A positioning processing method and device

Abstract

The present application provides a positioning processing method, device, electronic device and computer-readable storage medium; it relates to artificial intelligence positioning technology, and the method includes: determining the observation position of the positioning target based on the carrier phase; determining the identification position of the positioning target based on the road elements identified from the image of the road environment and the position of the road elements in the map; determining the deviation between the observation position of the positioning target and the identification position of the positioning target based on the observation position of the positioning target and the identification position of the positioning target; correcting the observation position based on the deviation to obtain the updated observation position of the positioning target. Through the present application, the positioning deviation can be eliminated and the positioning accuracy can be improved.

Description

Technical Field

[0001] The present application relates to artificial intelligence positioning technology, and in particular to a positioning processing method, device, electronic device and computer-readable storage medium. Background Art

[0002] With the research and progress of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as autonomous driving. Autonomous driving technology usually includes high-precision maps, environmental perception, behavioral decision-making, path planning, motion control and other technologies. Autonomous driving technology has broad application prospects.

[0003] In the vehicle positioning products for autonomous driving, positioning is achieved through visual fusion of high-precision maps and real-time kinematic (RTK) carrier differential technology. By default, the global position (such as longitude and latitude) of road elements in the high-precision map is consistent with the global observation longitude and latitude of RTK, that is, there is no deviation between the high-precision map and RTK. However, in actual application scenarios, due to factors such as the movement of the earth's crust, the position deviation of the GPS base station, or clock synchronization, there is a certain deviation between the position of the RTK and the position of the elements in the high-precision map, resulting in a decrease in the accuracy of the fusion positioning result, or even making the positioning result unusable. Summary of the invention

[0004] The embodiments of the present application provide a positioning processing method, device, electronic device and computer-readable storage medium, which can eliminate positioning deviation and thus improve positioning accuracy.

[0005] The technical solution of the embodiment of the present application is implemented as follows:

[0006] The present application provides a positioning processing method, including:

[0007] Determining an observed position of the positioning target based on a carrier phase;

[0008] Determining the identified position of the positioning target based on the road elements identified from the image of the road environment and the positions of the road elements in the map;

[0009] Determining a deviation between the observed position of the positioning target and the identified position of the positioning target based on the observed position of the positioning target and the identified position of the positioning target;

[0010] The observed position is corrected based on the deviation to obtain an updated observed position of the positioning target.

[0011] The present application provides a positioning processing device, including:

[0012] An observation module, used to determine the observation position of the positioning target based on the carrier phase;

[0013] A positioning module, used to determine the recognition position of the positioning target based on the road elements recognized from the image of the road environment and the positions of the road elements in the map;

[0014] A positioning and online deviation estimation module, used to determine the deviation between the observed position of the positioning target and the identified position of the positioning target based on the observed position of the positioning target and the identified position of the positioning target;

[0015] The observed position is corrected based on the deviation to obtain an updated observed position of the positioning target.

[0016] In the above scheme, the observation module is also used to obtain the first carrier phase collected by the base station and the second carrier phase of the positioning target; the first carrier phase and the second carrier phase are subtracted to obtain a vector with the base station as the starting point and the positioning target as the end point; the vector is summed with the first coordinate of the base station, and the sum result is determined as the observation position of the positioning target.

[0017] In the above scheme, the positioning module is also used to identify at least one road element from the image of the road environment, determine a vector with the at least one road element as the starting point and the positioning target as the end point; obtain the second coordinate of the at least one road element in the map; superimpose the vector with the at least one road element as the starting point and the positioning target as the end point on the basis of the second coordinate to determine the third coordinate of the positioning target on the map; and use the third coordinate as the identification position of the positioning target.

[0018] In the above scheme, the positioning and online deviation estimation module is also used to determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the square of the observation error, the square of the initial observation error, and the sum of the square of the deviation initial error as the minimum constraint condition; wherein the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the deviation initial error is the difference between the deviation initial value and the deviation estimation initial value; the observation initial error is the difference between the observation initial value and the observation estimation initial value; wherein the first predicted position of the positioning target is the positioning result obtained by fusing the observation position of the positioning target and the identification position of the positioning target.

[0019] In the above scheme, the positioning and online deviation estimation module is also used to perform covariance normalization processing on the square of the observation error, the square of the initial observation error, and the square of the initial deviation error, respectively, to obtain the square of the observation error, the square of the initial observation error, and the square of the initial deviation error after the covariance normalization processing; with the minimum sum of the square of the observation error, the square of the initial observation error, and the square of the initial deviation error after the covariance normalization processing as the constraint condition, determine the deviation between the observed position of the positioning target and the identified position of the positioning target.

[0020] In the above scheme, the positioning and online deviation estimation module is also used to obtain the observation error corresponding to the target moment and each historical moment before the target moment, and add the squares of the observation errors corresponding to the target moment and each historical moment to obtain the square of the observation error at the first moment; determine the corresponding value when the deviation between the observed position of the positioning target and the identified position of the positioning target satisfies the following constraints: the sum of the square of the observation error at the first moment, the square of the initial observation error, and the square of the initial deviation error is minimized.

[0021] In the above scheme, the positioning and online deviation estimation module is also used to determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the minimum sum of the square of the observation error, the square of the initial observation error, the square of the initial deviation error, and the square of the motion error as the constraint condition; wherein the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the deviation initial error is the difference between the deviation initial value and the deviation estimation initial value; the observation initial error is the difference between the observation initial value and the observation estimation initial value; the motion error is the difference between the observed position of the positioning target and the second predicted position of the positioning target; wherein the second predicted position is the predicted position at the second moment, and is predicted based on the motion information and the observed position at the historical moment of the second moment.

[0022] In the above solution, the positioning processing device provided in the embodiment of the present application further includes:

[0023] The early warning module is used to generate early warning information when the deviation between the observed position of the positioning target and the identified position of the positioning target exceeds a deviation threshold, so as to prompt that the updated observed position of the positioning target is inaccurate.

[0024] In the above scheme, the early warning module is also used to determine that the map is inaccurate when the deviation between the observed position of the positioning target and the identified position of the positioning target exceeds a deviation threshold; and update the map with the deviation between the observed position of the positioning target and the identified position of the positioning target being less than the deviation threshold as a constraint condition.

