.

[0341] Based on the distance factor and the orientation factor, and combined with a preset fitting function, a reference probability is obtained that the current position belongs to one of the candidate lanes.

[0342] For ease of description, the above sections are divided into modules (or units) according to their functions and described separately. Of course, in implementing this application, the functions of each module (or unit) can be implemented in one or more software or hardware components.

[0343] Regarding the apparatus in the above embodiments, the specific manner in which each unit executes the request has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0344] Those skilled in the art will understand that various aspects of this application can be implemented as a system, method, or program product. Therefore, various aspects of this application can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, collectively referred to herein as a "circuit," "module," or "system."

[0345] Based on the same inventive concept, embodiments of this application also provide an electronic device. In one embodiment, the electronic device can be a server or a terminal device. See also... Figure 13 As shown, it is a schematic diagram of the structure of a possible electronic device provided in an embodiment of this application. Figure 13 In the electronic device 1300, there are: processor 1310 and memory 1320.

[0346] The memory 1320 stores a computer program that can be executed by the processor 1310. The processor 1310 can execute the steps of the lane matching method described above by executing the instructions stored in the memory 1320.

[0347] Memory 1320 may be volatile memory, such as random-access memory (RAM); memory 1320 may also be non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); or memory 1320 may be any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 1320 may also be a combination of the above-described memories.

[0348] The processor 1310 may include one or more central processing units (CPUs) or digital processing units, etc. The processor 1310 implements the lane matching method described above when executing a computer program stored in the memory 1320.

[0349] In some embodiments, the processor 1310 and the memory 1320 may be implemented on the same chip, while in other embodiments they may be implemented on separate chips.

[0350] This application embodiment does not limit the specific connection medium between the processor 1310 and the memory 1320. This application embodiment takes the connection between the processor 1310 and the memory 1320 via a bus as an example. Figure 13 The diagram uses thick lines to describe the connections between other components; these are merely illustrative and not intended to be limiting. Buses can be categorized as address buses, data buses, control buses, etc. For ease of description, Figure 13 It is described using only a thick line, but does not indicate that there is only one bus or one type of bus.

[0351] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium including a computer program. When the computer program is run on an electronic device, it causes the electronic device to perform the steps of the lane matching method described above. In some possible implementations, various aspects of the lane matching method provided in this application can also be implemented as a program product including a computer program. When the program product is run on an electronic device, the computer program causes the electronic device to perform the steps in the lane matching method described above. For example, the electronic device can perform actions such as... Figure 2 The steps are shown in the figure.

[0352] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, RAM, ROM, erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0353] The program product of the embodiments of this application may be a CD-ROM and include a computer program, and may run on an electronic device. However, the program product of this application is not limited thereto. In this document, the readable storage medium may be any tangible medium that contains or stores a computer program that may be used by or in conjunction with a command execution system, apparatus, or device.

[0354] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a readable computer program. This propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting a computer program for use by or in conjunction with a command execution system, apparatus, or device.

[0355] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0356] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A lane matching method, characterized in that, include: The real-time positioning information of the target vehicle generated by the terminal device is obtained, and the real-time positioning information includes: the current position of the target vehicle and at least one noise information of the current position; Obtain road network data from the map, and based on the road network data, filter out a reference lane that matches the current location from multiple candidate lanes, and based on the current location, determine the adsorption position of the target vehicle on the center line of the reference lane. Based on the at least one piece of noise information, error information is generated, which is used to measure the accuracy of the actual offset between the current position and the adsorption position, and based on the error information, the compensation offset required to adjust the current position to the adsorption position is predicted. If the compensation offset meets the set compensation conditions, the current position is corrected based on the compensation offset to obtain the corrected position of the target vehicle, and the target lane matching the corrected position is selected from multiple candidate lanes based on the road network data.

2. The method as described in claim 1, characterized in that, The step of predicting the compensation offset required to adjust the current position to the adsorption position based on the error information includes: Based on the actual offset, combined with the historical compensation offset predicted at the previous moment, the observation residual is obtained. The observation residual is used to characterize the difference between the actual offset and the compensation offset predicted at the previous moment. Based on the error information, the Kalman gain is obtained; Based on the Kalman gain, the observation residual, and the historical compensation offset, the compensation offset required to adjust the current position to the adsorption position is predicted.

