Vehicle positioning method in storage yard area, vehicle-mounted unit, server and vehicle positioning system

Abstract

The invention discloses a vehicle positioning method and a vehicle positioning system in a storage yard area, which can reduce transverse errors, longitudinal errors and course errors generated in the driving process of a vehicle in the storage yard area without being influenced by an external environment. The vehicle positioning method comprises the steps that based on initial vehicle position information of a vehicle, whether the vehicle is located in an entrance and exit section corresponding to a storage yard or not is judged, and the initial vehicle position information is obtained through a vehicle multi-sensor fusion filter; and when it is determined that the vehicle is located in the entrance and exit section corresponding to the storage yard, the vehicle is positioned at least based on the initial vehicle position information of the vehicle, the real-time information of the container at the position closest to the entrance and exit section of the storage yard and pre-stored default information.

Description

Technical Field

[0001] The present application relates to the field of port vehicle positioning and autonomous driving. Specifically, the present application relates to a vehicle positioning method, an on-vehicle unit, a server, and a vehicle positioning system in a yard area. Background Art

[0002] It is known that enabling vehicles to automatically drive and operate in the yard area of a port through autonomous driving technology can significantly reduce labor costs and capital costs. At the same time, whether the automatic operation of a vehicle in the yard area can proceed smoothly depends to a certain extent on whether the vehicle can drive on the roads in the yard area along a pre-established planned path without deviating from the road. Therefore, during the process of a vehicle automatically driving in the yard area of a port, there are relatively high requirements for the positioning accuracy of the vehicle, and it is expected that the vehicle has small lateral error, longitudinal error, and heading error.

[0003] Currently, when positioning a vehicle driving in the yard area of a port, when the vehicle is operating in a yard (each yard is at least 200 meters long, and some operations need to cross multiple yards), due to factors such as high containers and yard cranes blocking in the yard, the state of a satellite navigation system such as GPS is poor, resulting in the situation that it is unavailable for a long time or its accuracy is significantly reduced. Due to the long-term unavailability or significant reduction in accuracy of the satellite navigation system, the lateral observation, longitudinal observation, and heading observation of the vehicle driving are not sufficient. As a result, insufficient observation will lead to relatively large lateral error, longitudinal error, and heading error of the vehicle. Moreover, when the vehicle turns out of the yard area, the longitudinal error will turn into a lateral error, causing the vehicle to deviate from the lane, leading to a series of subsequent problems.

[0004] For this reason, the currently commonly used method is to manually arrange some observation sources, such as manually demarcating lane lines and manually drawing some arrows and yard exit stop lines along the lane lines, and identifying the lane lines, arrows, and yard exit stop lines through cameras and lidar, so as to reduce the lateral error, longitudinal error, and heading error generated during the process of the vehicle driving in the yard area respectively.

[0005] However, in the case of water accumulation on the ground on rainy days or the above-mentioned signs being covered by snow on snowy days, it is difficult for lidar and cameras to extract features related to these signs, resulting in the failure of the above-mentioned method. On the other hand, during the long-term operation of the vehicle, these signs may be severely worn due to the long-term action of vehicle tires and often need to be redrawn, increasing the on-site operation cost. Summary of the Invention

[0006] Technical Problems to be Solved by the Invention

[0007] This application is formed to solve the above problems, and its purpose is to provide a vehicle positioning method in a yard area, which can reduce the lateral error, longitudinal error and heading error generated during the vehicle's driving in the yard area without being affected by the external environment. On this basis, the present invention further provides an on-vehicle unit, a server, a vehicle positioning system, a computer-readable storage medium and a computer program product for implementing the above vehicle positioning method.

[0008] Technical solutions for solving technical problems

[0009] A technical solution of this application provides a vehicle positioning method in a yard area. The yard area includes a plurality of yards arranged in a grid pattern in two directions orthogonal to each other. There are roads extending in the other direction of the two directions and for vehicles to drive formed between the yards adjacent in one of the two directions.

[0010] It is characterized in that

[0011] The road includes import and export sections corresponding to each yard arranged in the extending direction of the road. The import section is located on one side of each yard in the extending direction.