[0025] An embodiment of the present application provides an electronic device, including:

[0026] A memory for storing executable instructions;

[0027] The processor is used to implement the positioning processing method provided in the embodiment of the present application when executing the executable instructions stored in the memory.

[0028] The embodiment of the present application provides a computer-readable storage medium storing executable instructions for implementing the positioning processing method provided in the embodiment of the present application when executed by a processor.

[0029] The embodiments of the present application have the following beneficial effects:

[0030] The observation position of the positioning target is corrected based on the deviation between the observation position of the positioning target and the identification position of the positioning target to obtain an updated observation position of the positioning target. The observation position of the positioning target can be corrected by the deviation between the high-precision map and the carrier phase to eliminate the influence of the deviation on high-precision positioning, thereby improving the positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a schematic diagram of the architecture of the positioning processing system 100 provided in an embodiment of the present application;

[0032] Figure 2 is a schematic diagram of the structure of the terminal 400 provided in an embodiment of the present application;

[0033] Figure 3A It is a flowchart of a positioning processing method provided in an embodiment of the present application;

[0034] Figure 3B It is a flowchart of a positioning processing method provided in an embodiment of the present application;

[0035] Figure 3C It is a flowchart of a positioning processing method provided in an embodiment of the present application;

[0036] Figure 4 It is a schematic diagram of an application scenario of the positioning processing method provided in an embodiment of the present application;

[0037] Figure 5It is a schematic diagram comparing the positioning result of directly fusing the high-precision map and RTK provided in the embodiment of the present application and the positioning result after eliminating the deviation between the high-precision map and RTK using the method provided in the embodiment of the present application. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0039] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0040] In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0042] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations.

[0043] 1) High-precision maps, in layman's terms, are electronic maps with higher precision and more data dimensions. The higher precision is reflected in at least centimeter accuracy, and the data dimensions reflect more road element location information. High-precision maps store a large amount of driving assistance information as structured data, such as lane lines, traffic signs and other road element location information.

[0044] 2) Real-time kinematic (RTK) carrier phase differential technology is a method for real-time processing of the differential carrier phase observations of two measurement stations. The carrier phase collected by the base station is sent to the user receiver to calculate the coordinates by difference.

[0045] 3) Autonomous driving refers to the ability to provide guidance and make decisions on vehicle driving tasks without the need for the test driver to perform physical driving operations, and to replace the test driver's control behavior to enable the vehicle to complete safe driving.

[0046] 4) The least squares method is a mathematical tool that is widely used in many disciplines such as error estimation, uncertainty, system identification, prediction, and forecasting, and other data processing fields.

[0047] In the current higher-level autonomous driving, or other high-precision vehicle positioning products, positioning is generally achieved by visual fusion of high-precision maps and real-time kinematic (RTK) carrier differential technology (for example, positioning is achieved by directly fusing high-precision maps with vision and RTK through filtering or optimization methods). In the usual fusion positioning algorithm, the global position (such as longitude and latitude) of road elements in the high-precision map is assumed to be consistent with the global observation longitude and latitude of RTK, that is, there is no deviation between the high-precision map and RTK. However, in actual application scenarios, due to the movement of the earth's crust, the position deviation of the global positioning system (GPS) base station, or clock synchronization, there is a certain deviation between the position of the RTK and the position of the elements in the high-precision map. Failure to consider this deviation will cause the accuracy of the final fusion positioning result to decrease, or even make the positioning result unusable.

[0048] It can be seen that during the implementation of this application, the following problems were found in the related technology:

[0049] 1) There are very high requirements for consistency between high-precision maps and RTK. When high-precision maps deviate from RTK due to untimely updates, the positioning accuracy will decrease, which will have a great impact on autonomous driving and other high-precision positioning products.

[0050] 2) The only way to eliminate the inaccuracy of high-precision maps is to iterate and update the high-precision maps offline. It takes a long time to collect data and make and publish maps. The cost of updating maps over a large area is very high, the update frequency is very slow, and the influence of deviations cannot be eliminated quickly in high-precision positioning.

[0051] In view of the above technical problems, the embodiments of the present application provide a positioning processing method, device, electronic device and computer-readable storage medium, which can eliminate positioning deviation and thus improve positioning accuracy. The following describes an exemplary application of the positioning processing method provided by the embodiments of the present application. The positioning processing method provided by the embodiments of the present application can be implemented by various electronic devices, for example, it can be implemented as various types of terminals such as smart phones, tablet computers, smart car terminals, smart wearable devices, etc., and can also be implemented as a server. Below, an exemplary application when the electronic device is implemented as a terminal will be described.

[0052] See also Figure 1 , Figure 1 It is an architectural diagram of the positioning processing system 100 provided in an embodiment of the present application. To support a positioning processing application, the terminal 400 is connected to the server 200 via the network 300. The network 300 can be a wide area network or a local area network, or a combination of the two.

[0053] The terminal 400 is used to determine the observed position of a positioning target based on the carrier phase; determine the identification position of the positioning target based on the road elements identified from the image of the road environment and the position of the road elements in the map; send the identification position and the observed position of the positioning target to the server 200 for correction processing, thereby receiving the updated observed position of the positioning target returned by the server 200, so as to display it in the client of the terminal 400 for the user to view the positioning related information.

[0054] The server 200 is used to receive the identification position and observation position of the positioning target sent by the terminal 400 to perform deviation determination and correction processing to obtain an updated observation position of the positioning target, and then return the updated observation position to the terminal 400.

[0055] In some embodiments, the terminal 400 implements the positioning processing method provided in the embodiments of the present application by running a computer program, and the computer program can be a native program or software module in the operating system; it can be a native application (APP, Application), that is, a program that needs to be installed in the operating system to run; it can also be a small program, that is, a program that can be run only by downloading it to a browser environment; it can also be an instant messaging small program that can be embedded in any APP. In short, the above-mentioned computer program can be an application, module or plug-in in any form.

[0056] In some embodiments, the server 200 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal and the server may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiments of the present application.

