Method and apparatus for generating global map vector fragments

Abstract

The present application relates to the technical field of intelligent driving. Disclosed are a method and apparatus for generating global map vector fragments. In the present application, the method comprises: on the basis of a plurality of global poses included in global pose data and their corresponding global times, performing pose matching on a plurality of local map vector fragments in local map data, so as to obtain respective pose matching pair sequences of the plurality of local map vector fragments; then, on the basis of the obtained plurality of pose matching pair sequences, determining target transformation matrices respectively corresponding to the plurality of local map vector fragments; and finally, on the basis of the plurality of local map vector fragments and their corresponding target transformation matrices, obtaining global map vector fragments respectively corresponding to the plurality of local map vector fragments. In this way, global map vector fragments are generated by using information in local map vector fragments and information between the local map vector fragments (i.e. target transformation matrices), thereby improving the efficiency and accuracy of the generation of global map vector fragments.

Description

Method and device for generating global map vector segments Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a method and device for generating global map vector segments. Background Art

[0002] As one of the key supports for achieving autonomous driving, high-precision maps (HD Maps) can provide autonomous vehicles with lane-level planning and vehicle positioning assistance within road sections.

[0003] Summary of the Invention

[0004] The embodiments of the present application provide a method and apparatus for generating global map vector segments.

[0005] In a first aspect, an embodiment of the present application provides a method for generating a global map vector fragment, the method comprising:

[0006] Obtaining global pose data and local map data of the vehicle in the target scene; wherein the global pose data includes: multiple global poses of the vehicle in a standard coordinate reference system; and the local map data includes: multiple local map vector segments, each local map vector segment includes: at least one local pose of the vehicle in a local coordinate reference system;

[0007] Based on the multiple global poses and their corresponding global times, pose matching is performed on the multiple local map vector fragments to obtain pose matching pair sequences of the multiple local map vector fragments;

[0008] Based on the obtained multiple pose matching pair sequences, determine the target transformation matrix corresponding to each of the multiple local map vector segments; wherein, for each target transformation matrix, the target transformation matrix is ​​used to ensure that: the pose coincidence between at least one candidate global pose corresponding to at least one local pose in the local map vector segment corresponding to the target transformation matrix and at least one target global pose corresponding to the at least one local pose among the multiple global poses meets a set pose coincidence requirement, and the at least one candidate global pose is obtained by converting at least one local pose in the local map vector segment corresponding to the target transformation matrix into the standard coordinate reference system;

[0009] Based on the multiple local map vector segments and their corresponding target transformation matrices, global map vector segments corresponding to the multiple local map vector segments are obtained.

[0010] In a second aspect, an embodiment of the present application further provides a device for generating a global map vector fragment, the device comprising:

[0011] an acquisition module, configured to acquire global pose data and local map data of the vehicle in a target scene; wherein the global pose data includes: a plurality of global poses of the vehicle in a standard coordinate reference system; and the local map data includes: a plurality of local map vector segments, each local map vector segment including: at least one local pose of the vehicle in a local coordinate reference system;

[0012] A matching module is used to perform pose matching on the multiple local map vector segments based on the multiple global poses and their corresponding global times, so as to obtain a pose matching pair sequence of the multiple local map vector segments;

[0013] a determination module, configured to determine, based on the obtained plurality of pose matching pair sequences, a target transformation matrix corresponding to each of the plurality of local map vector segments; for each target transformation matrix, the target transformation matrix is ​​configured to ensure that: a pose coincidence between at least one candidate global pose corresponding to at least one local pose in the local map vector segment corresponding to the target transformation matrix and at least one target global pose corresponding to the at least one local pose among the plurality of global poses meets a set pose coincidence requirement, and the at least one candidate global pose is obtained by converting the at least one local pose in the local map vector segment corresponding to the target transformation matrix into the standard coordinate reference system;

[0014] The generating module is configured to obtain global map vector segments corresponding to the plurality of local map vector segments based on the plurality of local map vector segments and their corresponding target transformation matrices.

[0015] In an optional embodiment, when performing pose matching on multiple local map vector segments based on multiple global poses and their corresponding global times, the matching module is specifically configured to:

[0016] For multiple local map vector fragments, perform the following operations:

[0017] Obtain at least one local pose in the local map vector segment and the local time corresponding to each pose;

[0018] For each local time, in response to determining that there is a global time identical to the local time among the multiple global times, the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

[0019] In an optional embodiment, the matching module is further configured to:

[0020] In response to determining that none of the multiple global times is the same as the local time, acquiring two global times adjacent to the local time from the multiple global times;

[0021] Based on the two global times and their respective corresponding global poses, and the local time, the global pose corresponding to the local time is obtained, and the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

[0022] In an optional embodiment, when determining the target transformation matrices corresponding to the plurality of local map vector segments based on the obtained plurality of pose matching pair sequences, the determination module is specifically configured to:

[0023] For each pose matching pair sequence in the multiple pose matching pair sequences, perform the following operations:

[0024] Obtain the next pose matching pair sequence adjacent to the local time corresponding to the end of the sequence in the pose matching pair sequence;

[0025] Based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, obtaining the initial joint pose overlap corresponding to the pose matching pair sequence and the next pose matching pair sequence;

[0026] Iteratively modifying the first transformation matrix and the second transformation matrix multiple times based on gradient information of multiple degrees of freedom postures corresponding to the initial joint posture coincidence until a target joint posture coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix satisfies a preset joint posture coincidence condition;

[0027] The modified first transformation matrix and the modified second transformation matrix are used as the target transformation matrices corresponding to the pose matching pair sequence and the next pose matching pair sequence, respectively.

[0028] In an optional embodiment, when obtaining the initial joint pose overlap corresponding to the pose matching pair sequence and the next pose matching pair sequence based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, the determination module is specifically configured to:

[0029] Based on each pose matching pair included in the pose matching pair sequence and the first transformation matrix, obtaining a first sequence pose coincidence degree, and based on each pose matching pair included in the next pose matching pair sequence and the second transformation matrix, obtaining a second sequence pose coincidence degree;

[0030] Obtaining a sub-initial joint pose overlap based on the sequence-end local pose in the pose matching pair sequence, the sequence-start local pose in the next pose matching pair sequence, and the first transformation matrix and the second transformation matrix;

[0031] The initial joint pose coincidence is obtained based on the first sequence pose coincidence, the second sequence pose coincidence and the sub-initial joint pose coincidence.