[0012] The vehicle positioning method includes:

[0013] Based on the initial vehicle position information of the vehicle, determine whether the vehicle is located in the import section corresponding to a yard. The initial vehicle position information is obtained through a filter system of vehicle multi-sensor fusion; and

[0014] In the case where it is determined that the vehicle is located in the import section corresponding to the one yard, position the vehicle based on at least the initial vehicle position information of the vehicle, the real-time information of a container at the position closest to the import section of the one yard, and the pre-stored default information.

[0015] Advantages of the invention

[0016] According to the vehicle positioning method of this technical solution, high-precision positioning of the vehicle at the yard import and export can be achieved without any manual intervention and completely unaffected by weather changes. Description of the drawings

[0017] Figure 1 It is a partial schematic diagram showing an example of the application scenario of this application, namely the yard area.

[0018] Figure 2 It is a partial schematic diagram showing another example of the application scenario of this application, namely the yard area.

[0019] Figure 3 It is a network topology diagram of a vehicle positioning system showing an embodiment of the present application.

[0020] Figure 4 It is a schematic diagram showing a first modification example of the vehicle positioning system according to an embodiment of the present application.

[0021] Figure 5 It is a schematic diagram showing a second modification example of the vehicle positioning system according to an embodiment of the present application.

[0022] Figure 6 It is a flowchart showing a vehicle positioning method according to an embodiment of the present application.

[0023] Figure 7 It is a flowchart showing a first modification example of the vehicle positioning method according to an embodiment of the present application.

[0024] Figure 8 It is a flowchart showing a second modification example of the vehicle positioning method according to an embodiment of the present application. Detailed implementation manners

[0025] Before describing the embodiments of the present application, for the convenience of understanding, first, the meaning of the "yard area" involved in the present application will be described.

[0026] The "yard area" described in the present application refers to the yard area of containers, which is the area where the yards dedicated to storing containers are located in ports, terminals or inland freight stations. The yard area can have multiple yards, each yard can have multiple bays, each bay can have multiple columns, and each column can have multiple layers. Figure 1 A partial schematic diagram showing an example of the yard area is shown. As Figure 1 shown, in this partial area of the yard area, multiple ( Figure 1 nine in the figure) yards, namely the first yard S1 to the ninth yard S9, are provided, and the floor areas of each yard are the same. Taking the first yard S1 as an example, the first yard S1 has four bays, namely the first bay B1 to the fourth bay B4, and the first bay B1 to the fourth bay B4 all have four columns. In addition, Figure 1 the figure shows the number of layers of containers at each column of each bay. For example, at the first column of the first bay B1, the number of layers of containers is four, which means that four layers of containers are stacked at the first column of the first bay B1. In addition, from Figure 1 it can be observed that the nine yards S1 to S9 are arranged in a grid pattern in two orthogonal directions, namely the X direction and the Y direction, and roads for vehicles to extend, namely the first road R1 to the fourth road R4, are formed between adjacent yards in the X direction or the Y direction and extend along the Y direction or the X direction. It should be noted that Figure 1Only one layout of the yard area is shown in the figure. The number of yards is not limited to nine and can be any number as long as the area has at least one yard. In addition, the yards in the yard area are arranged in a grid pattern in two directions orthogonal to each other. It is not limited to the case where a yard is arranged at the intersection of each row and each column, and also includes the case where there is no yard arranged at the intersection of some rows and columns (see Figure 2 ), and also includes the case where each row or each column is offset. In addition, in Figure 1 , the case where the floor areas of all yards are the same is shown, but it is not limited to this. That is to say, the floor areas of the yards do not have to be the same and there can be certain differences.