[0057] Next, the structure of the electronic device for implementing the positioning processing method provided in the embodiment of the present application is described. As mentioned above, the electronic device provided in the embodiment of the present application may be Figure 1 Terminal 400 in . Figure 2 , Figure 2is a schematic diagram of the structure of the terminal 400 provided in an embodiment of the present application, Figure 2 The terminal 400 shown includes: at least one processor 410, a memory 450, at least one network interface 420 and a user interface 430. The various components in the terminal 400 are coupled together via a bus system 440. It is understood that the bus system 440 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 440 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, the bus system 440 is not shown in FIG. Figure 2 Various buses are labeled as bus system 440 .

[0058] Processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0059] The user interface 430 includes one or more output devices 431 that enable presentation of media content, including one or more speakers and / or one or more visual display screens. The user interface 430 also includes one or more input devices 432, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0060] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state memory, hard disk drives, optical disk drives, etc. The memory 450 may optionally include one or more storage devices that are physically remote from the processor 410.

[0061] The memory 450 includes a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), and the volatile memory may be a random access memory (RAM). The memory 450 described in the embodiments of the present application is intended to include any suitable type of memory.

[0062] In some embodiments, memory 450 can store data to support various operations, examples of which include programs, modules, and data structures, or a subset or superset thereof, as exemplarily described below.

[0063] Operating system 451, including system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks;

[0064] A network communication module 452, for reaching other computing devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 include: Bluetooth, Wireless Compatibility Authentication (WiFi), and Universal Serial Bus (USB);

[0065] a presentation module 453 for enabling presentation of information via one or more output devices 431 (e.g., display screen, speaker, etc.) associated with the user interface 430 (e.g., a user interface for operating peripherals and displaying content and information);

[0066] The input processing module 454 is used to detect one or more user inputs or interactions from one of the one or more input devices 432 and translate the detected inputs or interactions.

[0067] In some embodiments, the positioning processing device provided in the embodiments of the present application can be implemented in software. Figure 2 The positioning processing device 455 stored in the memory 450 is shown, which can be software in the form of a program and a plug-in, etc., including the following software modules: an observation module 4551, a positioning module 4552, a positioning and online deviation estimation module 4553 and an early warning module 4554. These modules are logical, so they can be arbitrarily combined or further split according to the functions implemented. The functions of each module will be explained below.

[0068] The positioning processing method provided in the embodiment of the present application can be Figure 1 The terminal 400 or the server 200 in the embodiment can be executed separately, or the terminal 400 or the server 200 in the embodiment can be executed separately. Figure 1 The terminal 400 and the server 200 cooperate to execute.

[0069] Below, by Figure 1 The terminal 400 in the embodiment performs the positioning processing method provided in the embodiment of the present application as an example. Figure 3A , Figure 3A is a flowchart of the positioning processing method provided in the embodiment of the present application, which will be combined with Figure 3A The steps shown are explained.

[0070] In step 101, the observed position of the positioning target is determined based on the carrier phase.

[0071] In some embodiments, the above-mentioned determination of the observed position of the positioning target based on the carrier phase can be achieved in the following manner: obtaining the first carrier phase collected by the base station and the second carrier phase of the positioning target; subtracting the first carrier phase from the second carrier phase to obtain a vector with the base station as the starting point and the positioning target as the end point; summing the vector with the first coordinate of the base station, and determining the sum result as the observed position of the positioning target.

[0072] For example, the terminal device and the base station of the positioning target are both equipped with a global positioning system (GPS) receiver, and the GPS receiver in the terminal device and the GPS receiver in the base station simultaneously detect and observe the same navigation satellite. Among them, the three-dimensional coordinates of the base station are known (for example, the 2200+ Beidou ground-based signal enhancement stations built by Qianxun Positioning in the country). Since the atmospheric errors and satellite errors carried by the base station and the terminal device are approximately the same, the difference between the carrier phase observation values ​​of the base station and the terminal device of the positioning target can offset the error between the two. Therefore, the base station sends the measured carrier phase observation value to the terminal device, and the terminal device subtracts the two carrier phase observation values ​​to obtain the carrier phase observation equation; then the carrier phase observation equation is solved by the positioning solution algorithm (such as Kalman filtering method, least squares method, etc.), and a vector with the base station coordinates as the starting point and the terminal device coordinates as the end point is obtained. The vector is summed with the first coordinate of the base station, and the sum result is determined as the observation position of the positioning target. Among them, the first coordinate is the three-dimensional coordinate of the base station.

[0073] In step 102, based on the road elements identified from the image of the road environment and the positions of the road elements in the map, the identified position of the positioning target is determined.

[0074] In some embodiments, at least one road element is identified from an image of a road environment, and a vector with the at least one road element as a starting point and a positioning target as an end point is determined; a second coordinate of the at least one road element in the map is obtained; a vector with the at least one road element as a starting point and a positioning target as an end point is superimposed on the second coordinate to determine a third coordinate of the positioning target on the map; and the third coordinate is used as the identification position of the positioning target.

[0075] For example, the terminal device is provided with an image sensor. The image sensor in the terminal device collects road environment images in real time. The terminal device identifies at least one road element (such as a traffic light near a positioning target) from the image of the road environment, establishes a three-dimensional coordinate system based on the image of the road environment, and subtracts the coordinates of the positioning target in the three-dimensional coordinate system from the coordinates of the traffic light in the three-dimensional coordinate system to obtain a vector with the traffic light as the starting point and the positioning target as the end point; obtains the second coordinate of the traffic light in a map (such as a high-precision map), wherein the second coordinate is the three-dimensional coordinate of the road element in the high-precision map; superimposes a vector with the traffic light as the starting point and the positioning target as the end point on the basis of the second coordinate to determine the third coordinate of the positioning target on the high-precision map, wherein the third coordinate is the three-dimensional coordinate of the positioning target in the high-precision map; and uses the third coordinate as the identification position of the positioning target.

[0076] It should be noted that road elements may include road-related signs, such as various types of lane lines in the road environment (e.g., white solid lane lines, yellow dashed lane lines, left edge lane lines, right edge measuring lines), turn lines, stop lines, road edge lines, and traffic signs set beside or on the road (e.g., slow-moving traffic signs, no parking traffic signs, speed limit traffic signs), street lights, traffic lights, etc. Therefore, the examples of lane lines recorded above should not be regarded as limitations on the embodiments of the present application. Moreover, the number of road elements can be multiple. Generally speaking, the number of road elements is positively correlated with the accuracy of the identification position of the determined positioning target, that is, the more road elements there are, the higher the accuracy of the identification position of the positioning target.

