Lane level positioning method, device, equipment, vehicle and medium

Abstract

The invention provides a lane level positioning method, device and equipment, a vehicle and a medium, and relates to the technical field of computers, in particular to the technical fields of artificial intelligence, computer vision, intelligent vehicles, automatic driving vehicles and the like. The method comprises the following steps: acquiring map road information and visual perception road information around a vehicle based on positioning information of the vehicle; determining a plurality of candidate lanes based on the positioning information; determining the topology recursion probability of the vehicle in each candidate lane by using map road information; utilizing visual perception road information to determine the perception observation probability of the vehicle in each candidate lane; determining the positioning probability of the vehicle in each candidate lane by using the positioning information; and determining a target lane where the vehicle is located in the plurality of candidate lanes based on the topology recursion probability, the perception observation probability and the positioning probability that the vehicle is located in each candidate lane.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, specifically to the technical fields of artificial intelligence, computer vision, intelligent vehicles and autonomous driving vehicles, and in particular to a lane-level positioning method, a lane-level positioning device, an electronic device, a vehicle, a computer-readable storage medium and a computer program product. Background Art

[0002] Due to its broad application prospects, assisted driving technology has received extensive attention in the industry, especially the automotive industry, in recent years, and has been deeply studied and explored. Assisted driving technology is the concentrated embodiment of current automation and intelligent technology, which is obtained by integrating a large number of advanced technologies. Lane-level positioning technology is an indispensable component of it, playing a pivotal role and directly affecting the reliability and safety of subsequent decision-making and control.

[0003] The methods described in this section are not necessarily methods that have been previously conceived or employed. Unless otherwise indicated, it should not be assumed that any method described in this section is considered to be prior art simply because it is included in this section. Similarly, unless otherwise indicated, the issues mentioned in this section should not be considered to have been recognized in any prior art. Summary of the invention

[0004] The present disclosure provides a lane-level positioning method, a lane-level positioning device, an electronic device, a vehicle, a computer-readable storage medium, and a computer program product.

[0005] According to one aspect of the present disclosure, a lane-level positioning method is provided, including: based on the positioning information of the vehicle, obtaining map road information and visually perceived road information around the vehicle; determining multiple candidate lanes based on the positioning information; using the map road information, determining the topological recursive probability that the vehicle is located in each candidate lane; using the visually perceived road information, determining the perceptual observation probability that the vehicle is located in each candidate lane; using the positioning information, determining the positioning probability that the vehicle is located in each candidate lane; and determining the target lane where the vehicle is located among the multiple candidate lanes based on the topological recursive probability, perceptual observation probability and positioning probability that the vehicle is located in each candidate lane.

[0006] According to another aspect of the present disclosure, a lane-level positioning device is provided, including: an acquisition unit, configured to acquire map road information and visual perception road information around a vehicle based on the positioning information of the vehicle; a candidate lane determination unit, configured to determine multiple candidate lanes based on the positioning information; a topological recursive probability determination unit, configured to determine the topological recursive probability that the vehicle is located in each candidate lane using the map road information; a perceptual observation probability determination unit, configured to determine the perceptual observation probability that the vehicle is located in each candidate lane using the visual perception road information; a positioning probability determination unit, configured to determine the positioning probability that the vehicle is located in each candidate lane using the positioning information; and a first target lane determination unit, configured to determine the target lane where the vehicle is located among multiple candidate lanes based on the topological recursive probability, perceptual observation probability and positioning probability that the vehicle is located in each candidate lane.

[0007] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0008] According to another aspect of the present disclosure, a vehicle is provided, comprising the above electronic device.

[0009] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the above method.

[0010] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program implements the above method when executed by a processor.

[0011] According to one or more embodiments of the present disclosure, the present disclosure obtains map road information and visually perceived road information around the vehicle based on the vehicle's positioning information, and determines multiple candidate lanes, and then uses the map road information, visually perceived road information and positioning information to give the probability of the vehicle being located in each candidate lane from three dimensions, and finally obtains the target lane where the vehicle is located based on these probabilities, thereby achieving more accurate lane-level positioning.

[0012] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings exemplarily illustrate the embodiments and constitute a part of the specification, and together with the text description of the specification, are used to explain the exemplary implementation of the embodiments. The embodiments shown are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0014] Figure 1 A schematic diagram showing an exemplary system in which the various methods described herein may be implemented according to an embodiment of the present disclosure;

[0015] Figure 2 A flow chart of a lane-level positioning method according to an exemplary embodiment of the present disclosure is shown;

[0016] Figure 3 A flowchart of determining the topological recursive probability of a vehicle being located in each candidate lane by using map road information according to an exemplary embodiment of the present disclosure is shown;

[0017] Figure 4 A schematic diagram showing a filtering process according to an exemplary embodiment of the present disclosure is shown;

[0018] Figure 5 A flowchart of determining the perceived observation probability of a vehicle being located in each candidate lane by using visually perceived road information according to an exemplary embodiment of the present disclosure is shown;

[0019] Figure 6 A flowchart of determining the positioning probability of a vehicle being located in each candidate lane by using positioning information according to an exemplary embodiment of the present disclosure is shown;

[0020] Figure 7 A flowchart of determining a target lane in which a vehicle is located among a plurality of candidate lanes according to an exemplary embodiment of the present disclosure is shown;

[0021] Figure 8 A flowchart of determining the topological recursive weight, the perceptual observation weight and the positioning weight of each of a plurality of candidate roads according to an exemplary embodiment of the present disclosure is shown;

[0022] Fig. 9 A schematic diagram of a hidden Markov model according to an exemplary embodiment of the present disclosure is shown;

[0023] Fig.10 A flow chart of a lane-level positioning method according to an exemplary embodiment of the present disclosure is shown;

[0024] Fig.11 An overall technical flow chart according to an exemplary embodiment of the present disclosure is shown;

[0025] Fig.12A schematic diagram showing lane-level positioning according to an exemplary embodiment of the present disclosure is shown;

[0026] Fig.13 A structural block diagram of a lane-level positioning device according to an exemplary embodiment of the present disclosure is shown; and

[0027] Fig.14 A structural block diagram of an exemplary electronic device that can be used to implement the embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0028] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0029] In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, timing relationship, or importance relationship of these elements, and such terms are only used to distinguish one element from another element. In some examples, the first element and the second element may refer to the same instance of the element, and in some cases, based on the description of the context, they may also refer to different instances.

[0030] The terms used in the description of various examples in this disclosure are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element can be one or more. In addition, the term "and / or" used in this disclosure covers any one of the listed items and all possible combinations.

[0031] In the related technology, the accuracy of existing lane-level positioning is low.

[0032] To solve the above problems, the present invention obtains map road information and visually perceived road information around the vehicle based on the vehicle's positioning information, and determines multiple candidate lanes, and then uses the map road information, visually perceived road information and positioning information from three dimensions to give the probability of the vehicle being located in each candidate lane, and finally obtains the target lane where the vehicle is located based on these probabilities, thereby achieving more accurate lane-level positioning.