[0027] In addition, different from the prior art, in the present application, each of the first road R1 to the fourth road R4 is divided into sections. Hereinafter, taking the first road R1 in the yard area shown in Figure 1 as an example for description. As shown in Figure 1 , the first road R1 includes an internal section, an import section, and an export section corresponding to each yard arranged in the extending direction of the first road R1, that is, the X direction. Specifically, when the vehicle travels from the left side to the right side of the first road R1, for the first yard S1, the first road R1 includes a first yard internal section R1A, a first yard import section R1B, and a first yard export section R1C corresponding to the first yard S1. Among them, the first yard internal section R1A is located within the range of the first yard S1 in the X direction, and the first yard import section R1B and the first yard export section R1C are respectively located on both sides of the first yard S1 in the X direction. For other yards, the same applies. At the same time, for the fourth yard S4, the first yard internal section R1A, the first yard import section R1B, and the first yard export section R1C of the first road R1 also respectively constitute the fourth yard internal section, the fourth yard import section, and the fourth yard export section corresponding to the fourth yard S4. In other words, in the yard area shown in Figure 1 , for two yards located on both sides of the same road and opposite to each other, the import section, the export section, and the internal section are the same and completely coincide with each other. In addition, it can be noted that in the yard area shown in Figure 1 , the first yard export section R1C also constitutes the second yard import section, and so on. In addition, it should be noted that the above import section and export section are relative to the driving direction of the vehicle. When the driving direction of the vehicle is reversed, the import section becomes the export section, and the export section becomes the import section.

[0028] Hereinafter, with reference to Figures 3 to 5 , the main components of the vehicle positioning system A of an embodiment of the present application and its modified examples will be described.

[0029] As Figure 3 shown, the vehicle positioning system A of this embodiment adopts a C / S network topology architecture (i.e., client / server network topology architecture). In this embodiment, the vehicle positioning system A includes a server SV and multiple in-vehicle units Cn (where n ∈ natural numbers) serving as clients.

[0030] The server SV realizes the positioning of vehicles traveling in the yard area by communicating with each in-vehicle unit Cn and other devices via the network. As Figure 3 shown, the server SV mainly includes a control unit SV0, a storage unit SV1, and a communication module SV2. The control unit SV0 performs various controls according to instructions or commands sent by other units of the server SV. The storage unit SV1 stores at least a digital map of the yard area, and the digital map at least includes default information of each yard in the yard area, default information of a container located at the position of the import section closest to each yard, and geographical information of each road. The default information of each yard includes yard number information, geographical coordinate information, floor area information, geometric information, boundary range information, etc. The default information of a container located at the position of the import section closest to each yard at least includes the container number, type code, center point plane coordinate information (sometimes also called position information), and geometric information. Among them, the container number is a unique identification number used to determine the container, the type code indicates the type of the container (for example, general container, refrigerated container, tank container, etc.), the center point plane coordinate information represents the two-dimensional coordinate information of the center point (for example, centroid) of the container in the map coordinate system of the digital map, and the geometric information includes the length dimension information, width dimension information, and height dimension information of the container, etc. Of course, the default information of the above-mentioned container may also include other information, for example, it may also include the material information, color information, cargo information, usage status information, etc. of the container. The geographical information of each road includes, for example, road number, road length, road width, road direction (one-way or two-way), road geometric information, road sign information, road surface marking information (for example, lane line information), etc. The communication module SV2 is used to send electronic data such as the digital map of the yard area stored in the storage unit SV1, and receive requests from the outside. Specifically, the communication module SV2 can distribute electronic data such as the digital map of the yard area stored in the storage unit SV1 to each in-vehicle module Cn installed in each vehicle, and can receive various information (for example, the updated digital map described later) and requests (for example, update requests) from each in-vehicle module Cn.