[0077] In step 103, based on the observed position of the positioning target and the recognized position of the positioning target, a deviation between the observed position of the positioning target and the recognized position of the positioning target is determined.

[0078] In some embodiments, see Figure 3B , Figure 3B is a flow chart of a positioning processing method provided in an embodiment of the present application, showing Figure 3A Step 103 in the embodiment can be implemented by executing step 1031. The steps will be described in combination.

[0079] In step 1031, the deviation between the observed position of the positioning target and the identified position of the positioning target is determined with the minimum sum of the square of the observation error, the square of the initial observation error, and the square of the deviation initial error as the constraint condition.

[0080] Among them, the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the deviation initial error is the difference between the deviation initial value and the deviation estimated initial value; the observation initial error is the difference between the observation initial value and the observation estimated initial value; among them, the first predicted position of the positioning target is the positioning result obtained by fusing the observed position of the positioning target and the recognized position of the positioning target.

[0081] For example, the deviation between the observed position of the positioning target and the identified position of the positioning target is determined according to the constraint condition when the sum of the square of the observation error, the square of the initial observation error, the square of the initial deviation error, and the square of the deviation error is minimized. The calculation formula (1) is:

[0082]

[0083] Where i is a natural number greater than zero; z i is the observed position of the positioning target determined based on the carrier phase at the i-th moment; h(x i ,b i ) is the predicted positioning result at the i-th moment, and is predicted by fusing the observed position and the identified position of the positioning target at the i-th moment, that is, the first predicted position; x0 is the initial value of the deviation; is the initial value of the deviation estimate; b0 is the initial value of the observation; Estimate initial values ​​for the observations.

[0084] Among them, h(x i ,b i ) is implemented by an observation model. For example, the observation model can be a geometric model. The observed position is predicted based on the deviation between the observed position of the positioning target at the i-th moment and the identified position of the positioning target at the i-th moment. For example, the deviation between the observed position of the positioning target at the i-th moment and the identified position of the positioning target at the i-th moment is represented by a deviation vector. The deviation vector is superimposed on the observed position of the positioning target at the i-th moment to obtain the predicted observed position. It can also be implemented by a neural network model. The identified position of the positioning target at the i-th moment and the observed position of the positioning target, the deviation therebetween, and the updated observed position of the positioning target are used as sample features to train the neural network model. The first predicted position is predicted by the trained neural network model. x0 and b0 are pre-set initial values. It is the result predicted based on the initial value.

[0085] In the embodiment of the present application, when determining the deviation between the observed position of the positioning target and the identified position of the positioning target, only the observation error at the current moment is considered, which can simply and conveniently determine the deviation between the observed position of the positioning target and the identified position of the positioning target.

[0086] In some examples, the square of the observation error, the square of the initial observation error, and the square of the initial deviation error are respectively covariance normalized to obtain the square of the observation error, the square of the initial observation error, and the square of the initial deviation error after covariance normalization; the deviation between the observed position of the positioning target and the identified position of the positioning target is determined with the minimum sum of the square of the observation error, the square of the initial observation error, and the square of the initial deviation error after covariance normalization as a constraint condition.

[0087] Taking the covariance normalization of the observation error as an example, the covariance is used to describe the correlation between the observation position and the first predicted position. For example, the observation position z is calculated. i and the first predicted position h(x i ,b i ), z i and h(x i ,b i ) i ,h(x i ,b i )) is calculated by formula (2) as

[0088]

[0089] Where k is a natural number greater than 1. For z i The mean of is h(x i ,b i ). Similarly, we can obtain the covariance between the initial value of the deviation and the initial value of the deviation estimate, the covariance between the initial value of the observation and the initial value of the observation estimate, etc.

[0090] In the embodiment of the present application, since the units of paired data such as the observed position and the first predicted position, the covariance of the initial deviation value and the initial deviation estimation value, the initial observation value and the initial observation estimation value are not unified, the amplitude influence caused by the unit change between the two is eliminated by normalizing the covariance, so as to purely reflect the correlation between the paired data.

[0091] In some examples, the observation error corresponding to the target moment and each historical moment before the target moment is obtained, and the squares of the observation errors corresponding to the target moment and each historical moment are summed to obtain the square of the observation error at the first moment; and the corresponding value is determined when the deviation between the observed position of the positioning target and the identified position of the positioning target satisfies the following constraints: the sum of the square of the observation error at the first moment, the square of the initial observation error, and the square of the deviation initial error is minimized.

[0092] For example, the observation error O corresponding to the target time I and each historical time before the target time I I The calculation formula (3) is:

[0093]

[0094] It should be noted that the observation error O corresponding to the target time I and each historical time before the target time I is I , the squares of the observation errors corresponding to the target time I and each historical time before the target time I are added to obtain the square of the observation error at the first time. The observation error at the first time here represents the observation error determined by considering all historical times from the 0th time to the Ith time.

[0095] In an embodiment of the present application, observation errors at all times are taken into account when determining the deviation between the observed position of the positioning target and the identified position of the positioning target, so that the deviation between the observed position of the positioning target and the identified position of the positioning target is determined with higher accuracy.

[0096] In some embodiments, the deviation can also be determined based on the motion information. The deviation between the observed position of the positioning target and the identified position of the positioning target is determined with the minimum sum of the square of the observation error, the square of the initial observation error, the square of the initial deviation error, and the square of the motion error as the constraint condition.

[0097] Among them, the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the initial deviation error is the difference between the initial deviation value and the initial deviation estimation value; the initial observation error is the difference between the initial observation value and the initial observation estimation value; the motion error is the difference between the observed position of the positioning target and the second predicted position of the positioning target; wherein the second predicted position is the predicted position at the second moment, and is predicted based on the motion information and the observed position at the historical moment of the second moment. Among them, the second moment refers to the target moment, and the historical moment of the second moment refers to the moment before the target moment.