[0032] In an optional embodiment, if the following conditions are met, determining the target joint pose coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix satisfies a preset joint pose coincidence condition:

[0033] Obtaining the modified first transformation matrix and the modified second transformation matrix, and the corresponding multiple degrees of freedom postures;

[0034] If multiple degree-of-freedom postures all belong to the posture intervals associated with the multiple degree-of-freedom postures, it is determined that the target joint posture coincidence satisfies the joint posture coincidence condition; wherein each posture interval is set according to the gradient information of the corresponding degree-of-freedom posture.

[0035] In a third aspect, an electronic device is proposed, comprising a processor and a memory, wherein the memory stores program code, and when the program code is executed by the processor, the processor executes the steps of the method for generating global map vector fragments described in the first aspect above.

[0036] In a fourth aspect, a computer-readable storage medium is proposed, which includes a program code. When the program code is run on an electronic device, the program code is used to enable the electronic device to execute the steps of the method for generating global map vector fragments described in the first aspect.

[0037] In a fifth aspect, a computer program product is provided. When the computer program product is called by a computer, the computer is caused to execute the steps of the method for generating global map vector fragments as described in the first aspect.

[0038] The beneficial effects of this application are as follows:

[0039] In the method for generating global map vector fragments provided in an embodiment of the present application, pose matching is performed on multiple local map vector fragments in the local map data based on multiple global poses contained in the global pose data and their respective corresponding global times, so as to obtain pose matching pair sequences of the multiple local map vector fragments; then, based on the obtained multiple pose matching pair sequences, the target transformation matrices corresponding to the multiple local map vector fragments are determined; finally, based on the multiple local map vector fragments and their respective corresponding target transformation matrices, the global map vector fragments corresponding to the multiple local map vector fragments are obtained.

[0040] This approach utilizes information within local map vector segments and information between local map vector segments (i.e., the target transformation matrix) to generate global map vector segments. This avoids the technical drawbacks of related technologies that optimize vehicle perception results based on coordinate points, resulting in a large amount of data involved in the calculation, low optimization efficiency, and an inability to ensure the accuracy of the generated global map vector segments due to world coordinate errors. Consequently, the efficiency and accuracy of generating global map vector segments are improved.

[0041] In addition, other features and advantages of the present application will be described in the following description, and in part will become apparent from the description, or may be understood by practicing the present application. The objectives and other advantages of the present application can be realized and obtained through the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without inventive work. In the drawings:

[0043] FIG1 is a schematic diagram of an optional system architecture applicable to an embodiment of the present application;

[0044] FIG2 is a schematic diagram of an implementation flow of a method for generating a global map vector segment provided in an embodiment of the present application;

[0045] FIG3 is a schematic diagram of a specific application scenario for matching a local pose and a global pose provided by an embodiment of the present application;

[0046] FIG4 is a schematic diagram of a specific application scenario for matching a local pose and a global pose provided by an embodiment of the present application;

[0047] FIG5 is a schematic diagram of a specific application scenario based on FIG3 and FIG4 provided in an embodiment of the present application;

[0048] FIG6 is a schematic diagram of an implementation flow of a method for obtaining a target change matrix provided in an embodiment of the present application;

[0049] FIG7 is a logic diagram of obtaining the initial joint posture overlap provided by an embodiment of the present application;

[0050] FIG8 is a schematic diagram of a specific scenario for constructing an optimized edge according to an embodiment of the present application;

[0051] FIG9 is a logic diagram based on FIG2 provided in an embodiment of the present application;

[0052] FIG10 is a schematic structural diagram of a device for generating a global map vector segment according to an embodiment of the present application;

[0053] FIG11 is a schematic structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of the technical solutions of this application, but not all of them. Based on the embodiments described in this application document, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the technical solutions of this application.

[0055] It should be noted that in the description of this application, "multiple" is understood to mean "at least two." "And / or" describes the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. A and B are connected to each other, which can mean: A and B are directly connected, and A and B are connected through C. In addition, in the description of this application, words such as "first" and "second" are used only for the purpose of distinguishing descriptions and should not be understood as indicating or implying relative importance or order.

[0056] With the rapid development of intelligent driving technology, high-precision maps, as the key to determining whether intelligent driving technology can be implemented, have become an important research direction in the field of intelligent driving.

[0057] The current mainstream method for generating high-precision maps is to use timestamps to obtain the world coordinates corresponding to the vehicle perception results (i.e., local perception results), and then generate the perception map results in the world coordinate system (i.e., global map vector fragments) based on the vehicle perception results and their corresponding world coordinates. Then, using a preset fusion algorithm, the multiple perception map results obtained are fused to generate the corresponding high-precision map.

[0058] However, the aforementioned method for generating global map vector fragments optimizes vehicle perception results based on coordinate points, resulting in a large amount of data involved in the calculation and low optimization efficiency. Furthermore, since mapping is performed using world coordinates and vehicle perception results, there is a high probability of introducing world coordinate errors. Therefore, under the influence of world coordinate errors, the accuracy of the generated global map vector fragments cannot be guaranteed.