[0033] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0034] Figure 1FIG. 1 is a schematic diagram of an exemplary system 100 in which various methods and apparatuses described herein may be implemented according to an embodiment of the present disclosure. Figure 1 , the system 100 includes a motor vehicle 110 , a server 120 , and one or more communication networks 130 coupling the motor vehicle 110 to the server 120 .

[0035] In an embodiment of the present disclosure, the motor vehicle 110 may include a computing device according to an embodiment of the present disclosure and / or be configured to perform a method according to an embodiment of the present disclosure.

[0036] The server 120 may run one or more services or software applications of the lane-level positioning method. In some embodiments, the server 120 may also provide other services or software applications, which may include non-virtual environments and virtual environments. Figure 1 In the configuration shown, the server 120 may include one or more components that implement the functions performed by the server 120. These components may include software components, hardware components, or a combination thereof that can be executed by one or more processors. A user of the motor vehicle 110 may in turn utilize one or more client applications to interact with the server 120 to utilize the services provided by these components. It should be understood that a variety of different system configurations are possible, which may differ from the system 100. Therefore, Figure 1 is one example of a system for implementing the various methods described herein and is not intended to be limiting.

[0037] Server 120 may include one or more general purpose computers, dedicated server computers (e.g., PC (personal computer) servers, UNIX servers, mid-range servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and / or combination. Server 120 may include one or more virtual machines running virtual operating systems, or other computing architectures involving virtualization (e.g., one or more flexible pools of logical storage devices that may be virtualized to maintain a server's virtual storage device). In various embodiments, server 120 may run one or more services or software applications that provide the functionality described below.

[0038] The computing units in the server 120 may run one or more operating systems including any of the above operating systems and any commercially available server operating systems. The server 120 may also run any of a variety of additional server applications and / or middle-tier applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, etc.

[0039] In some embodiments, server 120 may include one or more applications to analyze and consolidate data feeds and / or event updates received from motor vehicle 110. Server 120 may also include one or more applications to display data feeds and / or real-time events via one or more display devices of motor vehicle 110.

[0040] The network 130 may be any type of network known to those skilled in the art that may support data communications using any of a variety of available protocols, including, but not limited to, TCP / IP, SNA, IPX, etc. By way of example only, one or more networks 130 may be a satellite communications network, a local area network (LAN), an Ethernet-based network, a token ring, a wide area network (WAN), the Internet, a virtual network, a virtual private network (VPN), an intranet, an extranet, a blockchain network, a public switched telephone network (PSTN), an infrared network, a wireless network (including, for example, Bluetooth, WiFi), and / or any combination of these and other networks.

[0041] The system 100 may also include one or more databases 150. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 150 may be used to store information such as audio files and video files. The data repository 150 may reside in various locations. For example, the data repository used by the server 120 may be local to the server 120, or may be remote from the server 120 and may communicate with the server 120 via a network-based or dedicated connection. The data repository 150 may be of different types. In some embodiments, the data repository used by the server 120 may be a database, such as a relational database. One or more of these databases may store, update, and retrieve data to and from the database in response to a command.

[0042] In some embodiments, one or more of the databases 150 may also be used by an application to store application data. The databases used by the application may be different types of databases, such as a key-value store, an object store, or a conventional store backed by a file system.

[0043] The motor vehicle 110 may include a sensor 111 for sensing the surrounding environment. The sensor 111 may include one or more of the following sensors: a visual camera, an infrared camera, an ultrasonic sensor, a millimeter wave radar, and a laser radar (LiDAR). Different sensors may provide different detection accuracy and range. The camera may be installed at the front, rear, or other location of the vehicle. The visual camera may capture the situation inside and outside the vehicle in real time and present it to the driver and / or passengers. In addition, by analyzing the images captured by the visual camera, information such as traffic light indications, intersection conditions, and the operating status of other vehicles may be obtained. The infrared camera may capture objects under night vision conditions. Ultrasonic sensors may be installed around the vehicle to measure the distance of objects outside the vehicle from the vehicle using the characteristics of strong ultrasonic directionality. Millimeter wave radars may be installed at the front, rear, or other location of the vehicle to measure the distance of objects outside the vehicle from the vehicle using the characteristics of electromagnetic waves. Laser radars may be installed at the front, rear, or other location of the vehicle to detect object edges and shape information, thereby identifying and tracking objects. Due to the Doppler effect, the radar device may also measure the speed change of the vehicle and the moving object.

[0044] The motor vehicle 110 may also include a communication device 112. The communication device 112 may include a satellite positioning module capable of receiving satellite positioning signals (e.g., Beidou, GPS, GLONASS, and GALILEO) from satellites 141 and generating coordinates based on these signals. The communication device 112 may also include a module for communicating with a mobile communication base station 142, and the mobile communication network may implement any suitable communication technology, such as GSM / GPRS, CDMA, LTE, and other current or evolving wireless communication technologies (e.g., 5G technology). The communication device 112 may also have a vehicle-to-vehicle or vehicle-to-everything (V2X) module, which is configured to implement, for example, vehicle-to-vehicle (V2V) communication with other vehicles 143 and vehicle-to-infrastructure (V2I) communication with infrastructure 144. In addition, the communication device 112 may also have a module configured to communicate with a user terminal 145 (including but not limited to a smartphone, a tablet computer, or a wearable device such as a watch) by, for example, using a wireless LAN or Bluetooth of the IEEE802.11 standard. Using the communication device 112, the motor vehicle 110 may also access the server 120 via the network 130.

[0045] The motor vehicle 110 may also include a control device 113. The control device 113 may include a processor that communicates with various types of computer-readable storage devices or media, such as a central processing unit (CPU) or a graphics processing unit (GPU), or other dedicated processors. The control device 113 may include an automatic driving system for automatically controlling various actuators in the vehicle. The automatic driving system is configured to control the powertrain, steering system, and braking system of the motor vehicle 110 (not shown) via multiple actuators in response to inputs from multiple sensors 111 or other input devices to control acceleration, steering, and braking, respectively, without human intervention or limited human intervention. Some processing functions of the control device 113 can be implemented through cloud computing. For example, some processing can be performed using an on-board processor, while other processing can be performed using computing resources in the cloud. The control device 113 can be configured to perform the lane-level positioning method according to the present disclosure. In addition, the control device 113 can be implemented as an example of a computing device on the motor vehicle side (client) according to the present disclosure.

[0046] Figure 1 The system 100 may be configured and operated in various ways to enable the application of various methods and apparatuses described in the present disclosure.

[0047] According to one aspect of the present disclosure, a lane-level positioning method is provided. Figure 2 As shown, the lane-level positioning method includes: step S201, based on the positioning information of the vehicle, obtaining the map road information and visual perception road information around the vehicle; step S202, determining multiple candidate lanes based on the positioning information; step S203, using the map road information, determining the topological recursive probability that the vehicle is located in each candidate lane; step S204, using the visual perception road information, determining the perceptual observation probability that the vehicle is located in each candidate lane; step S205, using the positioning information, determining the positioning probability that the vehicle is located in each candidate lane; and step S206, based on the topological recursive probability, perceptual observation probability and positioning probability that the vehicle is located in each candidate lane, determining the target lane where the vehicle is located among the multiple candidate lanes.