[0098] For example, the deviation between the observed position of the positioning target and the identified position of the positioning target is determined by taking the sum of the square of the observation error, the square of the initial observation error, the square of the initial deviation error, the square of the deviation error, and the square of the motion error as the minimum constraint condition. The calculation formula (4) is:

[0099]

[0100] Among them, z i is the observed position of the positioning target determined based on the carrier phase at the i-th moment; h(x i ,b i ) is the predicted positioning result at the i-th moment, and is predicted by fusing the observed position and the identified position of the positioning target at the i-th moment, i.e., the first predicted position; x i is the observed position of the positioning target updated at the i-th moment; f(x i-1 ) is the motion prediction model, that is, based on the motion information and the optimal positioning result x at the i-1th moment i-1 Recursively obtain the optimal positioning result x at the i-1th moment i , i.e., the second predicted position, where the motion information may be the acceleration of the positioning target measured by the inertial measurement unit (IMU) and the angular velocity of the target vehicle measured by the gyroscope, thereby predicting the positioning point at the i-th moment according to the acceleration and angular velocity of the positioning target and the positioning point at the i-1th moment; x0 is the initial value of the deviation; is the initial value of the deviation estimate; b0 is the initial value of the observation; Estimate initial values ​​for the observations.

[0101] In some embodiments, taking the deviation between the observed position of the positioning target and the identified position of the positioning target as an example, the deviation between the observed position of the positioning target and the identified position of the positioning target is determined according to the constraint condition when the sum of the square of the observation error at the first moment, the square of the initial observation error, and the square of the deviation error is minimized. The calculation formula (5) is:

[0102]

[0103] Among them, z i is the observed position of the positioning target by RTK at the i-th moment. i ,b i ) is the predicted positioning result at the i-th moment, and is predicted by fusing the observed position and the identified position of the positioning target at the i-th moment, that is, the first predicted position h(x i ,b i), can also be obtained by the following method: the recognition position of the positioning target at the historical moment (the historical moment here refers to all moments before and including the i-th moment) and the observation position of the positioning target, the deviation between the two and the updated observation position of the positioning target are used as sample features to train the neural network model, and h(x i ,b i ).

[0104] In step 104, the observed position of the positioning target is corrected based on the deviation between the observed position of the positioning target and the recognized position of the positioning target to obtain an updated observed position of the positioning target.

[0105] In some embodiments, the optimal deviation between the observed position of the positioning target and the recognized position of the positioning target is determined. The observed position of the positioning target is superimposed to obtain the updated observed position of the positioning target.

[0106] For example, assuming that the observed position (i.e., three-dimensional coordinates) of the positioning target is (x1, y1, z1), the optimal deviation (i.e., vector) is (x2, y2, z2), and the updated observation position (i.e., three-dimensional coordinates) of the positioning target is (x1+x2, y1+y2, z1+z2).

[0107] In an embodiment of the present application, based on the deviation between the observation position of the positioning target by the high-precision map and the RTK, the observation position of the positioning target is corrected to obtain an updated observation position of the positioning target, so as to improve the accuracy of positioning the positioning target.

[0108] In some embodiments, see Figure 3C , Figure 3C It is a flow chart of the positioning processing method provided in an embodiment of the present application, showing that after step 104 in 3A, steps 1041 to 1043 may also be executed, and each step will be explained in combination.

[0109] In step 1041, when the deviation between the observed position of the positioning target and the identified position of the positioning target exceeds a deviation threshold, a warning message is generated to indicate that the updated observed position of the positioning target is inaccurate.

[0110] In some examples, the deviation threshold is used to characterize whether the deviation between the observed position of the positioning target and the identified position of the positioning target exceeds the normal deviation range. The warning information can be prompt information such as inaccurate high-precision map, so that the prompt information can be displayed through the terminal device to let the user know that the updated observed position of the positioning target is inaccurate, so as to facilitate the user to perform corresponding operations to overcome the inaccuracy of the high-precision map.

[0111] In step 1042, when the deviation between the observed position of the positioning target and the recognized position of the positioning target exceeds a deviation threshold, it is determined that the map is inaccurate.

[0112] In the embodiment of the present application, it is possible to quickly detect and perceive the deviation between the high-precision map of each place where the positioning target passes and the RTK's observation position of the positioning target, and for places with large deviations, it is possible to quickly give an early warning of inaccurate high-precision maps, thereby reducing detection costs.

[0113] In step 1043, the map is updated with the constraint that the deviation between the observed position of the positioning target and the recognized position of the positioning target is less than the deviation threshold.

[0114] In some embodiments, for the deviation between the high-precision map and the RTK observed position of the positioning target, when the deviation between the observed position of the positioning target and the identified position of the positioning target is less than the deviation threshold, a new and more accurate high-precision map version is updated and released to reduce this deviation, thereby reducing the impact of this deviation on high-precision positioning, so as to improve the accuracy of the determined observation position of the positioning target.

[0115] Below, an exemplary application of the embodiment of the present application in an actual application scenario will be described. Taking the map navigation client as an example, the user's positioning request for the target vehicle is obtained, and the position of the target vehicle is determined in real time based on the perception results based on the image sensor (such as a camera) (i.e., the position of the road elements identified from the image of the road environment) and the position of the road elements in the high-precision map; the deviation between the position of the target vehicle and the observed position of the target vehicle provided in real time based on the carrier phase is determined in real time; based on this deviation, the observed position of the target vehicle is corrected in real time so that the user can obtain the updated observed position of the target vehicle. The embodiment of the present application can still ensure accurate and stable high-precision positioning output when there is a certain deviation between the high-precision map and the RTK positioning technology; and through this technology, the deviation between the high-precision map and the RTK positioning technology of each place where the target vehicle passes can be quickly detected. For places with large deviations, an early warning of inaccurate high-precision maps can be quickly given to reduce detection costs.

[0116] See also Figure 4 , Figure 4 This is a schematic diagram of an application scenario of the positioning processing method provided in the embodiment of the present application. Figure 4 The specific implementation scenario of the positioning processing method provided in the embodiment of the present application is described.