[0059] In view of this, in an embodiment of the present application, in order to improve the efficiency of generating global map vector segments and, to a certain extent, to improve the accuracy of the generated global map vector segments, a method for generating global map vector segments is proposed, which specifically includes: obtaining global pose data and local map data of a vehicle in a target scene; wherein the global pose data includes: multiple global poses of the vehicle in a standard coordinate reference system, and the local map data includes: multiple local map vector segments, each local map vector segment includes: at least one local pose of the vehicle in a local coordinate reference system; then, based on the multiple global poses and their respective corresponding global times, pose matching is performed on the multiple local map vector segments to obtain pose matching pair sequences of the multiple local map vector segments; further, based on the obtained multiple A pose matching pair sequence is prepared to determine a target transformation matrix corresponding to each of the multiple local map vector segments; wherein, for each target transformation matrix, the target transformation matrix is ​​used to ensure that: a pose coincidence between at least one candidate global pose corresponding to at least one local pose in the local map vector segment corresponding to the target transformation matrix and at least one target global pose corresponding to the at least one local pose among the multiple global poses meets a set pose coincidence requirement, and the at least one candidate global pose is obtained by converting at least one local pose in the local map vector segment corresponding to the target transformation matrix into the standard coordinate reference system; finally, based on the multiple local map vector segments and their corresponding target transformation matrices, the global map vector segments corresponding to the multiple local map vector segments are obtained.

[0060] In particular, the following describes the preferred embodiments of the present application in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application and are not intended to limit the present application. Furthermore, the embodiments of the present application and the features therein may be combined with each other unless there is a conflict.

[0061] Referring to FIG. 1 , which is a schematic diagram of a system architecture applicable to an embodiment of the present application, the system architecture includes a target terminal 101 and a server 102. The target terminal 101 and the server 102 can exchange information via a communication network, wherein the communication network may employ wireless communication and wired communication.

[0062] Exemplarily, the target terminal 101 can access the network through cellular mobile communication technology and communicate with the server 102, wherein the cellular mobile communication technology includes, for example, the fifth generation mobile communication (5th Generation Mobile Networks, 5G) technology.

[0063] Optionally, the target terminal 101 may access the network and communicate with the server 102 via short-range wireless communication, wherein the short-range wireless communication may include, for example, Wireless Fidelity (Wi-Fi) technology.

[0064] The embodiments of the present application do not impose any restrictions on the number of communication devices involved in the above system architecture. For example, the system architecture may include multiple target terminals, or no target terminal, or may also include other network devices. As shown in Figure 1, only the target terminal 101 and the server 102 are used as examples for description. The following briefly introduces each of the above devices and their respective functions.

[0065] The target terminal 101 is a device that can provide voice and / or data connectivity to the user, and can be a device that supports wired and / or wireless connection.

[0066] Exemplarily, the target terminal 101 includes but is not limited to: mobile phones, tablet computers, laptop computers, PDAs, mobile Internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminal devices in industrial control, wireless terminal devices in unmanned driving, wireless terminal devices in smart grids, wireless terminal devices in transportation safety, wireless terminal devices in smart cities, or wireless terminal devices in smart homes, etc.

[0067] In addition, the target terminal 101 may be installed with a relevant client, which may be software, such as an application (APP), a browser, a short video software, etc., or a web page, a mini-program, etc. It should be noted that in the embodiment of the present application, the target terminal 101 may be the above-mentioned client related to the generation of the global map vector fragment, sending the global pose data and local map data of the vehicle in the target scene to the server 102 for subsequent method steps such as the generation of the global map vector fragment.

[0068] Server 102 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0069] It is worth mentioning that in an embodiment of the present application, the server 102 is used to obtain the global pose data and local map data of the vehicle in the target scene; then, based on multiple global poses and their respective corresponding global times, pose matching is performed on multiple local map vector fragments to obtain pose matching pair sequences of each of the multiple local map vector fragments; further, based on the obtained multiple pose matching pair sequences, the target transformation matrix corresponding to each of the multiple local map vector fragments is determined; finally, based on the multiple local map vector fragments and their respective corresponding target transformation matrices, the global map vector fragments corresponding to each of the multiple local map vector fragments are obtained.

[0070] The following describes the method for generating global map vector fragments provided by an exemplary embodiment of the present application in combination with the above-mentioned system architecture and with reference to the accompanying drawings. It should be noted that the above-mentioned system architecture is only shown to facilitate understanding of the spirit and principles of the present application, and the implementation of the present application is not limited in this respect.

[0071] Refer to FIG. 2 , which is a schematic diagram of an implementation flow of a method for generating a global map vector segment provided in an embodiment of the present application. The execution entity is a server as an example. The specific implementation flow of the method includes the following steps S201 - S204 .

[0072] S201: Obtain global pose data and local map data of the vehicle in the target scene.

[0073] Among them, the global pose data includes: multiple global poses of the vehicle in the standard coordinate reference system; the local map data includes: multiple local map vector fragments; each local map vector fragment includes: at least one local pose of the vehicle in the local coordinate reference system; it should be noted that the above-mentioned standard coordinate reference system is the world coordinate system.

[0074] It should also be noted that the above-mentioned global pose data can be data obtained through the vehicle-side positioning algorithm, which includes global time (i.e., global timestamp / global time identifier) ​​and global pose (i.e., global positioning pose); the above-mentioned local map vector fragments can be local map information of fixed mileage length including local time (i.e., local timestamp / local time identifier), trajectory (i.e., local pose / local positioning pose) and various traffic elements obtained through the vehicle-side reconstruction algorithm.

[0075] Exemplarily, the above-mentioned types of traffic elements include but are not limited to: target objects, vehicles and roads; among them, target objects include: drivers, pedestrians, passengers, etc.; vehicles include: motor vehicles and non-motor vehicles, etc.; roads include: highways, urban roads, entrance and exit roads and their related facilities and environment, off-road landscapes, management facilities and climatic conditions, etc.

[0076] Optionally, the specific forms of global pose and local pose can be expressed as follows:

[0077] Among them, R is in the form of: Indicates rotation information, and Represents the rotation vectors on the X-axis, Y-axis, and Z-axis respectively; the form of h is: Represents translation information, where h0, h1, and h2 represent the offsets on the X, Y, and Z axes, respectively.

[0078] S202: Based on the multiple global poses and their corresponding global times, pose matching is performed on the multiple local map vector segments to obtain pose matching pair sequences of the multiple local map vector segments.

[0079] In an optional implementation, when executing step S202, after obtaining the global pose data and local map data of the vehicle in the target scene, the server can perform the following operations for any local map vector fragment among multiple local map vector fragments: obtain at least one local pose in the local map vector fragment and the local time corresponding to each local time; for each local time, if there is a global time that is the same as the local time among multiple global times, then the global pose and local pose corresponding to the local time are regarded as a pose matching pair; in this way, all global poses and local poses of the same time (global time is the same as local time) can be successfully paired.