[0031] Regarding the digital map of the yard area, in the case where the on-vehicle unit without a vehicle sends an update request and the updated digital map to the server SV, the initial digital map is stored in the storage unit SV1 of the server SV. Regarding the formation and acquisition of the initial digital map, first, the overall topographic map of the yard area and a part of the default information of each yard in the yard area (at least including the number information, geographical coordinate information, floor area information, geometric information, boundary range information of each yard) and the geographical information of all roads located between the yards are obtained according to the satellite navigation system and / or other technical means, thereby forming the default information of all yards and the geographical information of all roads located between the yards. Usually, in order to ensure the efficiency and safety of port operation, the above-mentioned default information of each yard is predefined, so these default information can be obtained. On this basis, the length or width information and other information of the containers in the peripheral area of each yard are collected by a handheld device around each yard as the default information of the containers. Since the lengths and widths of containers of various specifications (such as 20 feet, 40 feet, 45 feet, etc.) are determined, as long as the length or width of one container can be collected, the size specification of the container can be determined. In addition, since each bay in each yard usually places containers of the same size specification, although the size information of the containers in the middle bay and far from the road side in each yard cannot be collected by the handheld device, because the size information of the containers in the middle bay and close to the road side can be collected by the handheld device, it can be assumed that the size specifications of the containers in each column of this middle bay are the same as those of the containers close to the road side collected. In addition, through the collection by the handheld device, the highest container layer height of each yard can be determined, and thus it is assumed that the number of layers of the containers in each column of each bay is this number of layers. That is to say, it is assumed that each column of each bay in each yard is filled with containers. On the basis of the above information, the default information of the containers is added, thereby forming an initial digital map including the default information of each yard in the yard area, the geographical information of each road, and the default information of the containers in each yard based on the above assumptions. However, the information included in the initial digital map of the yard area is not limited to the above information. For example, it is also possible to collect only the size information of the containers at the position of the import section closest to the road in each yard as the default information of these containers. That is to say, the initial digital map of the yard area at least includes the default information of each yard, the default information of one container at the position of the import section closest to the road in each yard, and the geographical information of each road.On the other hand, as described later, when a vehicle traveling within the yard area detects a change in certain information, for example, when a vehicle traveling within the yard area detects a change in certain information of a container (e.g., a change in dimensions), the corresponding information in the received digital map of the yard area is updated, and the updated digital map is sent back to the server SV as the updated initial digital map.

[0032] Regarding each on-vehicle unit Cn, hereinafter, in order to avoid repetitive description, taking the first on-vehicle unit C1 as an example, the main modules of each on-vehicle unit will be described. As Figure 3 shown, the first on-vehicle unit C1 mainly includes a receiving module C1A, a determination module C1B, a positioning module C1C, and a transmitting module C1D.

[0033] The receiving module C1A is used to receive the initial vehicle position information of the vehicle on which the above-mentioned first on-vehicle unit C1 is installed, as well as the digital map of the yard area where the vehicle is located sent from the outside. The initial vehicle position information is obtained, for example, from the global positioning system or the Beidou satellite navigation system through a satellite navigation receiver installed on the vehicle, and the digital map of the yard area where the vehicle is located is sent from the server SV, for example. When the receiving module C1A receives the latest digital map of the yard area distributed from the server SV, the above-mentioned latest digital map of the yard area is passed to the positioning program of the positioning module C1C for update.

[0034] The determination module C1B determines whether the vehicle on which the first on-vehicle unit C1 is installed is located in the import section corresponding to a yard within the yard area according to the initial vehicle position information of the vehicle received from the receiving module C1A. For example, in Figure 1 the case of the yard area shown, when the vehicle is located on the first road R1, the determination module C1B determines whether the vehicle on which the first on-vehicle unit C1 is installed is located in the first yard import section S1B corresponding to the first yard S1 according to the initial vehicle position information of the vehicle received from the receiving module C1A.

[0035] The positioning module C1C has a positioning program. When the determination module C1B determines that the vehicle equipped with the first vehicle-mounted unit C1 is located in the import section corresponding to a yard in the yard area, it positions the vehicle based on at least the initial vehicle position information of the vehicle and the real-time information or default information of a container at the position closest to the import section of the above-mentioned yard to obtain the actual position information of the vehicle. The "real-time information" described here may include the center point plane coordinate information, geometric information, etc. of the container obtained in real time. The "default information" described here refers to the center point plane coordinate information, geometric information, etc. of the container in the above-mentioned default information pre-stored in the digital map. In addition, the real-time information about the container can be obtained in real time through other devices installed on the vehicle. For example, the geometric information and center point plane coordinate information of the container can be obtained through a lidar or a multi-modal camera installed on the vehicle. Specifically, the initial digital map from the server SV is transmitted to the positioning program of the positioning module C1C via the receiving module C1A as the original digital map loaded when the vehicle positioning program is started. At this time, the state of the container in the default digital map is not involved in positioning. During the subsequent driving process, the latest digital map stored in the server SV is transmitted to the positioning module C1C in real time via the receiving module C1A and updated to the positioning program. At this time, the state of the container in the digital map becomes involved in positioning.

[0036] The sending module C1D sends, for example, the updated digital map and other electronic data and / or various requests (such as update requests).