[0117] Among them, the positioning and online deviation estimation module has three inputs, including the observed position of the target vehicle by RTK, the position information of road elements such as lane lines perceived by the image sensor, and the position of road elements in the high-precision map. Positioning and online deviation estimation are performed through these three inputs to output the optimal positioning result of the target vehicle. The specific process of the positioning and online deviation estimation module is as follows: first, based on the image perception results of the image sensor (such as a camera, an optical sensor, etc.) (i.e., the road elements identified from the image of the road environment) and the position of the road element in the high-precision map, the position of the target vehicle is determined in real time; secondly, the observed position of the target vehicle is obtained in real time based on the carrier phase, and the deviation between the position of the target vehicle and the observed position based on the carrier phase is determined in real time; then, the observed position of the target vehicle is corrected based on the deviation between the position of the target vehicle and the observed position, and finally the updated observed position of the target vehicle is output. The positioning and online deviation estimation module can be implemented in the following ways:

[0118] The functions implemented by the positioning and online deviation estimation modules are converted into the following mathematical model: The calculation formula (6) is:

[0119]

[0120] Among them, x 0:k represents the optimal positioning result (i.e., the updated observed position of the target vehicle) output by the positioning and online deviation estimation module at each moment from the 0th moment to the kth moment, b 0:k represents the deviation between the RTK observation position of the target vehicle and the high-precision map (the position of the target vehicle calculated by the high-precision map and the image perception result) at each moment from the 0th moment to the kth moment, z 1:k represents the RTK observation position of the target vehicle at each moment from the 1st moment to the kth moment, u 0:k-1 Represents the motion information at each moment from the 0th moment to the k-1th moment (for example, when a conventional measurement unit is used, the acceleration and angular velocity of the target vehicle measured by the conventional measurement unit).

[0121] It should be noted that the positioning and online deviation estimation module can take motion information into account or not when correcting the observed position of the target vehicle. The deviation between the observed position of the target vehicle by RTK at each moment from the 1st moment to the kth moment and the identified position of the target vehicle calculated by the high-precision map and the image perception result can be determined, and the observed position of the target vehicle by RTK can be corrected based on the deviation; the observed position of the target vehicle by RTK, the identified position of the target vehicle calculated by the high-precision map and the image perception result, and the motion information can also be combined to determine the deviation between the two, and the observed position of the target vehicle by RTK can be corrected based on the deviation. Therefore, the positioning and online deviation estimation module can obtain the estimation of the deviation between the observed position of the target vehicle by RTK and the high-precision map (the position of the target vehicle calculated by the high-precision map and the image perception result), and the optimal positioning result (i.e., the updated observed position of the target vehicle).

[0122] For the above problem of optimizing the positioning results, by using the derivation of the Bayesian formula, the problem of optimizing the positioning results can be equivalent to the least squares optimization problem of solving formula (7), as shown in formula (7):

[0123]

[0124] Among them, z i is the RTK observation position of the target vehicle at the i-th moment. i ,b i ) is the prediction result of the observation model (i.e., the first predicted position), that is, the observation model integrates the RTK observation position of the target vehicle and the identification position of the target vehicle calculated by the high-precision map and image perception results, and predicts the positioning result. i is the observed position of the target vehicle updated at the i-th moment. i-1 ) is a motion prediction model, which is based on the motion information and the optimal positioning result x at the i-1th moment. i-1 Recursively obtain the optimal positioning result x at the i-1th moment i It should be noted that the motion information here can be the acceleration of the target vehicle measured by the inertial measurement unit (IMU), or the angular velocity of the target vehicle measured by the gyroscope; x0 is the initial value of the deviation; is the initial value of the deviation estimate; b0 is the initial value of the observation; Estimate initial values ​​for the observations.

[0125] Among them, R, Q, P0 and P b0In order to perform covariance normalization, since the units of the squares in formula (7) are not uniform, normalization is performed through covariance operation.

[0126] In some embodiments, whenever new RTK input data is received, the positioning and online deviation estimation module reconstructs the optimization problem in the above formula (6) and converts it into the least squares optimization problem in formula (7) to obtain the current optimal deviation Then superimpose the deviation with the RTK positioning result to obtain the optimal positioning result.

[0127] In some embodiments, z i It can also be the observation position of the image sensor on the road element at the i-th moment (i.e., the image perception result). i ,b i ) is the prediction result of the observation model (i.e., the first predicted position), that is, the recognition position of the road element predicted by the observation model based on the high-precision map and the observation position of the road element by the image sensor at the i-th moment. i is the observed position of the road element updated at the i-th moment. i-1 ) is the motion prediction model. Since the motion is relative, it is the same as the motion prediction model for the RTK observation position in formula (7). x0 is the initial value of the deviation; is the initial value of the deviation estimate; b0 is the initial value of the observation; Estimate initial values ​​for the observations.

[0128] Therefore, the current optimal deviation is obtained based on the input data of the image perception result; since the road elements and the target vehicle correspond to each other, the deviation of the road elements is the deviation of the target vehicle, and the RTK observation position of the target vehicle is corrected based on the deviation to update the observation position of the target vehicle.

[0129] In the embodiment of the present application, for RTK input data or image sensor input data, the positioning and online deviation estimation modules can output the current optimal deviation and the optimal positioning result in real time, which can eliminate the positioning deviation and thus improve the positioning accuracy.

[0130] In some examples, there are two common methods for solving optimization problems: one is filtering and the other is optimization. When using the filtering method, always keep the optimal predictions of x and b and their covariance at the previous moment. Recurse to the current observation moment through motion information, and then use the current observation for correction and update (for example, the Extended Kalman Filter (EKF) method). When using the optimization method, the least squares optimization problem shown above can be constructed, and the optimal x can be obtained by solving the least squares optimization problem. i and b i It should be noted that the constructed least squares optimization problem can only include some states (x i and b i ), or it can contain x at all times from the beginning to now i and b i status.