[0048] Therefore, the present invention obtains map road information and visually perceived road information around the vehicle based on the vehicle's positioning information, and determines multiple candidate lanes, and then uses the map road information, visually perceived road information and positioning information to give the probability of the vehicle being located in each candidate lane from three dimensions, and finally obtains the target lane where the vehicle is located based on these probabilities, thereby achieving more accurate lane-level positioning.

[0049] The method proposed in the present disclosure can be used for accurate and comprehensive lane-level positioning under conditions of poor map quality.

[0050] Before step S201, the positioning information of the vehicle may be obtained.

[0051] The present disclosure involves two levels of positioning, namely road-level positioning and lane-level positioning. Road-level positioning obtains the approximate position of the vehicle from the positioning result of the vehicle itself, and the positioning result may come from the positioning system or the position after the navigation positioning system and the inertial measurement unit are integrated. The positioning information used in step S201, step S202, and step S205 may all refer to the road-level positioning result. By implementing the method provided by the present disclosure, a lane-level positioning result may be obtained, that is, the target lane where the vehicle is located.

[0052] In an exemplary embodiment, an inertial measurement unit, a navigation positioning system, a wheel speed sensor, and map matching information can be used to perform fusion positioning through a Kalman filter to obtain the vehicle's position information at all times. In step S201, the position can be used to obtain high-precision map data in a certain area nearby. The acquired high-precision map data is mainly given in the form of lanes, so that preliminary road-level positioning information can be obtained. The result of road-level positioning is given as a link identifier (ID).

[0053] In step S202, multiple candidate lanes may be determined in the high-precision map based on the positioning information.

[0054] In some embodiments, the map road information may include road topology information, such as lane edges, center lines, road edges of multiple map lanes, and the successive relationship between lanes / roads, etc., and the visually perceived road information may include lane edges, road edges of multiple perceived lanes, and the successive relationship between perceived lanes / roads, etc. It should be noted that the candidate lane determined in step S202 is a lane in the high-precision map.

[0055] After obtaining multiple candidate lanes, in steps S203 to S205, the probability of the vehicle being located in each candidate lane is predicted from three levels: topological recursion, perception observation, and positioning. Then, in step S206, the target lane where the vehicle is most likely to be located can be determined based on these probabilities, thereby completing lane-level positioning.

[0056] According to some embodiments, Figure 3 As shown, step S203, using map road information to determine the topological recursive probability that the vehicle is located in each candidate lane may include: step S301, using map road information to determine the topological recursive relationship between multiple historical lanes and multiple candidate lanes; and step S304, based on the topological recursive relationship and the topological recursive probability that the vehicle was located in each historical lane at the previous moment, determining the topological recursive probability that the vehicle is located in each candidate lane.

[0057] Therefore, through the above method, the map road information can be effectively used to predict the topological recursive probability of the vehicle being located in each candidate lane from the topological recursive layer of the lane, thereby improving the accuracy of lane-level positioning.

[0058] In some embodiments, the topological recursive relationship may refer to a successor relationship between multiple historical lanes and multiple candidate lanes. In step S302, for each candidate lane, the historical lane corresponding to the candidate lane may be determined based on the topological recursive relationship, and then the topological recursive probability of the historical lane may be determined as the topological recursive probability that the vehicle is located in the candidate lane at the current moment.

[0059] According to some embodiments, Figure 3 As shown, step S203, using the map road information, determining the topological recursive probability that the vehicle is located in each candidate lane may also include: step S302, calculating the lane change probability of the vehicle based on visually perceived road information; and step S303, maintaining the state transition matrix of the topological recursive observation based on the lane change probability. The topological recursive probability that the vehicle is located in each candidate lane may be determined based on the state transition matrix, the topological recursive relationship, and the topological recursive probability that the vehicle was located in each historical lane at the last moment.

[0060] Therefore, through the above method, the possible lane change situation of the vehicle can be further introduced on the basis of the topological recursive relationship, thereby further improving the accuracy of lane-level positioning.

[0061] In some embodiments, the lane change probability may be calculated using the following formula:

[0062]

[0063] Among them, p k-1 represents the topological recursive probability that the vehicle was located in each historical lane at the previous moment, p k It represents the topological recursive probability that the vehicle is located on each candidate road at the current moment, T represents the state transition matrix, and lane_size represents the number of lanes.

[0064] In some embodiments, in step S303, the lane change probability of the vehicle can be calculated based on the lateral deviation of the host vehicle from the lane edge. Further, the lateral deviation can be used as an estimate of the Kalman filter for prediction and update, and the lane change to the left or right is determined according to the positive or negative value of the lateral deviation, and the corresponding lane change probability is set.

[0065] In some embodiments, Kalman filtering is applied to random stationary processes. The system's measurement noise and process noise sequences are not signals to be filtered out, and their distribution is the important information to be used in the estimation process. The signal to be processed does not distinguish between interference and usefulness, and the task of filtering is to extract all processed signals. Therefore, Kalman filtering is a method based on optimal estimation.

[0066] The discrete time system is as follows:

[0067] Equation of state:

[0068] X k =F k|k-1 X k-1 +L k|k-1 W k-1

[0069] Measurement equation:

[0070] Z k =H k X k +V k

[0071] Where: X k Represents the state quantity, F k|k-1 represents the state transfer matrix, L k|k-1 represents the process noise driving matrix, H k represents the measurement array, W k-1 With V k Represents process noise and measurement noise.

[0072] At the same time, W k and V k satisfy

[0073]

[0074] Where: Q k is the process noise matrix and is non-negative definite; R k is the measurement noise matrix and is positive definite, and we assume that the posterior probability distribution at each step is normally distributed.

[0075] State one-step prediction:

[0076] One-step prediction through k-1 time state estimator To predict the state at time k, we use k-1 measurements Z1, Z2, ... Z k-1 To estimate X k Linear minimum variance. After derivation, we can get:

[0077]

[0078] State Estimation:

[0079] Measure Z using the quantity obtained at time k k Corrected state one-step prediction value Use the one-step prediction value to replace the actual state value X k This will cause errors This leads to estimation errors in the quantity measurement. In filtering theory, this error is called residual (also called innovation). It can be seen that the residual (innovation) contains the error information of one-step prediction. By making corresponding weighted treatment on the innovation, the error term of one-step prediction can be extracted, as follows:

[0080]

[0081] Filter gain:

[0082] To obtain the state estimate at time k We need to obtain the weighted matrix K about the new information k , which is called the gain matrix of the filter. The principle of obtaining the gain matrix is ​​to let the estimated mean square error matrix

[0083] Reach the minimum, ( is the estimated error), the filter gain matrix is ​​deduced as follows:

[0084]

[0085] One-step forecast mean square error:

[0086] According to the definition Can be derived

[0087]

[0088] Estimate the mean squared error:

[0089] According to the definition Combining the relevant assumptions of the above system and through a series of derivations, we can get:

[0090]

[0091] The above five equations are the process of discrete Kalman filter, that is, the initial value is known and P0, using k time to measure Z k , then the estimated value of the state at time k can be derived Figure 4 A schematic diagram of a filtering process according to an exemplary embodiment of the present disclosure is shown.