[0080] For example, referring to FIG3 , assuming that the local time corresponding to a local pose (e.g., Local.Pose1) in the above-mentioned local map vector fragment is: 2023.09.06 14:20:58, and there is also a global pose (e.g., Global.Pose1 or World.Pose1) with a global time of 2023.09.06 14:20:58 in multiple global poses, then the server may consider that at least one of the obtained local times has the same local time as the time in the multiple global times (e.g., 2023.09.06 14:20:58), and therefore, the local pose Local.Pose1 and the global pose Global.Pose1 can be regarded as a pose matching pair.

[0081] Optionally, for each local time, if there is no global time identical to the local time among multiple global times, two global times adjacent to the local time are obtained from the multiple global times, and the global pose corresponding to the local time is obtained based on the two global times and their respective corresponding global poses, as well as the local time, and the global pose corresponding to the local time and the local pose are taken as a pose matching pair; in this way, it is ensured that all local poses can be matched to the global pose.

[0082] For example, referring to FIG4 , it is assumed that the local time corresponding to a local pose (e.g., Local.Pose2) in the above-mentioned local map vector fragment is: 2023.09.06 14:24:37, and among multiple global poses, there are only global poses (e.g., Global.Pose2) with a global time of 2023.09.06 14:24:18 and 2023.09.06 14:24:52 (e.g., Global.Pose3) adjacent to the local time. Therefore, the server determines that in at least one local time, there is a local time that is different from multiple global times (e.g., 2023.09.06 14:24:37), and obtains two global times (i.e., 2023.09.06 14:24:18 and 2023.09.06 14:24:52) adjacent to the local time (i.e., 2023.09.06 14:24:37). Based on the two global times and their respective corresponding global poses (i.e., Global.Pose2 and Global.Pose3), as well as the local time, the server obtains the global pose corresponding to the local time, and treats the global pose and the local pose corresponding to the local time as a pose matching pair.

[0083] Specifically, the global pose corresponding to the above local time can be calculated as follows:

[0084] Among them, T represents the pose information obtained by interpolating the global pose according to time (i.e., global time and local time), i.e., local time The corresponding global pose, and For the local time Two adjacent global times, later than Represents global time The corresponding local pose, Represents global time The corresponding local pose.

[0085] Based on the above method, as shown in Figure 5, even if the frequencies (of the clocks of the global time and the local time) are different, the timestamps of the two (i.e., the global time and the local time) cannot completely correspond, that is, there is a global time that does not correspond to the local time, the global pose corresponding to the pose at each moment (i.e., each local pose) in the local map vector fragment can be obtained, that is, the local pose in the local map vector fragment and the global pose in the global positioning data can be associated through the timestamp (i.e., the global time and the local time).

[0086] It should also be noted that at least one pose matching pair in each of the above pose matching pair sequences is arranged in order of local time, that is, the earlier the local time, the closer it is to the beginning of the pose matching pair sequence, and conversely, the later the local time, the closer it is to the end of the pose matching pair sequence.

[0087] S203: Determine target transformation matrices corresponding to the plurality of local map vector segments based on the obtained plurality of pose matching pair sequences.

[0088] Among them, for each target transformation matrix, the target transformation matrix is ​​used to ensure that: the posture coincidence between at least one candidate global posture corresponding to at least one local posture in the local map vector segment corresponding to the target transformation matrix and at least one target global posture corresponding to the at least one local posture among the multiple global postures meets the set posture coincidence requirement, and the at least one candidate global posture is obtained by converting at least one local posture in the local map vector segment corresponding to the target transformation matrix into the standard coordinate reference system; exemplarily, the above-mentioned set posture coincidence requirement can be: after the above-mentioned at least one local posture is converted into the global posture under the standard coordinate reference system, the posture coincidence between the at least one global posture corresponding to at least one local posture among the multiple global postures is greater than the set posture coincidence threshold.

[0089] In an optional implementation, referring to FIG. 6 , when executing step S203 , after obtaining multiple pose matching pair sequences, the server may perform the following operations S601 - A604 for any one of the multiple pose matching pair sequences.

[0090] S601: Obtain the pose matching pair sequence and the next pose matching pair sequence that is adjacent to the pose matching pair sequence in local time.

[0091] Specifically, when executing step S601, after the server obtains the posture matching pair sequence, it can select the next posture matching pair sequence adjacent to the local time corresponding to the posture matching pair sequence from multiple posture matching pair sequences based on the local time range corresponding to the posture matching pair sequence.

[0092] S602: Based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, obtain the initial joint pose overlap corresponding to the pose matching pair sequence and the next pose matching pair sequence.

[0093] In an optional implementation, referring to Figure 7, when executing step S602, after obtaining the posture matching pair sequence and the next posture matching pair sequence, the server can obtain the first sequence posture coincidence based on the posture matching pairs contained in the posture matching pair sequence and the first transformation matrix, and obtain the second sequence posture coincidence based on the posture matching pairs contained in the next posture matching pair sequence and the second transformation matrix; and obtain the sub-initial joint posture coincidence based on the sequence tail local posture in the posture matching pair sequence, the sequence head local posture in the next posture matching pair sequence, and the first transformation matrix and the second transformation matrix; finally, obtain the initial joint posture coincidence based on the first sequence posture coincidence, the second sequence posture coincidence and the sub-initial joint posture coincidence.