[0131] See also Figure 5 , Figure 5 It is a comparative schematic diagram of the positioning result of the direct fusion of high-precision map and RTK provided in the embodiment of the present application and the positioning result after eliminating the deviation between high-precision map and RTK using the method provided in the embodiment of the present application. Among them, 501 is the RTK observation position (that is, the RTK observation position of the target vehicle), 502 is the positioning point obtained by the direct fusion method in the related art, that is, the RTK observation position of the target vehicle and the identification position of the target vehicle calculated based on the high-precision map and image perception results are directly fused to obtain the positioning point, 503 is the positioning point output by the positioning and online deviation estimation module provided in the embodiment of the present application, that is, the updated observation position obtained after the RTK observation position is corrected using the positioning processing method provided in the embodiment of the present application; 504 and 505 are the lane lines on the left and right sides of a lane in the high-precision map. It is worth noting that since the actual position of the target vehicle is close to the center of the lane in the lateral direction, it can be seen that the positioning point obtained by the positioning processing method provided in the embodiment of the present application is closer to the actual position of the target vehicle.

[0132] It should be noted that the deviation between high-precision maps and RTK positioning technology (or other high-precision GPS) can also be continuously updated and released offline to continuously reduce the deviation, thereby continuously reducing the impact of the deviation on high-precision positioning. The deviation size of different locations in the high-precision map can also be achieved through other manual methods.

[0133] The following is a description of an exemplary structure of the positioning processing device 455 provided in the embodiment of the present application implemented as a software module. In some embodiments, for example, Figure 2As shown, the software modules stored in the positioning processing device 455 of the memory 450 may include:

[0134] The observation module 4551 is used to determine the observation position of the positioning target based on the carrier phase; the positioning module 4552 is used to determine the identification position of the positioning target based on the road elements identified from the image of the road environment and the positions of the road elements in the map; the positioning and online deviation estimation module 4553 is used to determine the deviation between the observation position of the positioning target and the identification position of the positioning target based on the observation position of the positioning target and the identification position of the positioning target; the observation position is corrected based on the deviation to obtain an updated observation position of the positioning target.

[0135] In the above scheme, the observation module 4551 is also used to obtain the first carrier phase collected by the base station and the second carrier phase of the positioning target; the first carrier phase and the second carrier phase are subtracted to obtain a vector with the base station as the starting point and the positioning target as the end point; the vector is summed with the first coordinate of the base station, and the sum result is determined as the observation position of the positioning target.

[0136] In the above scheme, the positioning module 4552 is also used to identify at least one road element from the image of the road environment, determine a vector with the at least one road element as the starting point and the positioning target as the end point; obtain the second coordinate of the at least one road element in the map; superimpose the vector with the at least one road element as the starting point and the positioning target as the end point on the basis of the second coordinate to determine the third coordinate of the positioning target on the map; and use the third coordinate as the identification position of the positioning target.

[0137] In the above scheme, the positioning and online deviation estimation module 4553 is also used to determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the square of the observation error, the square of the initial observation error, and the sum of the square of the deviation initial error as the minimum constraint condition; wherein the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the deviation initial error is the difference between the deviation initial value and the deviation estimation initial value; the observation initial error is the difference between the observation initial value and the observation estimation initial value; wherein the first predicted position of the positioning target is the positioning result obtained by fusing the observation position of the positioning target and the identification position of the positioning target.

[0138] In the above scheme, the positioning and online deviation estimation module 4553 is also used to perform covariance normalization on the square of the observation error, the square of the initial observation error, and the square of the initial deviation error, respectively, to obtain the square of the observation error, the square of the initial observation error, and the square of the initial deviation error after the covariance normalization processing; with the minimum sum of the square of the observation error, the square of the initial observation error, and the square of the initial deviation error after the covariance normalization processing as the constraint condition, determine the deviation of the observed position of the positioning target and the identified position of the positioning target.

[0139] In the above scheme, the positioning and online deviation estimation module 4553 is also used to obtain the observation error corresponding to the target moment and each historical moment before the target moment, and add the squares of the observation errors corresponding to the target moment and each historical moment to obtain the square of the observation error at the first moment; determine the corresponding value when the deviation between the observed position of the positioning target and the identified position of the positioning target satisfies the following constraints: the sum of the square of the observation error at the first moment, the square of the initial observation error, and the square of the initial deviation error is minimized.

[0140] In the above scheme, the positioning and online deviation estimation module 4553 is also used to determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the minimum sum of the square of the observation error, the square of the initial observation error, the square of the initial deviation error, and the square of the motion error as the constraint condition; wherein the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the initial deviation error is the difference between the initial deviation value and the initial deviation estimation value; the initial observation error is the difference between the initial observation value and the initial observation estimation value; the motion error is the difference between the observed position of the positioning target and the second predicted position of the positioning target; wherein the second predicted position is the predicted position at the second moment, and is predicted based on the motion information and the observed position at the historical moment of the second moment.

[0141] In the above scheme, the positioning processing device provided in the embodiment of the present application also includes: an early warning module 4554, which is used to generate early warning information when the deviation between the observation position of the positioning target and the identification position of the positioning target exceeds the deviation threshold, so as to prompt that the updated observation position of the positioning target is inaccurate.

[0142] In the above scheme, the early warning module is also used to determine that the map is inaccurate when the deviation between the observed position of the positioning target and the identified position of the positioning target exceeds a deviation threshold; and update the map with the deviation between the observed position of the positioning target and the identified position of the positioning target being less than the deviation threshold as a constraint condition.

[0143] The embodiment of the present application provides a computer program product or a computer program, 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 positioning processing method described in the embodiment of the present application.

[0144] The present application embodiment provides a computer-readable storage medium storing executable instructions, wherein the executable instructions are stored. When the executable instructions are executed by a processor, the processor will execute the positioning processing method provided by the present application embodiment, for example, Figure 3A , 3B , the positioning processing method shown in 3C.

[0145] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EP ROM, EEPROM, flash memory, magnetic surface storage, optical disk, or CD-ROM; or it may be various devices including one or any combination of the above memories.

[0146] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0147] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0148] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.

[0149] In summary, through the embodiment of the present application, the observed position of the positioning target is corrected based on the deviation between the observed position of the positioning target and the identified position of the positioning target to obtain an updated observed position of the positioning target, taking into account the deviation between the high-precision map and the RTK, and the observed position of the positioning target can be corrected by the deviation to eliminate the influence of the deviation on the high-precision positioning, thereby improving the positioning accuracy; since the units between the observed position and the first predicted position, the covariance of the deviation initial value and the deviation estimated initial value, the observation initial value and the observation estimated initial value are not uniform, the amplitude influence caused by the unit change between the two is eliminated by normalizing the covariance, so as to purely reflect the correlation between the paired data; when the deviation between the observed position of the positioning target and the identified position of the positioning target is less than the deviation threshold, a new and more accurate high-precision map version is updated and released to reduce the deviation, thereby reducing the influence of the deviation on the high-precision positioning, so as to improve the accuracy of the determined observation position of the positioning target.