[0092] The filter essentially has two parts: prediction and update. The prediction part uses the equations that describe the dynamic characteristics of the system to calculate the transition probability density of the predicted state from the previous moment to the current moment. There is unpredictable random noise that affects the state estimation, so the prediction process usually needs to be deformed to propagate the probability distribution of the state. The update part uses the newly acquired quantity measurement to correct the probability distribution of the prediction stage.

[0093] According to some embodiments, the visually perceived road information may include a plurality of perceived lanes, and the plurality of perceived lanes may include a main lane and an adjacent lane where the vehicle is located. Figure 5 As shown, step S204, using visual perception of road information to determine the perceived observation probability that the vehicle is located in each candidate lane may include: step S501, using visual perception of road information to determine a first distance between the vehicle and the left and right sidelines of the main lane; and step S502, based on the first distance, determining the perceived observation probability that the vehicle is located in the candidate lane corresponding to the main lane and the candidate lane corresponding to the adjacent lane.

[0094] Therefore, through the above method, visually perceived road information can be effectively utilized to predict the topological recursive probability of the vehicle being located in each candidate lane from the road perception level, thereby improving the accuracy of lane-level positioning.

[0095] In some embodiments, in step S502, in response to determining that the first distance indicates that the vehicle is not approaching any adjacent lane, a perceived observation probability exceeding a first probability value may be set for a candidate lane corresponding to the main lane, and a perceived observation probability below a second probability value may be set for a candidate lane corresponding to the adjacent lane. In response to determining that the first distance indicates that the vehicle is approaching an adjacent lane, a perceived observation probability exceeding a second probability value may be set for a candidate lane corresponding to the adjacent lane, and a perceived observation probability below the first probability value may be set for a candidate lane corresponding to the main lane.

[0096] According to some embodiments, Figure 6 As shown, step S205, using the positioning information to determine the positioning probability of the vehicle being located in each candidate lane may include: step S601, using the positioning information to determine the second distance between the vehicle and the center line of each candidate lane; and step S602, based on the second distance between the vehicle and each candidate lane, determining the positioning probability of the vehicle being located in each candidate lane.

[0097] Therefore, through the above method, the positioning information can be effectively used to predict the positioning probability of the vehicle in each candidate lane from the positioning level, thereby improving the accuracy of lane-level positioning.

[0098] In some embodiments, the center line of the candidate lane may be obtained from the map road information. In step S602, the determined positioning probability may be negatively correlated with the second distance. That is, the closer the vehicle is to the center line of the candidate lane, the higher the positioning probability corresponding to the candidate lane.

[0099] In step S206, the topological recursive probability, the perception observation probability and the positioning probability corresponding to each candidate lane may be combined to obtain the target probability corresponding to the candidate lane. Furthermore, based on the target probability of each candidate lane, the target lane where the vehicle is most likely to be located may be determined.

[0100] In some embodiments, the above steps S203 to S206 may also be regarded as a process of probability confidence allocation.

[0101] According to some embodiments, Figure 7 As shown, step S206, based on the topological recursive probability, perceptual observation probability and positioning probability of the vehicle being located in each candidate lane, determining the target lane where the vehicle is located among multiple candidate lanes may include: step S701, determining the topological recursive weight, perceptual observation weight and positioning weight of each of the multiple candidate roads; step S702, for each of the multiple candidate roads, normalizing the topological recursive probability, perceptual observation probability and positioning probability of the vehicle being located on the candidate road using the topological recursive weight, perceptual observation weight and positioning weight of the candidate road to obtain the target probability of the vehicle being located on the candidate road; and step S703, determining the target lane based on the target probability of the vehicle being located on each candidate road.

[0102] Therefore, through the above method, the topological recursive weight, the perception observation weight and the positioning weight can be combined more effectively, so that the target road can be determined more accurately among multiple candidate roads.

[0103] In some embodiments, the target probability may be expressed as:

[0104] P 目标 =L 拓扑递推 P 拓扑递推 +L 感知观测 P 感知观测 +L 定位 P 定位

[0105] Among them, P is the probability and L is the weight.

[0106] In some embodiments, the topology recursion weight, the perception observation weight, and the positioning weight may be set based on prior knowledge or may be adjusted dynamically.

[0107] According to some embodiments, Figure 8As shown, step S701, determining the topological recursive weight, perception observation weight and positioning weight of each of multiple candidate roads may include: step S801, in response to determining that the topological recursive probability that the vehicle is located in a historical lane at the previous moment is greater than a preset threshold, setting the topological recursive weight of the vehicle located in a candidate lane corresponding to a historical lane to be not less than a first preset value.

[0108] The topology recursion probability corresponding to the historical lane is greater than the preset threshold, indicating that when performing lane-level positioning in the historical lane, the confidence of the topology recursion-based method is higher, so the weight of this method can be set to a higher value in subsequent predictions. In this way, it is possible to rely more on the topology recursion information for lane-level positioning.

[0109] In some embodiments, in response to determining that the topological recursive probability that the vehicle is located in a historical lane at the last moment is less than a preset threshold, the topological recursive weight of the vehicle being located in a candidate lane corresponding to the historical lane is set to be less than a first preset value. In other words, if the confidence level of the topological recursive method is low when performing lane-level positioning in the historical lane, the corresponding weight can be reduced in subsequent predictions to reduce the use of topological recursive information.

[0110] According to some embodiments, Figure 8 As shown, step S701, determining the topological recursive weights, perceptual observation weights and positioning weights of multiple candidate roads respectively may include: step S802, in response to determining that the number of lanes in the visually perceived road information and the map road information is the same, and in response to determining that the visually perceived road information indicates that the left and right curbs are detected, setting the perceptual observation weight of the vehicle in each candidate lane to not less than a second preset value.

[0111] When the number of lanes in the visual perception result is the same as that in the high-precision map, and the left and right edges of the current road are successfully detected, it means that the confidence of the current visual perception road information is high. Therefore, a higher weight can be set for the lane-level positioning method based on perception observation, thereby improving the accuracy of lane-level positioning.

[0112] According to some embodiments, Figure 8 As shown, step S701, determining the topological recursive weight, perceptual observation weight and positioning weight of each of multiple candidate roads may include: step S803, determining the positioning weight based on the duration of the vehicle entering the lane-level positioning recursive mode, wherein the lane-level positioning recursive mode indicates a mode of performing lane-level positioning using topological recursive probability, perceptual observation probability and positioning probability, and the positioning weight is negatively correlated with the duration.

[0113] The mode of lane-level positioning using topological recursive probability, perception observation probability and positioning probability is called lane-level positioning recursive mode. This mode does not completely rely on positioning information and map information, but performs comprehensive reasoning based on multiple aspects of information. Figure 2 The lane-level positioning recursive mode is described in steps S201 to S206. Since this mode relies less on positioning information and map information, the longer the time in this mode is, the lower the reliability of the recursive information is. Therefore, a smaller weight can be set to ensure the overall accuracy of lane-level positioning.

[0114] After step S206, post-processing operations may be performed. In some embodiments, the post-processing operations may use multi-frame temporal results to maintain a Hidden Markov Model (HMM), realize temporal information transfer, realize spatial alignment of multi-source information (positioning information, map road information, and visual perception road information) through the candidate link segment (1ink) road topology structure, and improve the safety attributes of lane-level positioning post-processing.