[0037] Figure 4 The schematic diagram of the first variant of the vehicle positioning system A, namely the vehicle positioning system A1, of the above-mentioned embodiment is shown. The difference from the vehicle positioning system A of the above-mentioned embodiment is that in this first variant, the first vehicle-mounted unit C1 further includes a matching module C1E and a correction module C1F, and the same applies to other vehicle-mounted units. The matching module C1E is used to match the real-time information of the obtained container with the default information of the container pre-stored in the digital map when the determination module C1B determines that the vehicle equipped with the first vehicle-mounted unit C1 is located in the import section corresponding to a yard in the yard area. For example, the matching module C1E matches the geometric information of the container obtained in real time with the geometric information of the container pre-stored in the digital map to obtain a matching result. The correction module C1F corrects the initial vehicle information of the vehicle equipped with the first vehicle-mounted unit C1 according to the matching result obtained through the matching module. After obtaining the correction result, the positioning module C1C of the first vehicle-mounted unit C1 positions the vehicle according to the above correction result.

[0038] Figure 5The figure shows a schematic diagram of a second variant of the vehicle positioning system A of the above-described embodiment, namely the vehicle positioning system A2. The difference from the vehicle positioning system A1 of the first variant described above is that in this second variant, the first on-vehicle unit C1 further includes an update module C1G, and the same applies to other on-vehicle units. The update module C1G has the following functions:

[0039] (1) Determine whether the size of the container located at this position has changed according to the matching result obtained by the matching module C1E; and

[0040] (2) In the case where it is determined that the size of the container located at the above position has changed, replace the default information of the container located at this position with the above-obtained real-time information as the new default information, and update the digital map of the yard area according to the new default information.

[0041] When the update module C1G updates the digital map of the yard area, the transmission module C1D of the first on-vehicle unit C1 transmits the updated digital map, for example, transmits the updated digital map to the server SV or other on-vehicle units or other devices.

[0042] On the other hand, it should be noted that when the update module C1G determines that the size of the container located at the above position has changed, the matching module C1C still matches the geometric information of the container located at this position obtained in real time with the geometric information of this container pre-stored in the digital map. However, in the case where it is determined that the size of the container located at a certain position has not changed, the size specifications of the containers being compared are the same, while in the case where it is determined that the size of the container located at this position has changed, the size specifications of the containers being compared are different.

[0043] Hereinafter, with reference to Figures 6 to 8 , the main processes of the vehicle positioning method of an embodiment of the present application and its variants will be described. It should be noted that the following embodiments and their variants are only examples, and do not limit the main processes of the vehicle positioning method and the order of each step in the process. That is to say, as long as the technical purpose of the present application can be achieved, the main process may further include other steps, and the order of the steps in the main process can be swapped.

[0044] First, with reference to Figure 6 , the main process of the vehicle positioning method of an embodiment of the present application will be described.

[0045] As Figure 6As shown, first, in step ST1, based on the initial vehicle position information of the vehicle within the yard area, it is determined whether the vehicle is located in the import section corresponding to one of the yards within the yard area. As described above, the initial vehicle position information of the vehicle is obtained, for example, by receiving the position information sent by the satellite navigation system through a satellite navigation receiver (e.g., GPS receiver or Beidou satellite receiver) installed on the vehicle. Specifically, as described above, each road within the yard area is divided into an import section, an internal section, and an export section corresponding to each yard arranged in the extending direction of the road. On the digital map of the yard area, it is possible to determine which belong to the import section, the internal section, or the export section based on the coordinate information of the grids constituting each road. Thus, based on the initial vehicle information of the vehicle, it is possible to determine which section of the road the vehicle is located in, and more specifically, it is possible to determine which section of the road the vehicle is located in and which yard it corresponds to. In the case where it is determined that the vehicle is not located in the import section corresponding to a certain yard, a re-determination is made at a preset time interval. On the other hand, in the case where it is determined that the vehicle is located in the import section corresponding to a certain yard, step ST2 is entered.