[0094] Optionally, the calculation formula for the initial joint pose overlap can be expressed as follows:

[0095] Among them, Error represents the initial joint pose overlap, represents the first sequence pose overlap, represents the overlap of the second sequence poses, represents the overlap of the initial joint poses, ΔT0 represents the first transformation matrix, Indicates the i-th pose matching pair in the sequence of n pose matching pairs (i.e., at t i The local pose of the pose matching pair at the moment, Indicates the i-th pose matching pair in the sequence (i.e. at t i The global pose of the pose matching pair at the moment, inv() represents the inverse operation of the matrix, for example, for The inverse matrix of , ΔT1 represents the second transformation matrix, Indicates the jth pose matching pair in the next pose matching pair sequence containing (mn) pose matching pairs (i.e., at t j The local pose of the pose matching pair at the moment, Indicates the jth pose in the next pose matching sequence (i.e., at t j The global pose of the pose matching pair at the moment, Indicates the nth pose matching pair in the sequence (i.e. at t n-1The local pose of the pose matching pair at the moment, that is, the local pose at the end of the sequence in the pose matching pair sequence, Indicates the first one in the next pose matching sequence (i.e. at t n The local pose of the pose matching pair at the moment, that is, the first local pose in the next pose matching pair sequence.

[0096] It should be noted that the above-mentioned first transformation matrix ΔT0 and second transformation matrix ΔT1 are in the same form as the matrix forms of the global pose and the local pose, and can be initially set to the unit matrix; and the product of the above-mentioned transformation matrix (first transformation matrix / second transformation matrix) and the local pose, and then multiplied by the inverse matrix of the global pose corresponding to the local pose, can measure the pose coincidence between the corresponding local pose after being transformed into the standard coordinate reference system through the transformation matrix and the global pose corresponding to the local pose.

[0097] Based on the method for obtaining the initial joint pose coincidence recorded in S602 above, as shown in FIG8 , after the server obtains the pose matching pair sequence and the next pose matching pair sequence through matching, it can construct the optimized edge between the pose matching pairs. And through the pose relationship between the local map vector segments, it can construct the optimized edge between the local map vector segments. It should be noted that the above-mentioned method for obtaining the initial joint pose coincidence is also the optimization function constructed according to the optimization graph shown in FIG8 . Therefore, the above-mentioned first sequence pose coincidence can be regarded as a measure of the optimized edge in the pose matching sequence, the above-mentioned second sequence pose coincidence can be regarded as a measure of the optimized edge in the next pose matching sequence, and the above-mentioned sub-initial joint pose coincidence can be regarded as a measure of the optimized edge between the sequence-end local pose in the pose matching pair sequence and the sequence-first local pose in the next pose matching pair sequence.

[0098] It should also be noted that the function output value of the above optimization function can be: a 1×6 matrix; where 6 is the number of degrees of freedom of the posture.

[0099] S603: Based on the gradient information of multiple degrees of freedom postures corresponding to the initial joint posture coincidence, the first transformation matrix and the second transformation matrix are iteratively modified multiple times until the target joint posture coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix meets the preset joint posture coincidence condition.

[0100] Among them, the gradient information of the above-mentioned multiple degrees of freedom posture is the gradient information of 6 degrees of freedom related to the posture, and the 6 degrees of freedom include 3 degrees of freedom of displacement (i.e., translation) and 3 degrees of freedom of spatial rotation, i.e., displacement and spatial rotation in the three directions of X-axis, Y-axis and Z-axis.

[0101] In an optional implementation, during the process of iteratively modifying the first transformation matrix and the second transformation matrix multiple times based on the gradient information of multiple degree-of-freedom postures corresponding to the initial joint posture overlap, if the multiple degree-of-freedom postures all belong to the posture intervals associated with the multiple degree-of-freedom postures, it can be determined that the target joint posture overlap meets the joint posture overlap condition; wherein each posture interval is set according to the gradient information of the corresponding degree-of-freedom posture.

[0102] S604: Using the modified first transformation matrix and the modified second transformation matrix as target transformation matrices corresponding to the pose matching pair sequence and the next pose matching pair sequence, respectively.

[0103] Based on the method steps recorded in S603 to S604 above, after obtaining the optimization function, the optimization function is differentiated to obtain the derivative function of the optimization function with respect to each degree of freedom in the pose information (i.e., the gradient information of multiple degree-of-freedom poses), and then the optimization function is iteratively optimized using a graph optimization (e.g., General Graph Optimization (G2O)) library to obtain the final transformation matrix (i.e., the target transformation matrix) between the local map vector fragment and the next local map vector fragment.

[0104] S204: Based on the multiple local map vector segments and their corresponding target transformation matrices, obtain the global map vector segments corresponding to the multiple local map vector segments.

[0105] Specifically, when executing step S204, after determining the target transformation matrices corresponding to the multiple local map vector segments, the server can obtain the global map vector segments corresponding to the multiple local map vector segments based on the multiple local map vector segments and their corresponding target transformation matrices. Exemplarily, the transformation relationship between the local map vector segments and the global map vector segments can be expressed as follows: ins =ΔT*T ndm

[0106] Among them, ΔT represents the target transformation matrix corresponding to the local map vector fragment, T ndm Represents a local map vector segment, T ins Represents a local map vector segment T ndm The corresponding global map vector fragment.

[0107] For example, the server can update the trajectory and traffic element targets in the local map vector segment according to the obtained target transformation matrix. The specific update formula is as follows: world =ΔT*T local P world =ΔT*P local

[0108] Among them, ΔT represents the target transformation matrix corresponding to the local map vector fragment, T local Represents the pose information corresponding to the trajectory point in the local map vector segment, T world Represents the pose information T corresponding to the trajectory point in the local map vector segment local The pose information corresponding to the trajectory point in the corresponding global map vector segment, P local Represents the coordinate information of the points that constitute the traffic elements in the local map vector segment, P world Represents the coordinate information P of the points that constitute the traffic elements in the local map vector segment local The coordinate information of the points that constitute the traffic elements in the corresponding global map vector fragment.

[0109] Based on the steps of the method for generating global map vector fragments recorded in S201 to S204 above, as shown in Figure 9, the server can obtain pose matching pairs by timestamp interpolation and combining local map data and global pose data (i.e., global positioning pose), and construct an optimization graph based on this; then, in the process of iterative optimization of the optimization graph, a transformation matrix (i.e., target transformation matrix) that can describe the coincidence of local pose to global pose is generated to achieve the fusion of two types of information (global information and local information), i.e., data update, thereby generating high-precision map global vector fragments (i.e., vector information).