3. The method as described in claim 2, characterized in that, Determining that the compensation offset meets the set compensation conditions includes: Based on the Kalman gain and combined with the historical prediction error from the previous moment, the current prediction error is obtained; wherein, the historical prediction error is used to measure the accuracy of the historical compensation offset, and the current prediction error is used to measure the accuracy of the compensation offset; If the current prediction error is less than the set error threshold, then the compensation offset is determined to meet the set compensation conditions.

4. The method according to any one of claims 1-3, characterized in that, The step of generating error information based on the at least one piece of noise information includes at least one of the following methods: If the at least one piece of noise information includes location-related information, then based on the location-related information, the offset noise is evaluated, and using the offset noise, error information is generated; If the at least one piece of noise information includes speed-related information, then based on the speed-related information, speed noise is evaluated, and error information is generated using the speed noise; If the at least one noise information includes driving behavior information, then driving behavior noise is evaluated based on the driving behavior information, and error information is generated using the driving behavior noise.

5. The method as described in claim 4, characterized in that, Location-related information includes: the positioning accuracy of the terminal device; then, evaluating the offset noise based on the location-related information includes: The lane width of the reference lane is obtained from the road network data; Select target information that meets the set selection criteria from the lane width of the reference lane and the actual offset. Based on the target information and the positioning accuracy, and combined with the accuracy weight set for the positioning accuracy, the offset noise is obtained.

6. The method as described in claim 4, characterized in that, Speed-related information includes: direction of movement and speed of movement.

7. The method as described in claim 4, characterized in that, The assessment of driving behavior noise based on the driving behavior information includes: If the driving behavior information indicates that the target vehicle was in a candidate lane other than the reference lane at the previous moment, or that the target vehicle deviated from the road boundary line, then the driving behavior noise is set to a first preset value; otherwise, the driving behavior noise is set to a second preset value, wherein the first preset value is higher than the second preset value.

8. The method according to any one of claims 1-3, characterized in that, The road network data includes: the lane positions of each of the multiple candidate lanes; the step of filtering a reference lane that matches the current position from the multiple candidate lanes based on the road network data includes: Based on the lane positions of each of the multiple candidate lanes, the distance between each of the multiple candidate lanes and the current position is obtained, and based on the obtained multiple distances, a reference probability is obtained that the current position belongs to the multiple candidate lanes; Based on the obtained multiple reference probabilities, candidate lanes that meet the set lane selection conditions are selected from the multiple candidate lanes and used as reference lanes that match the current position.

9. The method as described in claim 7, characterized in that, The step of obtaining a reference probability that the current position belongs to one of the multiple candidate lanes based on the obtained multiple distances includes: For each of the multiple candidate lanes, perform the following operations: Based on the distance between a candidate lane and the current position, the distance factor between the target vehicle and the candidate lane is obtained; Based on the distance factor, a reference probability is obtained that the current position belongs to one of the candidate lanes.

10. The method as described in claim 9, characterized in that, The step of obtaining the reference probability that the current position belongs to one of the candidate lanes based on the distance factor includes: Based on the driving direction of the target vehicle and the lane direction of the candidate lane, the orientation factor between the target vehicle and the candidate lane is obtained. Based on the distance factor and the orientation factor, and combined with a preset fitting function, a reference probability is obtained that the current position belongs to one of the candidate lanes.

11. A lane matching device, characterized in that, include: The data acquisition unit is used to acquire real-time positioning information of the target vehicle generated by the terminal device, the real-time positioning information including: the current position of the target vehicle and at least one noise information of the current position; and to acquire road network data of the map; The first matching unit is used to filter out a reference lane that matches the current position from multiple candidate lanes based on the road network data, and to determine the adsorption position of the target vehicle on the lane centerline of the reference lane based on the current position. A position compensation unit is configured to generate error information based on the at least one noise information, the error information being used to measure the accuracy of the actual offset between the current position and the adsorption position, and based on the error information, to predict the compensation offset required to adjust the current position to the adsorption position; The second matching unit is used to correct the current position based on the compensation offset if the compensation offset meets the set compensation conditions, to obtain the corrected position of the target vehicle, and to select the target lane from multiple candidate lanes based on the road network data and match the corrected position with the multiple candidate lanes to obtain the lane matching result.

12. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of any of the methods described in claims 1 to 10.

13. A computer-readable storage medium, characterized in that, It includes a computer program that, when run on an electronic device, causes the electronic device to perform the steps of any of the methods described in claims 1 to 10.

14. A computer program product, characterized in that, It includes a computer program stored in a computer-readable storage medium, and a processor of an electronic device reads from and executes the computer program, causing the electronic device to perform the steps of any of the methods described in claims 1 to 10.

Images