[0046] In step ST2, the real-time information of a container located at the position closest to the import section of the above-mentioned certain yard is obtained. Specifically, for example, the point cloud of the target scene located in front of the import section is obtained by scanning the target scene with a lidar installed on the vehicle. Then, target recognition is performed on the point cloud to identify a container at the position closest to the import section of the yard in the point cloud, and at the same time, at least the geometric information and the center point plane coordinate information of the container are identified. The "geometric information of the container obtained in real time" mentioned here is, for example, the fitting plane information formed by fitting the four sides of the container through the point cloud, which is shown by a plane equation. After completing the operation of step ST2, step ST3 is entered.

[0047] In step ST3, default information matching is performed based on the initial vehicle position information of the vehicle and the identified geometric information and center point plane coordinate information of the container. Specifically, based on the initial vehicle position information of the vehicle and the geometric information and center point plane coordinate information of the container obtained in real time, for example, through the kdtree retrieval algorithm, the default information of each container pre-stored in the digital map is searched to find a container located at a substantially same position, as well as the geometric information and center point coordinate information of the container. The "geometric information of the container pre-stored in the digital map" mentioned here is, for example, the default plane information of the four sides of the container, which is also shown by a plane equation. After completing step ST3, step ST4 is entered.

[0048] In step ST4, the vehicle is positioned based on at least the initial vehicle position information of the vehicle described above, the real-time information of the above-mentioned one container obtained, and the corresponding default information. As described above, when there are high containers or yard cranes on the driving section of the vehicle, the navigation of the satellite navigation system may have poor accuracy or even malfunction. Therefore, if only based on the initial vehicle position information of the vehicle obtained by the satellite navigation system, it will cause deviations in the longitudinal and lateral observations of the vehicle, which may trigger a series of subsequent problems. For this reason, in the present embodiment, based on the initial vehicle position information of the vehicle, the vehicle is accurately positioned further according to the real-time information of one container at the position closest to the import section of the yard and the pre-stored default information. Specifically, first, the four sides of the above-mentioned one container obtained in real time and the four sides of the container stored in the digital map in advance are placed in the same coordinate system (for example, the map coordinate system) through coordinate conversion. Then, for example, perpendicular lines are drawn from the coordinate points shown in the initial vehicle position information of the vehicle to the fitting plane and the default plane of each of the four sides of the above-mentioned one container, and two distances from the vehicle to the fitting plane and the default plane can be obtained, thereby obtaining the difference between the two distances. Then, according to this difference, the initial vehicle position information, that is, the coordinates of the initial vehicle position, of the vehicle is corrected, thereby obtaining the corrected vehicle position information.

[0049] According to the vehicle positioning method described in the present embodiment, the vehicle is positioned by simultaneously based on the initial vehicle position information of the vehicle, the real-time information of one container at the position closest to the import section of a yard in front of the import section of the road, and the corresponding default information. In this way, it is not necessary to draw ground marking lines on the road, and even not necessary to draw lane lines, and the vehicle can be accurately positioned. As a result, there is no need for manual intervention at all, and the positioning accuracy is not affected by weather changes.

[0050] Next, with reference to Figure 7 , the main process of the first modification example of the vehicle positioning method of the above embodiment will be described. To avoid repeated description, only the differences between this first modification example and the above embodiment will be described here.

[0051] As Figure 7As shown, as an alternative to step ST1, this first modified example includes step ST1A. Specifically, in step ST1A, based on both the initial vehicle position information of the vehicle and the image information or point cloud information of the scene in front of the vehicle, it is determined whether the vehicle is located in the import section corresponding to a yard. Sometimes, due to poor or even malfunctioning positioning accuracy of the satellite navigation system, there may be a large deviation in the initial positioning coordinates of the vehicle. In this case, if the determination of whether it is in the import section is made only based on the initial vehicle position information, misjudgment may occur. Therefore, the image information or point cloud information of the scene in front of the vehicle is added to the initial vehicle position information for the determination. For example, by performing object recognition on the image or point cloud of the scene in front of the vehicle to obtain real-time information of the container (for example, a certain right-angled side of the container), based on the real-time information of the container, it can be determined whether the vehicle is located in the import section of the yard, in the internal section of the yard, or in the export section of the yard. Preferably, based on both the initial vehicle position information of the vehicle and the image information or point cloud information of the above-mentioned scene obtained multiple times (for example, three times), it is determined whether the vehicle is located in the import section corresponding to a yard.