[0150] The above is only an embodiment of the present application and is not intended to limit the protection scope of the present application. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. A positioning processing method, characterized in that: include: Determine the observed position of the positioning target based on the carrier phase; Determining the identified position of the positioning target based on the road elements identified from the image of the road environment and the positions of the road elements in the map; Determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the minimum sum of the square of the observation error, the square of the initial observation error, and the square of the initial deviation error as a constraint condition; or, Determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the sum of the square of the observation error, the square of the initial observation error, the square of the initial deviation error, and the square of the motion error being minimized as a constraint condition; Among them, the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the deviation initial error is the difference between the deviation initial value and the deviation estimation initial value; the observation initial error is the difference between the observation initial value and the observation estimation initial value; the first predicted position of the positioning target is the positioning result obtained by fusing the observed position of the positioning target and the identified position of the positioning target; the motion error is the difference between the observed position of the positioning target and the second predicted position of the positioning target; the second predicted position is the predicted position at the second moment, and is predicted based on the motion information of the historical moment of the second moment and the observed position; The observed position is corrected based on the deviation to obtain an updated observed position of the positioning target.

2. The method according to claim 1, characterized in that The method of determining the observed position of the positioning target based on the carrier phase includes: Acquire a first carrier phase collected by a reference station and a second carrier phase of the positioning target; Subtracting the first carrier phase from the second carrier phase to obtain a vector with the reference station as a starting point and the positioning target as an end point; The vector is summed with the first coordinate of the reference station, and the summation result is determined as the observed position of the positioning target.

3. The method according to claim 1, characterized in that The determining the identification position of the positioning target based on the road elements identified from the image of the road environment and the positions of the road elements in the map includes: Identify at least one road element from the image of the road environment, and determine a vector with the at least one road element as a starting point and the positioning target as an end point; Acquire a second coordinate of the at least one road element in the map; superimposing the vector having the at least one road element as a starting point and the positioning target as an end point on the basis of the second coordinate, so as to determine a third coordinate of the positioning target on the map; The third coordinate is used as the identification position of the positioning target.

4. The method according to claim 1, characterized in that The method of determining the deviation between the observed position of the positioning target and the identified position of the positioning target by taking the sum of the square of the observation error, the square of the initial observation error, and the square of the initial deviation error as a constraint condition to be minimized comprises: Performing covariance normalization processing on the square of the observation error, the square of the initial observation error, and the square of the initial deviation error, respectively, to obtain the square of the observation error, the square of the initial observation error, and the square of the initial deviation error after the covariance normalization processing; The deviation between the observed position of the positioning target and the identified position of the positioning target is determined with the minimum sum of the square of the observation error after covariance normalization, the square of the initial observation error, and the square of the initial deviation error as a constraint condition.

5. The method according to claim 1, characterized in that The method of determining the deviation between the observed position of the positioning target and the identified position of the positioning target by taking the sum of the square of the observation error, the square of the initial observation error, and the square of the initial deviation error as a constraint condition to be minimized comprises: Obtain the observation error corresponding to the target moment and each historical moment before the target moment, and add the squares of the observation errors corresponding to the target moment and each historical moment to obtain the square of the observation error at the first moment; Determine a corresponding value when the deviation between the observed position of the positioning target and the identified position of the positioning target satisfies the following constraint conditions: the sum of the square of the observation error at the first moment, the square of the initial observation error, and the square of the initial deviation error is minimized.

6. The method according to claim 1, characterized in that The method further comprises: When the deviation between the observed position of the positioning target and the identified position of the positioning target exceeds a deviation threshold, an early warning message is generated to indicate that the updated observed position of the positioning target is inaccurate.

7. The method according to claim 6, characterized in that The method further comprises: When the deviation between the observed position of the positioning target and the identified position of the positioning target exceeds a deviation threshold, determining that the map is inaccurate; The map is updated with the constraint that the deviation between the observed position of the positioning target and the recognized position of the positioning target is less than the deviation threshold.

8. The method according to any one of claims 1 to 7, characterized in that: The correcting the observed position based on the deviation to obtain an updated observed position of the positioning target includes: The deviation is superimposed on the observed position of the positioning target to obtain an updated observed position of the positioning target.

9. A positioning processing device, characterized in that: include: An observation module, used to determine the observation position of the positioning target based on the carrier phase; A positioning module, used to determine the recognition position of the positioning target based on the road elements recognized from the image of the road environment and the positions of the road elements in the map; A positioning and online deviation estimation module, used to determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the square of the observation error, the square of the initial observation error, and the square of the initial deviation error as the constraint condition; or, to determine the deviation between the observed position of the positioning target and the identified position of the positioning target with the square of the observation error, the square of the initial observation error, the square of the initial deviation error, and the square of the motion error as the constraint condition; Among them, the observation error is the difference between the observed position of the positioning target and the first predicted position of the positioning target; the deviation initial error is the difference between the deviation initial value and the deviation estimation initial value; the observation initial error is the difference between the observation initial value and the observation estimation initial value; the first predicted position of the positioning target is the positioning result obtained by fusing the observed position of the positioning target and the identified position of the positioning target; the motion error is the difference between the observed position of the positioning target and the second predicted position of the positioning target; the second predicted position is the predicted position at the second moment, and is predicted based on the motion information of the historical moment of the second moment and the observed position; The observed position is corrected based on the deviation to obtain an updated observed position of the positioning target.

10. An electronic device, characterized in that: The electronic device comprises: A memory for storing executable instructions; A processor, configured to implement the positioning processing method according to any one of claims 1 to 8 when running the executable instructions stored in the memory.

11. A computer-readable storage medium storing executable instructions, characterized in that: When the executable instructions are executed by a processor, the positioning processing method described in any one of claims 1 to 8 is implemented.

12. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the positioning processing method described in any one of claims 1 to 8 is implemented.

Images