[0115] Hidden Markov model map matching has been established as the state-of-the-art approach for both offline and online map matching. A first-order hidden Markov model defines the behavior of a system over time as a sequence of system states (s0, ..., s T ), they are hidden because the actual system state can only be known through a series of measurements (z0, ..., z T ) to observe. Fig. 9 Schematic diagram of a hidden Markov model according to an exemplary embodiment of the present disclosure is shown. Fig. 9 As shown, each measurement z t All are system status t Observation result (wavy arrow) at time t∈[0,t]. The state s at time i t is a finite state candidate set s t Since the observations are affected by measurement noise, their uncertainty can be expressed as the likelihood probability p(z t |s t ) to model the state transition. State transitions, as shown by straight arrows, are transitions with probability p(s t |s t-1 ) is part of a random process.

[0116] The first-order HMM makes the following assumptions about state transitions and estimation:

[0117] Markov state assumption: The system state s at time t∈[1, t] t Depends only on the previous state s t-1 , so that p(s t |s0,...,st-1 )=p(s t |s t-1 ).

[0118] Markov sensor assumption: At time t∈[0,t], the likelihood z t The probability depends only on the current state s t , so that p(z t |z0,...,z t-1 ,s0,...,s t )=p(z t |s t ).

[0119] In the context of map matching, states refer to the vehicle's map position and estimates refer to position measurements. Transitions between system states correspond to paths between map positions. The solution to the online map matching problem follows a recursive approach of state filtering.

[0120] Maximum likelihood location probability Corresponds to the position measurement z t :

[0121] According to the definition:

[0122]

[0123] It can be inferred that:

[0124]

[0125] Where p(s t |z0,...,z t ) is the position candidate s t ∈S t The probability that, given the position measurement (z0,...,z t ). In addition, p(s t-1 |z0,...,z t-1 ) is the position candidate s t-1 ∈S t-1 The probability that, given the position measurement (z0,...,z t-1 ). Position candidates t The likelihood probability is p(z t |s t ), from position s t-1 To location t The transition probability is p, and α is a normalization constant. The probability of the initial position candidate s0∈S0 is:

[0126] p(s0|z0)=αp(z0|s0)

[0127] According to some embodiments, Fig.10 As shown, the lane-level positioning method may further include: step S1002, matching the visually perceived road information with the map road information. Determining multiple candidate lanes and their corresponding topological recursive probabilities, perceptual observation probabilities, and positioning probabilities may be performed in response to determining that the visually perceived road information does not match the map road information. It is understandable that Fig.10 The operations and effects of steps S1001, S1005-S1009 can be referred to Figure 2 Steps S201 to S206 in the above are not described in detail here.

[0128] After obtaining the visual perception road information and the map road information, the map matching module can be used to match the two first. If the match fails, it triggers the determination of candidate lanes and calculates the topological recursive probability, perception observation probability and positioning probability, and then uses these probabilities for lane-level positioning. In other words, if the visual perception road information and the map road information are not successfully matched, the lane-level positioning recursive mode is entered.

[0129] In some embodiments, the visually perceived road information includes visually perceived lane lines, and the map road information includes map lane lines. In step S1002, the visually perceived road information and the map road information may be filtered through histograms to provide matching point pairs, and high-precision availability information may be generated based on the matching point pairs. The high-precision availability information may include: 1) whether the visually perceived lane line matches the map lane line successfully; 2) whether the deviation between the visually perceived lane line and the map lane line is less than a threshold; 3) whether the main vehicle has driven out of the road boundary; 4) whether it is a Y-shaped intersection.

[0130] In response to determining that at least one of the four conditions of 1) the visually perceived lane line matches the map lane line successfully, 2) the deviation is less than a preset deviation threshold, 3) the vehicle has not driven out of the road boundary, and 4) it is not a Y-shaped intersection is not met, it can be considered that the visually perceived road information and the map road information do not match, and steps S1005 to S1009 are executed, that is, recursive lane-level positioning is performed using multi-source sensors.

[0131] In response to determining that all four conditions are met, it can be considered that the visually perceived road information and the map road information match, so that lane positioning can be performed using the high-precision map.

[0132] According to some embodiments, the visually perceived road information may include a plurality of perceived lanes, and the plurality of perceived lanes may include a main lane and an adjacent lane where the vehicle is located. Fig.10As shown, the lane-level positioning method may also include: step S1003, in response to determining that the visually perceived road information matches the map road information, using the visually perceived road information to determine a first distance between the vehicle and the left and right edges of the main lane and a second distance between the vehicle and the center line of at least one of the multiple candidate lanes, the at least one candidate lane at least including a candidate lane corresponding to the main lane; and step S1004, based on the first distance and the second distance, determining the target lane where the vehicle is located among the candidate lanes corresponding to the main lane and the candidate lanes corresponding to the adjacent lanes.

[0133] In some embodiments, the road-level positioning calculation is large and the frequency is low, but the downstream requires a high frequency for lane-level positioning output. Therefore, the link information provided by the road-level positioning may be delayed. Therefore, it is necessary to perform lane subsequent update judgment, that is, to use the lane topology provided by the high-precision map and the position information provided by the fusion positioning to judge whether the main vehicle enters the subsequent lane, and further confirm the main lane where the main vehicle is currently located, so as to prevent lane judgment errors caused by road-level positioning lag and improve the robustness of lane calculation.

[0134] In some embodiments, lane subsequent update judgment may include: calculating the left and right intercepts of the main vehicle from the edge line of the subsequent lane, judging the possible main lane by the different signs of the left and right intercepts; and calculating the angle between the main vehicle and the center line of the lane. If it is an acute angle, the main vehicle has entered the subsequent lane, and then the road information is updated.

[0135] Fig.11 The overall technical flow chart according to an exemplary embodiment of the present disclosure is shown. The input layer 1102 may include a high-precision map 1104, positioning information 1106, map matching information 1108, and visually perceived road information 1110. Among them, the visually perceived road information may further include lane lines 1112 and curbs 1114. According to the map matching information 1108, scene control 1116 may be performed, that is, whether to perform high-precision map lane-level positioning 1110 or recursive lane-level positioning 1118 is determined according to the matching situation. Recursive lane-level positioning 1118 further requires calculation of topological recursive probability 1120, perception observation probability 1122, and positioning probability 1124, and then normalize and post-process the three 1126. Finally, the result output 1128 is performed after high-precision map lane-level positioning 1110 or recursive lane-level positioning 1118.

[0136] The method of the present disclosure is introduced below in conjunction with an exemplary embodiment.

[0137] Fig.12A schematic diagram of lane-level positioning according to an exemplary embodiment of the present disclosure is shown. The white lane line is the HD map lane line, the green is the HD map lane center line, the yellow is the visual perception lane line, and the pink arcs are 5m, 10m, and 50m radius distances from the main vehicle, respectively. This figure shows that the yellow line and the white line have a significant deviation (high offset), that is, the HD map quality is poor. The map matching module will output high-precision unavailable information accordingly, and the lane-level positioning enters the recursive mode.