[0052] According to the vehicle positioning method described in this first modified example, the determination accuracy of whether the vehicle is located in the import section can be improved, thereby further improving the positioning accuracy of the vehicle.

[0053] Next, with reference to Figure 8 , the main process of the second modified example of the vehicle positioning method of the above embodiment will be described. To avoid repeated description, only the differences between this second modified example and the above embodiment will be described here.

[0054] As Figure 8As shown, this second variant example further includes step ST2A. When it is determined in step ST1 that the vehicle is located in the import section corresponding to a certain yard, based on the real-time information of a container at the position closest to the import section of the certain yard and the default information of the container at the same position in the digital map, it is determined whether the size of the container at this position has changed. Sometimes, due to the frequent movement of containers in the yard, it may cause changes in the size or placement position of the container at the position closest to the import section of the yard. Therefore, it is necessary to update the corresponding information on the digital map. Specifically, when it is determined that the vehicle is located in the import section corresponding to a certain yard, a camera or lidar installed on the vehicle captures the scene ahead to obtain a digital image or point cloud of the scene ahead. Then, object recognition is performed on the digital image or point cloud of the scene ahead, and a container closest to the vehicle (i.e., closest to the import section) is identified from the image or point cloud and its real-time information (e.g., real-time geometric information) is determined. In addition, based on the initial vehicle position information of the vehicle and through algorithms such as kdtree retrieval, etc., the default information (e.g., default geometric information) of a container located at the same position in the digital map can be determined. Then, by comparing the above real-time information with the above default information, it is determined whether the size of a container at this position has changed. When it is determined that the size of a container at this position has not changed at all, the digital map is not updated, and when it is determined that the size of a container at this position has changed, step ST3A is entered.

[0055] In step ST3A, the default information of the container at the above position in the digital map is updated to the above real-time information to update the digital map. After completing the update of the digital map, step ST4A is entered.

[0056] In step ST4A, the updated digital map is sent. In connection with Figure 5 the vehicle positioning system A2 shown, after the determination module C1B determines that the size of a container at the above position has changed and the update module C1G updates the relevant information in the digital map sent from the server SV, the sending module C1D transmits the updated digital map and the map update request together via the network to the communication module SV2 of the server SV. After the communication module SV2 of the server SV receives the above updated digital map, it further transmits the updated digital map to the storage unit SV1, and replaces the digital map originally stored in the storage unit SV1 with the above updated digital map.

[0057] According to the vehicle positioning method described in the second modification example, it is possible to achieve the accurate positioning of the vehicle while achieving the timely update of the digital map, thereby ensuring the positioning accuracy of subsequent vehicles.

[0058] In addition, the present application also provides a computer-readable storage medium, which may be included in the computer system described in the above embodiments, or may exist alone without being assembled into the computer system. The above computer-readable storage medium carries one or more programs. When the above one or more programs are executed by a computer system, the computer system implements the methods as described in the above embodiments and their modification examples. For example, the above computer system can implement Figures 6 to 8 each of the steps shown.

[0059] Furthermore, the present application also provides a computer program product, which includes computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in various alternative implementation manners of the above embodiments and modification examples.

[0060] It should be understood that the present invention is not limited to the exact structures that have been described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A vehicle positioning method in a yard area, where the yard area includes a plurality of yards arranged in a grid pattern in two directions orthogonal to each other. There are roads formed between adjacent yards in one of the two directions, extending along the other direction of the two directions for vehicles to travel. Characterized in that, The road includes import and export sections corresponding to each yard arranged in the extending direction of the road. The import and export sections are located on one side of each yard in the extending direction. The vehicle positioning method includes: Based on the initial vehicle position information of the vehicle, determining whether the vehicle is located in the import and export section corresponding to a yard. The initial vehicle position information is obtained through a filter of multi-sensor fusion; and When it is determined that the vehicle is located in the import and export section corresponding to the yard, positioning the vehicle at least based on the initial vehicle position information of the vehicle, the real-time information of a container at the position closest to the import section of the yard, and the pre-stored default information.