[0110] It should also be noted that the above-mentioned method, which performs optimization based on the information of local map vector segments, reduces the optimization calculation amount of the target (i.e., point, such as a traffic element) and the trajectory (i.e., line), thereby improving the optimization efficiency; and, by utilizing the relationship between the local map vector segments, reduces the introduction of errors in the standard coordinate reference system (i.e., world coordinates).

[0111] In summary, in the method for generating global map vector fragments provided in the embodiment of the present application, the global pose data and local map data of the vehicle in the target scene are obtained; then, based on the multiple global poses contained in the global pose data and their respective corresponding global times, pose matching is performed on the multiple local map vector fragments in the local map data to obtain pose matching pair sequences of the multiple local map vector fragments; further, based on the obtained multiple pose matching pair sequences, the target transformation matrix corresponding to the multiple local map vector fragments is determined; finally, based on the multiple local map vector fragments and their respective corresponding target transformation matrices, the global map vector fragments corresponding to the multiple local map vector fragments are obtained.

[0112] This approach utilizes information within local map vector segments and information between local map vector segments (i.e., the target transformation matrix) to generate global map vector segments. This avoids the technical drawbacks of related technologies that optimize vehicle perception results based on coordinate points, resulting in a large amount of data involved in the calculation, low optimization efficiency, and an inability to ensure the accuracy of the generated global map vector segments due to world coordinate errors. Consequently, the efficiency and accuracy of generating global map vector segments are improved.

[0113] Furthermore, based on the same technical concept, the present embodiment also provides a device for generating global map vector segments. The device is used to implement the above-mentioned method for generating global map vector segments in the present embodiment. Referring to FIG10 , the device for generating global map vector segments includes: an acquisition module 1001, a matching module 1002, a determination module 1003, and a generation module 1004, wherein:

[0114] Acquisition module 1001 is configured to acquire global pose data and local map data of a vehicle in a target scene; wherein the global pose data includes: multiple global poses of the vehicle in a standard coordinate reference system; and the local map data includes: multiple local map vector segments, each local map vector segment including: at least one local pose of the vehicle in a local coordinate reference system;

[0115] A matching module 1002 is configured to perform pose matching on a plurality of local map vector segments based on a plurality of global poses and their corresponding global times, to obtain a pose matching pair sequence for each of the plurality of local map vector segments;

[0116] Determination module 1003 is configured to determine target transformation matrices corresponding to each of the plurality of local map vector segments based on the obtained plurality of pose matching pair sequences; wherein, for each target transformation matrix, the target transformation matrix is ​​configured to ensure that: a pose coincidence between at least one candidate global pose corresponding to at least one local pose in the local map vector segment corresponding to the target transformation matrix and at least one target global pose corresponding to the at least one local pose among the plurality of global poses meets a set pose coincidence requirement, and the at least one candidate global pose is obtained by converting the at least one local pose in the local map vector segment corresponding to the target transformation matrix into the standard coordinate reference system;

[0117] The generating module 1004 is configured to obtain global map vector segments corresponding to the plurality of local map vector segments based on the plurality of local map vector segments and their corresponding target transformation matrices.

[0118] In an optional embodiment, when performing pose matching on multiple local map vector segments based on multiple global poses and their corresponding global times, the matching module 1002 is specifically configured to:

[0119] For multiple local map vector fragments, perform the following operations:

[0120] Obtain at least one local pose in the local map vector segment and the local time corresponding to each pose;

[0121] For each local time, in response to determining that there is a global time identical to the local time among the multiple global times, the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

[0122] In an optional embodiment, the matching module 1002 is further configured to:

[0123] In response to determining that none of the multiple global times is the same as the local time, acquiring two global times adjacent to the local time from the multiple global times;

[0124] Based on the two global times and their respective corresponding global poses, and the local time, the global pose corresponding to the local time is obtained, and the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

[0125] In an optional embodiment, when determining the target transformation matrices corresponding to the plurality of local map vector segments based on the obtained plurality of pose matching pair sequences, the determination module 1003 is specifically configured to:

[0126] For each pose matching pair sequence in the multiple pose matching pair sequences, perform the following operations:

[0127] Obtain the next pose matching pair sequence adjacent to the local time corresponding to the end of the sequence in the pose matching pair sequence;

[0128] Based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, obtaining the initial joint pose overlap corresponding to the pose matching pair sequence and the next pose matching pair sequence;

[0129] Iteratively modifying the first transformation matrix and the second transformation matrix multiple times based on gradient information of multiple degrees of freedom postures corresponding to the initial joint posture coincidence until a target joint posture coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix satisfies a preset joint posture coincidence condition;

[0130] The modified first transformation matrix and the modified second transformation matrix are used as the target transformation matrices corresponding to the pose matching pair sequence and the next pose matching pair sequence, respectively.

[0131] In an optional embodiment, when obtaining the initial joint pose overlap corresponding to the pose matching pair sequence and the next pose matching pair sequence based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, the determination module 1003 is specifically used to:

[0132] Based on each pose matching pair included in the pose matching pair sequence and the first transformation matrix, obtaining a first sequence pose coincidence degree, and based on each pose matching pair included in the next pose matching pair sequence and the second transformation matrix, obtaining a second sequence pose coincidence degree;

[0133] Obtaining a sub-initial joint pose overlap based on the sequence-end local pose in the pose matching pair sequence, the sequence-start local pose in the next pose matching pair sequence, and the first transformation matrix and the second transformation matrix;

[0134] The initial joint pose coincidence is obtained based on the first sequence pose coincidence, the second sequence pose coincidence and the sub-initial joint pose coincidence.

[0135] In an optional embodiment, if the following conditions are met, determining the target joint pose coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix satisfies a preset joint pose coincidence condition:

[0136] Obtaining the modified first transformation matrix and the modified second transformation matrix, and the corresponding multiple degrees of freedom postures;

[0137] If multiple degree-of-freedom postures all belong to the posture intervals associated with the multiple degree-of-freedom postures, it is determined that the target joint posture coincidence satisfies the joint posture coincidence condition; wherein each posture interval is set according to the gradient information of the corresponding degree-of-freedom posture.