[0138] In some embodiments, the map road information can be obtained from the high-precision map according to the positioning information output by the road-level positioning, including all lane information bound under the link, i.e. lane_seq 1-3 in the figure. These lanes can be determined as multiple candidate lanes. In addition, the successor relationship of the lanes can be obtained according to the high-precision map near the main vehicle.

[0139] In some embodiments, current visually perceived road information may be obtained, including multiple perceived lanes, namely yellow percep_seq 1-3 in the figure. Lane change probability may be calculated, and the probability of lane change to the left and right of the illustrated position is 0.

[0140] In some embodiments, topology recursion information can be obtained, and the lane topology relationship, i.e., lane_topo, can be maintained through the link output by road-level positioning and considering the hysteresis of road-level positioning, i.e., cur_lane (candidate lane) and last_lane (historical lane) in the figure. Then, probability confidence can be assigned to obtain topology recursion probability, perception observation probability, and fusion positioning probability.

[0141] In some embodiments, regarding the topological recursive probability and weight calculation, since the lane change probability is 0, the probability state transfer matrix is ​​a unit matrix, and the weight is set according to the topological recursive probability used at the last moment. The lane change probability is calculated using the formula given above.

[0142] In some embodiments, regarding the perception observation probability and weight calculation, the current perception has the same number of lanes as the high-precision map and the left and right curbs are detected, and the perception observation weight is set to a larger value; the main vehicle is at the center of the lane and at a larger distance from the left and right side lines, so the probability of the main lane is 1 and the probability of the left and right lanes is 0.

[0143] In some embodiments, regarding the calculation of fusion positioning probability and weight, since the main vehicle has not left the map, the road-level positioning has not entered the recursive mode. The weight of the fusion positioning probability is only set according to the time of entering the lane-level positioning recursive mode. The longer the time, the lower the weight. The probability of fusion positioning is set according to the distance from the center line of the lane.

[0144] In some embodiments, the target probability formula given above can be used for normalization calculation to obtain the probabilities of all candidate lanes, and the lane identification with the highest probability is given as the final lane-level positioning output. Since the road-level positioning in this area does not enter the recursive mode and is not a ramp or tunnel scenario, no special post-processing is required.

[0145] The present disclosure proposes an active lane-level positioning method, which can be used as an important supplement to the lane-level positioning system based on high-precision maps in the assisted driving system. On the basis of the original framework, the lane-level positioning recursive calculation is introduced to deal with the scenes with poor quality of high-precision maps, and improve the effectiveness in the scenes of cornering and reality changes. In summary, the lane-level positioning technology as an important part of the assisted driving technology provides important technical support for the implementation of assisted driving technology, and can achieve the effects of no degradation of tunnel corners, no degradation of map convex hull, no degradation of longitudinal error out of the map, and keeping the positioning of reality changes intact. In addition, by using the method provided by the present disclosure, the availability of intelligent driving in tunnel scenarios is improved by 50%, and the positioning efficiency is improved by 42.9%; in the scene of reality changes, the availability of intelligent driving is improved by 58.8%, and the positioning efficiency is improved by 70.6%; and the problems of lane-level positioning failure and jump at the ramp entrance in normal scenarios are solved; the problem of multi-layer road misalignment is solved.

[0146] According to another aspect of the present disclosure, a lane level positioning device is provided. Fig.13 As shown, the lane-level positioning device 1300 includes: an acquisition unit 1310, configured to acquire map road information and visual perception road information around the vehicle based on the positioning information of the vehicle; a candidate lane determination unit 1320, configured to determine multiple candidate lanes based on the positioning information; a topological recursive probability determination unit 1330, configured to use the map road information to determine the topological recursive probability that the vehicle is located in each candidate lane; a perception observation probability determination unit 1340, configured to use the visual perception road information to determine the perception observation probability that the vehicle is located in each candidate lane; a positioning probability determination unit 1350, configured to use the positioning information to determine the positioning probability that the vehicle is located in each candidate lane; and a first target lane determination unit 1360, configured to determine the target lane where the vehicle is located among multiple candidate lanes based on the topological recursive probability, perception observation probability and positioning probability that the vehicle is located in each candidate lane.

[0147] According to some embodiments, the topological recursive probability determination unit may include: a first determination subunit, configured to determine the topological recursive relationship between multiple historical lanes and multiple candidate lanes using map road information; and a second determination subunit, configured to determine the topological recursive probability that the vehicle is located in each candidate lane based on the topological recursive relationship and the topological recursive probability that the vehicle was located in each historical lane at the previous moment.

[0148] According to some embodiments, the topological recursive probability determination unit may further include: a calculation subunit configured to calculate the lane change probability of the vehicle based on visually perceived road information; and a matrix maintenance subunit configured to maintain a state transition matrix of topological recursive observations based on the lane change probability. The topological recursive probability that the vehicle is located in each candidate lane may be determined based on the state transition matrix, the topological recursive relationship, and the topological recursive probability that the vehicle was located in each historical lane at the last moment.

[0149] According to some embodiments, the visually perceived road information includes a plurality of perceived lanes, and the plurality of perceived lanes include a main lane and an adjacent lane where the vehicle is located. The perceived observation probability determination unit may include: a third determination subunit configured to determine a first distance between the vehicle and the left and right sidelines of the main lane using the visually perceived road information; and a fourth determination subunit configured to determine, based on the first distance, the perceived observation probability that the vehicle is located in a candidate lane corresponding to the main lane and a candidate lane corresponding to the adjacent lane.

[0150] According to some embodiments, the positioning probability determination unit may include: a fifth determination subunit, configured to use the positioning information to determine the second distance between the vehicle and the center line of each candidate main lane; and a sixth determination subunit, configured to determine the positioning probability that the vehicle is located in the candidate lane corresponding to the main lane based on the second distance between the vehicle and each candidate lane.

[0151] According to some embodiments, the target lane determination unit may include: a weight determination subunit, configured to determine the topological recursive weight, perceptual observation weight and positioning weight of each of a plurality of candidate roads; a target probability determination subunit, configured to normalize the topological recursive probability, perceptual observation weight and positioning weight of each of the plurality of candidate roads, that the vehicle is located on the candidate road, to obtain a target probability that the vehicle is located on the candidate road, using the topological recursive weight, perceptual observation weight and positioning weight of the candidate road; and a target lane determination subunit, configured to determine the target lane based on the target probability that the vehicle is located on each candidate road.

[0152] According to some embodiments, the weight determination unit may include: a topological recursive weight determination subunit, configured to set the topological recursive weight of the vehicle being located in a candidate lane corresponding to a historical lane to not less than a first preset value in response to determining that the topological recursive probability that the vehicle was located in a historical lane at the previous moment is greater than a preset threshold.

[0153] According to some embodiments, the weight determination unit may include: a perceptual observation weight determination subunit, configured to set the perceptual observation weight of the vehicle in each candidate lane to not less than a second preset value in response to determining that the number of lanes in the visually perceived road information and the map road information is the same, and in response to determining that the visually perceived road information indicates that the left and right curbs are detected.

[0154] According to some embodiments, the weight determination unit may include: a positioning weight determination subunit, configured to determine the positioning weight based on the duration of the vehicle entering the lane-level positioning recursive mode, wherein the lane-level positioning recursive mode indicates a mode of performing lane-level positioning using topological recursive probability, perception observation probability and positioning probability, and the positioning weight is negatively correlated with the duration.