2. The vehicle positioning method in the yard area according to claim 1, characterized in that, When it is determined that the vehicle is located in the import and export section corresponding to the yard, matching the real-time information of a container at the position closest to the import section with the default information, and Correcting the initial vehicle position information according to the result of the matching, and positioning the vehicle according to the corrected result.

3. The vehicle positioning method in the yard area according to claim 2, characterized in that, The real-time information includes the fitting plane information formed by point cloud fitting or image processing on the four sides of a container at the position closest to the import section. The default information includes the default plane information of the four sides. Correct the initial vehicle position information according to the matching result of the fitting plane information and the default plane information.

4. The vehicle positioning method in the yard area according to claim 1, characterized in that, Determine whether the vehicle is located in the import and export section corresponding to the yard based on the initial vehicle position information of the vehicle and the image information or point cloud information of the scene in front of the vehicle at the same time.

5. The vehicle positioning method in the yard area according to claim 1, characterized in that, When it is determined that the vehicle is located in the import and export section corresponding to the yard, the vehicle positioning method further includes: Determine whether the size of a container at the position closest to the import and export section has changed according to the real-time information of the container at the position closest to the import and export section and the default information of the container at that position; and When it is determined that the size of the container at the position has changed, update the default information to the real-time information.

6. A vehicle-mounted unit is installed on a vehicle traveling in a yard area. The yard area includes a plurality of yards arranged in a grid pattern in two directions orthogonal to each other. A road for the vehicle to travel is formed between the adjacent yards in one of the two directions and extends in the other direction of the two directions. Characterized in that, The road includes an import / export section corresponding to each yard arranged in the extending direction of the road. The import / export section is located on one side of each yard in the extending direction. The vehicle-mounted unit includes: A receiving module that receives the initial vehicle position information of the vehicle and the digital map of the yard area. The initial vehicle position information is obtained through a filter system of vehicle multi-sensor fusion. The digital map at least includes the default information of each yard, the default information of a container located at the position closest to the import / export section of each yard, and the geographical information of each road. A determination module that determines whether the vehicle is located in the import / export section corresponding to a yard according to the initial vehicle position information of the vehicle received by the receiving module. And A positioning module. When the determination module determines that the vehicle is located in the import / export section corresponding to a yard, the positioning module positions the vehicle based at least on the initial vehicle position information of the vehicle, the real-time information and the pre-stored default information of a container located at the position closest to the import section of the yard.

7. The in-vehicle unit according to claim 6, characterized in that, It further includes: A matching module. When the determination module determines that the vehicle is located in the import / export section corresponding to the yard, the matching module matches the real-time information of a container closest to the import / export section with the default information. And A correction module that corrects the initial vehicle position information according to the matching result of the matching module. The positioning module positions the vehicle according to the correction result of the correction module.

8. The vehicle-mounted unit according to claim 7, characterized in that, The vehicle-mounted unit further includes an update module with the following functions: Determine whether the size of the container located at the position has changed according to the matching result; and When it is determined that the size of the container located at the position has changed, replace the default information with the real-time information to update the digital map.

9. The vehicle-mounted unit according to claim 8, characterized in that, The vehicle-mounted unit further includes a sending module that sends the updated digital map.

10. A server is used for positioning a vehicle traveling in a yard area. The yard area includes a plurality of yards arranged in a grid pattern in two directions orthogonal to each other. A road for the vehicle to travel is formed between the adjacent yards in one of the two directions and extends in the other direction of the two directions. Characterized in that, The road includes an import section corresponding to each yard arranged in the extending direction of the road, and the import section is located on one side of each yard in the extending direction. The server includes: a storage unit that stores a digital map of the yard area, the digital map at least including default information of each yard, default information of a container located at the position closest to the import section of each yard, and geographical information of each road; and a communication module that is configured to transmit the digital map stored in the storage unit, and is configured to receive information and / or requests from the outside.

11. A vehicle positioning system, characterized in that the vehicle positioning system includes the on-vehicle unit according to any one of claims 6 to 9 and the server according to claim 10.

12. A computer-readable storage medium storing a computer program, characterized in that the computer program is executed by a processor to implement the vehicle positioning method according to any one of claims 1 to 5.

13. A computer program product including a computer program, characterized in that the computer program implements the vehicle positioning method according to any one of claims 1 to 5 when executed by a processor.

Images