[0138] Based on the same technical concept, an embodiment of the present application further provides an electronic device that can implement the method for generating global map vector fragments provided in the above embodiment of the present application. In one embodiment, the electronic device can be a server, a terminal device, or other electronic device. As shown in Figure 11, the electronic device may include:

[0139] At least one processor 1101, and a memory 1102 connected to at least one processor 1101. In the embodiments of the present application, the specific connection medium between the processor 1101 and the memory 1102 is not limited. FIG11 takes the connection between the processor 1101 and the memory 1102 via the bus 1100 as an example. The bus 1100 is represented by a bold line in FIG11, and the connection between other components is only for schematic illustration and is not intended to be limiting. The bus 1100 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, FIG11 only uses a bold line to represent it, but it does not mean that there is only one bus or one type of bus. Alternatively, the processor 1101 can also be called a controller, and there is no restriction on the name.

[0140] In this embodiment of the present application, memory 1102 stores instructions executable by at least one processor 1101. At least one processor 1101 can execute the method for generating a global map vector fragment discussed above by executing the instructions stored in memory 1102. Processor 1101 can implement the functions of each module in the apparatus shown in FIG. 10 .

[0141] Among them, the processor 1101 is the control center of the device, which can use various interfaces and lines to connect the various parts of the entire control device, and monitor the device as a whole by running or executing instructions stored in the memory 1102 and calling data stored in the memory 1102, the various functions of the device and processing data.

[0142] In one possible design, processor 1101 may include one or more processing units. Processor 1101 may integrate an application processor and a modem processor. The application processor primarily processes the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into processor 1101. In some embodiments, processor 1101 and memory 1102 may be implemented on the same chip. In some embodiments, they may also be implemented on separate chips.

[0143] Processor 1101 can be a general-purpose processor, such as a CPU, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of this application. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method for generating a global map vector fragment disclosed in the embodiments of this application can be directly implemented and executed by a hardware processor, or by a combination of hardware and software modules within the processor.

[0144] Memory 1102 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules. Memory 1102 may include at least one type of storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory, a random access memory (Random Access Memory, RAM), a static random access memory (Static Random Access Memory, SRAM), a programmable read-only memory (Programmable Read Only Memory, PROM), a read-only memory (Read Only Memory, ROM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), a magnetic memory, a disk, an optical disk, etc. Memory 1102 is any other medium that can be used to carry or store a desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 1102 in the embodiment of the present application can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0145] By programming processor 1101, the code corresponding to the method for generating a global map vector fragment described in the aforementioned embodiment can be embedded within the chip, enabling the chip to execute the steps of the method for generating a global map vector fragment in the embodiment shown in FIG2 during operation. Designing and programming processor 1101 is well known to those skilled in the art and will not be further described here.

[0146] Based on the same inventive concept, an embodiment of the present application further provides a storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer executes the method for generating a global map vector fragment discussed above.

[0147] In some possible implementations, various aspects of the method for generating a global map vector fragment provided by the present application can also be implemented in the form of a program product, which includes program code. When the program product is run on an apparatus, the program code is used to enable the control device to execute the steps of the method for generating a global map vector fragment according to various exemplary embodiments of the present application described above in this specification.

[0148] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0149] The present application is described with reference to the flow chart and / or block diagram of the method, device (system), and computer program product according to the embodiment of the present application. It should be understood that each flow process and / or box in the flow chart and / or block diagram and the combination of the flow process and / or box in the flow chart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processing machine or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one flow chart flow or multiple flows and / or one box or multiple boxes of the block diagram.

[0150] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0151] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A method for generating a global map vector fragment, comprising: Acquire global pose data and local map data of the vehicle in the target scene; wherein the global pose data includes: a plurality of global poses of the vehicle in a standard coordinate reference system, and the local map data includes: a plurality of local map vector segments, each local map vector segment includes: at least one local pose of the vehicle in a local coordinate reference system; Based on the multiple global poses and their respective corresponding global times, pose matching is performed on the multiple local map vector fragments to obtain pose matching pair sequences of the multiple local map vector fragments; Based on the obtained multiple pose matching pair sequences, determine the target transformation matrix corresponding to each of the multiple local map vector fragments; wherein, for each target transformation matrix, the target transformation matrix is ​​used to ensure that: the pose coincidence between at least one candidate global pose corresponding to at least one local pose in the local map vector fragment corresponding to the target transformation matrix and at least one target global pose corresponding to the at least one local pose among the multiple global poses meets the set pose coincidence requirement, and the at least one candidate global pose is obtained by converting at least one local pose in the local map vector fragment corresponding to the target transformation matrix into the standard coordinate reference system; Based on the multiple local map vector fragments and their respective corresponding target transformation matrices, global map vector fragments corresponding to each of the multiple local map vector fragments are obtained.

2. The method according to claim 1, characterized in that The performing pose matching on the multiple local map vector fragments based on the multiple global poses and their respective corresponding global times includes: For each of the plurality of local map vector fragments, the following operations are performed respectively: Obtain at least one local pose in the local map vector fragment and the local time corresponding to each pose; For each local time, in response to determining that there is a global time identical to the local time among the multiple global times, the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

3. The method according to claim 2, characterized in that The method further comprises: In response to determining that there is no global time identical to the local time among the multiple global times, acquiring two global times adjacent to the local time from the multiple global times; Based on the two global times and their respective corresponding global poses, and the local time, the global pose corresponding to the local time is obtained, and the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

4. The method according to claim 1, characterized in that The step of determining target transformation matrices corresponding to the plurality of local map vector fragments based on the obtained plurality of pose matching pair sequences comprises: For each pose matching pair sequence in the multiple pose matching pair sequences, the following operations are performed respectively: Obtain the next pose matching pair sequence adjacent to the local time corresponding to the end of the sequence in the pose matching pair sequence; Based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, obtaining an initial joint pose overlap degree corresponding to the pose matching pair sequence and the next pose matching pair sequence; Based on the gradient information of the multiple degrees of freedom postures corresponding to the initial joint posture coincidence, the first transformation matrix and the second transformation matrix are iteratively modified multiple times until the target joint posture coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix meets the preset joint posture coincidence condition; The modified first transformation matrix and the modified second transformation matrix are respectively used as target transformation matrices corresponding to the pose matching pair sequence and the next pose matching pair sequence.