[0155] According to some embodiments, the lane-level positioning device may further include: a matching unit configured to match the visually perceived road information with the map road information. Determining a plurality of candidate lanes and their respective corresponding topological recursive probabilities, perceptual observation probabilities, and positioning probabilities may be performed in response to determining that the visually perceived road information does not match the map road information.

[0156] According to some embodiments, the visually perceived road information may include multiple perceived lanes, and the multiple perceived lanes may include a main lane and an adjacent lane where the vehicle is located. The lane-level positioning device may also include: a distance determination unit, configured to determine, in response to determining that the visually perceived road information matches the map road information, a first distance between the vehicle and the left and right edges of the main lane and a second distance between the vehicle and the center line of at least one of the multiple candidate lanes using the visually perceived road information, the at least one candidate lane at least including a candidate lane corresponding to the main lane; and a second determination unit, configured to determine, based on the first distance and the second distance, a target lane where the vehicle is located among the candidate lanes corresponding to the main lane and the candidate lanes corresponding to the adjacent lanes.

[0157] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0158] According to an embodiment of the present disclosure, an electronic device, a readable storage medium and a computer program product are also provided.

[0159] refer to Fig.14, a block diagram of an electronic device 1400 that can be used as a server or client of the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0160] like Fig.14 As shown, the device 1400 includes a computing unit 1401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1402 or a computer program loaded from a storage unit 1408 into a random access memory (RAM) 1403. In the RAM 1403, various programs and data required for the operation of the device 1400 can also be stored. The computing unit 1401, the ROM 1402, and the RAM 1403 are connected to each other via a bus 1404. An input / output (I / O) interface 1405 is also connected to the bus 1404.

[0161] Multiple components in device 1400 are connected to I / O interface 1405, including: input unit 1406, output unit 1407, storage unit 1408 and communication unit 1409. Input unit 1406 can be any type of device that can input information to device 1400. Input unit 1406 can receive input digital or character information and generate key signal input related to user settings and / or function control of the electronic device, and can include but is not limited to a mouse, keyboard, touch screen, track pad, track ball, joystick, microphone and / or remote control. Output unit 1407 can be any type of device that can present information, and can include but is not limited to a display, speaker, video / audio output terminal, vibrator and / or printer. Storage unit 1408 can include but is not limited to a disk, an optical disk. Communication unit 1409 allows device 1400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and can include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver and / or a chipset, such as Bluetooth TM devices, 802.11 devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0162] The computing unit 1401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning network algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1401 performs the various methods and processes described above, such as the lane-level positioning method. For example, in some embodiments, the lane-level positioning method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1408. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1400 via the ROM 1402 and / or the communication unit 1409. When the computer program is loaded into the RAM 1403 and executed by the computing unit 1401, one or more steps of the lane-level positioning method described above may be performed. Alternatively, in other embodiments, the computing unit 1401 may be configured to perform the lane-level positioning method in any other appropriate manner (eg, by means of firmware).

[0163] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0164] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0165] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0166] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0167] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0168] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server for a distributed system, or a server combined with a blockchain.

[0169] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0170] Although the embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it should be understood that the above-mentioned methods, systems and devices are merely exemplary embodiments or examples, and the scope of the present invention is not limited by these embodiments or examples, but only by the claims after authorization and their equivalent scope. Various elements in the embodiments or examples can be omitted or replaced by their equivalent elements. In addition, each step can be performed in an order different from that described in the present disclosure. Further, the various elements in the embodiments or examples can be combined in various ways. It is important that with the evolution of technology, many elements described herein can be replaced by equivalent elements that appear after the present disclosure.

Claims

1. A lane-level positioning method, comprising: Based on the positioning information of the vehicle, obtaining map road information and visually perceived road information around the vehicle; determining a plurality of candidate lanes based on the positioning information; Determine the topological recursive probability that the vehicle is located in each candidate lane by using the map road information; Determining a perceived observation probability that the vehicle is located in each candidate lane using the visually perceived road information; Determine the location probability of the vehicle being located in each candidate lane using the location information; as well as Based on the topological recursive probability, the perception observation probability and the positioning probability of the vehicle being located in each candidate lane, a target lane in which the vehicle is located is determined among the multiple candidate lanes.

2. The method according to claim 1, wherein: Using the map road information, determining the topological recursive probability that the vehicle is located in each candidate lane includes: Determining a topological recursive relationship between a plurality of historical lanes and the plurality of candidate lanes using the map road information; and Based on the topological recursive relationship and the topological recursive probability that the vehicle was located in each historical lane at a previous moment, a topological recursive probability that the vehicle is located in each candidate lane is determined.

3. The method according to claim 2, wherein: Using the map road information, determining the topological recursive probability that the vehicle is located in each candidate lane further includes: Calculating a lane change probability of the vehicle based on the visually perceived road information; and Based on the lane change probability, a state transition matrix of topological recursive observation is maintained, The topological recursive probability that the vehicle is located in each candidate lane is determined based on the state transition matrix, the topological recursive relationship, and the topological recursive probability that the vehicle was located in each historical lane at the previous moment.

4. The method according to claim 1, wherein: The visually perceived road information includes a plurality of perceived lanes, and the plurality of perceived lanes include a main lane and an adjacent lane where the vehicle is located, wherein, using the visually perceived road information, determining a perceived observation probability that the vehicle is located in each candidate lane includes: Determining a first distance between the vehicle and the left and right sidelines of the main lane using the visually perceived road information; and Based on the first distance, a perception observation probability that the vehicle is located in a candidate lane corresponding to the main lane and a candidate lane corresponding to the adjacent lane is determined.

5. The method according to claim 1, wherein: Using the positioning information, determining the positioning probability of the vehicle being located in each candidate lane includes: Determining a second distance between the vehicle and a center line of each candidate lane using the positioning information; and Based on the second distance between the vehicle and each candidate lane, a positioning probability of the vehicle being located in each candidate lane is determined.

6. The method according to any one of claims 1 to 5, wherein: Based on the topological recursive probability, the perception observation probability and the positioning probability that the vehicle is located in each candidate lane, determining the target lane where the vehicle is located among the multiple candidate lanes includes: Determine the topological recursive weight, the perception observation weight and the positioning weight of each of the plurality of candidate roads; For each candidate road among the multiple candidate roads, the topological recursive weight, the perceptual observation weight and the positioning weight of the candidate road are used to normalize the topological recursive probability, the perceptual observation probability and the positioning probability that the vehicle is located on the candidate road to obtain a target probability that the vehicle is located on the candidate road; and The target lane is determined based on the target probability that the vehicle is located on each candidate road.

7. The method according to claim 6, wherein: Determining the topological recursive weight, the perceptual observation weight, and the positioning weight of each of the plurality of candidate roads comprises: In response to determining that the topological recursion probability that the vehicle is located in a historical lane at a previous moment is greater than a preset threshold, a topological recursion weight that the vehicle is located in a candidate lane corresponding to the historical lane is set to be no less than a first preset value.

8. The method according to claim 6, wherein: Determining the topological recursive weight, the perceptual observation weight, and the positioning weight of each of the plurality of candidate roads comprises: In response to determining that the number of lanes in the visually perceived road information and the map road information is the same, and in response to determining that the visually perceived road information indicates that left and right curbs are detected, the perception observation weight of the vehicle in each candidate lane is set to be no less than a second preset value.

9. The method according to claim 6, wherein: Determining the topological recursive weight, the perceptual observation weight, and the positioning weight of each of the plurality of candidate roads comprises: The positioning weight is determined based on the duration that the vehicle enters the lane-level positioning recursive mode, wherein the lane-level positioning recursive mode indicates a mode of performing lane-level positioning using topological recursive probability, perception observation probability and positioning probability, and the positioning weight is negatively correlated with the duration.

10. The method according to any one of claims 1 to 5, further comprising: The visually perceived road information is matched with the map road information, wherein determining the multiple candidate lanes and their respective corresponding topological recursive probabilities, perceptual observation probabilities, and positioning probabilities is performed in response to determining that the visually perceived road information does not match the map road information.

11. The method according to claim 10, wherein: The visually perceived road information includes a plurality of perceived lanes, and the plurality of perceived lanes include a main lane and an adjacent lane where the vehicle is located. The method further includes: In response to determining that the visually perceived road information matches the mapped road information, determining, using the visually perceived road information, a first distance between the vehicle and left and right edges of the main lane and a second distance between the vehicle and a centerline of at least one of the plurality of candidate lanes, the at least one candidate lane including at least a candidate lane corresponding to the main lane; and Based on the first distance and the second distance, a target lane in which the vehicle is located is determined from candidate lanes corresponding to the main lane and candidate lanes corresponding to the adjacent lane.

12. A lane-level positioning device, comprising: an acquisition unit, configured to acquire map road information and visually perceived road information around the vehicle based on the positioning information of the vehicle; a candidate lane determining unit, configured to determine a plurality of candidate lanes based on the positioning information; A topology recursion probability determination unit is configured to determine a topology recursion probability that the vehicle is located in each candidate lane by using the map road information; a perception observation probability determination unit, configured to determine a perception observation probability that the vehicle is located in each candidate lane by using the visually perceived road information; a positioning probability determination unit, configured to determine the positioning probability of the vehicle being located in each candidate lane using the positioning information; as well as The first target lane determination unit is configured to determine the target lane where the vehicle is located among the multiple candidate lanes based on the topological recursive probability, the perception observation probability and the positioning probability of the vehicle being located in each candidate lane.

13. The device according to claim 12, wherein: The topology recursion probability determination unit includes: A first determining subunit is configured to determine a topological recursive relationship between a plurality of historical lanes and the plurality of candidate lanes using the map road information; and The second determining subunit is configured to determine a topological recursive probability that the vehicle is located in each candidate lane based on the topological recursive relationship and the topological recursive probability that the vehicle was located in each historical lane at a previous moment.

14. The device according to claim 13, wherein: The topology recursion probability determination unit also includes: a calculation subunit configured to calculate a lane change probability of the vehicle based on the visually perceived road information; and A matrix maintenance subunit is configured to maintain a state transition matrix of topological recursive observation based on the lane change probability, The topological recursive probability that the vehicle is located in each candidate lane is determined based on the state transition matrix, the topological recursive relationship, and the topological recursive probability that the vehicle was located in each historical lane at the previous moment.

15. The device according to claim 12, wherein: The visually perceived road information includes a plurality of perceived lanes, and the plurality of perceived lanes include a main lane and an adjacent lane where the vehicle is located, wherein the perceived observation probability determination unit includes: A third determining subunit is configured to determine a first distance between the vehicle and the left and right sidelines of the main lane by using the visually perceived road information; and The fourth determination subunit is configured to determine, based on the first distance, a perceived observation probability that the vehicle is located in a candidate lane corresponding to the main lane and a candidate lane corresponding to the adjacent lane.

16. The device according to claim 12, wherein: The positioning probability determination unit comprises: a fifth determining subunit, configured to determine a second distance between the vehicle and a center line of each candidate lane using the positioning information; and The sixth determination subunit is configured to determine a positioning probability that the vehicle is located in each candidate lane based on a second distance between the vehicle and each candidate lane.

17. The device according to any one of claims 12 to 16, wherein: The target lane determination unit comprises: A weight determination subunit, configured to determine a topological recursive weight, a perception observation weight, and a positioning weight of each of the plurality of candidate roads; a target probability determination subunit configured to normalize, for each candidate road among the plurality of candidate roads, the topological recursive weight, the perceptual observation weight and the positioning weight of the candidate road, the topological recursive probability, the perceptual observation probability and the positioning probability that the vehicle is located on the candidate road, to obtain a target probability that the vehicle is located on the candidate road; and The target lane determination subunit is configured to determine the target lane based on the target probability that the vehicle is located on each candidate road.

18. The device according to claim 17, wherein: The weight determination unit comprises: The topology recursion weight determination subunit is configured to set the topology recursion weight of the vehicle being located in a candidate lane corresponding to the historical lane to be not less than a first preset value in response to determining that the topology recursion probability that the vehicle was located in a historical lane at the previous moment is greater than a preset threshold.

19. The device according to claim 17, wherein: The weight determination unit comprises: The perception observation weight determination subunit is configured to set the perception observation weight of the vehicle in each candidate lane to not less than a second preset value in response to determining that the number of lanes in the visual perception road information and the map road information is the same, and in response to determining that the visual perception road information indicates that the left and right road edges are detected.

20. The device according to claim 17, wherein: The weight determination unit comprises: A positioning weight determination subunit is configured to determine the positioning weight based on the duration that the vehicle enters the lane-level positioning recursive mode, wherein the lane-level positioning recursive mode indicates a mode of performing lane-level positioning using topological recursive probability, perception observation probability and positioning probability, and the positioning weight is negatively correlated with the duration.

21. The apparatus according to any one of claims 12 to 16, further comprising: a matching unit configured to match the visually perceived road information with the map road information, Wherein, determining the multiple candidate lanes and their respective corresponding topological recursive probabilities, perception observation probabilities and positioning probabilities is performed in response to determining that the visually perceived road information does not match the map road information.

22. The device according to claim 21, wherein The visually perceived road information includes a plurality of perceived lanes, the plurality of perceived lanes including a main lane and an adjacent lane where the vehicle is located, and the device further includes: a distance determination unit configured to determine, in response to determining that the visually perceived road information matches the map road information, a first distance between the vehicle and the left and right edges of the main lane and a second distance between the vehicle and a center line of at least one of the plurality of candidate lanes using the visually perceived road information, the at least one candidate lane including at least a candidate lane corresponding to the main lane; and The second determining unit is configured to determine, based on the first distance and the second distance, a target lane where the vehicle is located from among candidate lanes corresponding to the main lane and candidate lanes corresponding to the adjacent lane.

23. An electronic device, comprising: at least one processor; as well as a memory communicatively coupled to the at least one processor; in The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 11.

24. A vehicle comprising the electronic device according to claim 23.

25. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-11.

26. A computer program product comprising a computer program, wherein: The computer program implements the method according to any one of claims 1-11 when executed by a processor.

Images