5. The method according to claim 4, characterized in that The obtaining, based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, an initial joint pose overlap degree corresponding to the pose matching pair sequence and the next pose matching pair sequence includes: Based on each pose matching pair included in the pose matching pair sequence and the first transformation matrix, a first sequence pose coincidence is obtained, and based on each pose matching pair included in the next pose matching pair sequence and the second transformation matrix, a second sequence pose coincidence is obtained; Obtaining a sub-initial joint pose overlap based on a sequence-end local pose in the pose matching pair sequence, a sequence-start local pose in the next pose matching pair sequence, and the first transformation matrix and the second transformation matrix; The initial joint pose coincidence is obtained based on the first sequence pose coincidence, the second sequence pose coincidence and the sub-initial joint pose coincidence.

6. The method according to claim 4, characterized in that If the following conditions are met, the target joint pose coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix is ​​determined to meet the preset joint pose coincidence condition: Obtaining the modified first transformation matrix and the modified second transformation matrix, and corresponding multiple degrees of freedom postures; If the multiple degrees of freedom postures all belong to the posture intervals associated with the multiple degrees of freedom postures respectively, it is determined that the target joint posture overlap satisfies the joint posture overlap condition; wherein each posture interval is set according to the gradient information of the corresponding degree of freedom posture.

7. A device for generating a global map vector fragment, comprising: An acquisition module is used to acquire global pose data and local map data of a vehicle in a target scene; wherein the global pose data includes: a plurality of global poses of the vehicle in a standard coordinate reference system, and the local map data includes: a plurality of local map vector segments, each of which includes: at least one local pose of the vehicle in a local coordinate reference system; A matching module, configured to perform pose matching on the plurality of local map vector segments based on the plurality of global poses and their respective corresponding global times, to obtain pose matching pair sequences of the plurality of local map vector segments; A determination module, configured to determine a target transformation matrix corresponding to each of the plurality of local map vector segments based on the obtained plurality of posture matching pair sequences; for each target transformation matrix, the target transformation matrix is ​​configured to ensure that: a posture coincidence degree between at least one candidate global posture corresponding to at least one local posture in the local map vector segment corresponding to the target transformation matrix and at least one target global posture corresponding to the at least one local posture among the plurality of global postures meets a set posture coincidence degree requirement, and the at least one candidate global posture is obtained by converting at least one local posture in the local map vector segment corresponding to the target transformation matrix into the standard coordinate reference system; A generating module is used to obtain global map vector segments corresponding to each of the multiple local map vector segments based on the multiple local map vector segments and their corresponding target transformation matrices.

8. The device according to claim 7, characterized in that When performing posture matching on the multiple local map vector fragments based on the multiple global postures and their respective corresponding global times, the matching module is specifically used to: For each of the plurality of local map vector fragments, the following operations are performed respectively: Obtain at least one local pose in the local map vector fragment and the local time corresponding to each pose; For each local time, in response to determining that there is a global time identical to the local time among the multiple global times, the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

9. The device according to claim 8, characterized in that The matching module is also used for: In response to determining that there is no global time identical to the local time among the multiple global times, acquiring two global times adjacent to the local time from the multiple global times; Based on the two global times and their respective corresponding global poses, and the local time, the global pose corresponding to the local time is obtained, and the global pose and the local pose corresponding to the local time are taken as a pose matching pair.

10. The device according to claim 7, characterized in that When determining the target transformation matrices corresponding to the plurality of local map vector fragments based on the obtained plurality of pose matching pair sequences, the determination module is specifically used to: For each pose matching pair sequence in the multiple pose matching pair sequences, the following operations are performed respectively: Obtain the next pose matching pair sequence adjacent to the local time corresponding to the end of the sequence in the pose matching pair sequence; Based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, obtaining an initial joint pose overlap degree corresponding to the pose matching pair sequence and the next pose matching pair sequence; Based on the gradient information of the multiple degrees of freedom postures corresponding to the initial joint posture coincidence, the first transformation matrix and the second transformation matrix are iteratively modified multiple times until the target joint posture coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix meets the preset joint posture coincidence condition; The modified first transformation matrix and the modified second transformation matrix are respectively used as target transformation matrices corresponding to the pose matching pair sequence and the next pose matching pair sequence.

11. The device according to claim 10, characterized in that When obtaining the initial joint pose overlap degree corresponding to the pose matching pair sequence and the next pose matching pair sequence based on the pose matching pair sequence and its associated first transformation matrix, and the next pose matching pair sequence and its associated second transformation matrix, the determination module is specifically used to: Based on each pose matching pair included in the pose matching pair sequence and the first transformation matrix, a first sequence pose coincidence is obtained, and based on each pose matching pair included in the next pose matching pair sequence and the second transformation matrix, a second sequence pose coincidence is obtained; Obtaining a sub-initial joint pose overlap based on a sequence-end local pose in the pose matching pair sequence, a sequence-start local pose in the next pose matching pair sequence, and the first transformation matrix and the second transformation matrix; The initial joint pose coincidence is obtained based on the first sequence pose coincidence, the second sequence pose coincidence and the sub-initial joint pose coincidence.

12. The device according to claim 10, characterized in that If the following conditions are met, the target joint pose coincidence obtained based on the modified first transformation matrix and the modified second transformation matrix is ​​determined to meet the preset joint pose coincidence condition: Obtaining the modified first transformation matrix and the modified second transformation matrix, and corresponding multiple degrees of freedom postures; If the multiple degrees of freedom postures all belong to the posture intervals associated with the multiple degrees of freedom postures respectively, it is determined that the target joint posture overlap satisfies the joint posture overlap condition; wherein each posture interval is set according to the gradient information of the corresponding degree of freedom posture.

13. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

14. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 6.

Images