Map data verification method and device, vehicle and storage medium

Abstract

The invention relates to a map data verification method and device, a vehicle and a storage medium. The method comprises the steps that when a user drives a vehicle, position information of the vehicle is collected; when the position information is collected, a real vehicle driving scene of the vehicle is collected through a visual sensor of the vehicle; constructing a first driving scene according to the position information and preset map data; and comparing the first driving scene with the real vehicle driving scene to obtain a first verification result of the map data. According to the map data verification method provided by the invention, the real vehicle driving scene collected by the sensor erected on the vehicle can be compared with the first driving scene obtained by rendering based on the positioning information and the map data during vehicle driving when the vehicle is driven, so that the accuracy of the map data is verified; the problems that existing map data verification is separated from an actual scene, the cost is high, and the map data verification accuracy is low are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of in-vehicle systems, and particularly to a method and device for verifying map data, a vehicle, and a storage medium. Background Art

[0002] Compared with the traditional parking function, the intelligent parking function after version update can select the optimal parking space by itself and automatically park on the basis of the existing parking lot map data. It extracts the driving information in real time through sensors such as surround cameras and millimeter-wave radars, and compares it with the high-precision map to realize the positioning of the vehicle in the high-precision map of the parking lot, and solve the problem of vehicle automatic driving from the parking lot entrance to the parking lot. However, the feasibility of these parking methods and functions depends on the high-precision map. For example, the accuracy of the existing high-precision map of the parking lot. When the accuracy of the map data is poor, it is difficult to ensure the actual feasibility of the intelligent parking function.

[0003] In view of the above problems, the existing technology conducts large-scale simulation tests by building a simulation cloud platform. Through a large number of scenario designs and constructions, and constructing specific data to verify the map data. However, this verification method is different from the actual scenario of vehicle driving, and the accuracy of simulation verification is low. Summary of the Invention

[0004] One of the purposes of the present invention is to provide a method for verifying map data to solve the problem of low accuracy of simulation verification in the existing technology; the second purpose is to provide a device for verifying map data; the third purpose is to provide an electronic vehicle capable of implementing the map data verification method provided by the present application; the fourth purpose is to provide a storage medium capable of implementing the map data verification method provided by the present application.

[0005] In order to achieve the above purposes, the technical solutions adopted by the present invention are as follows:

[0006] In the first aspect, the present invention provides a method for verifying map data, including:

[0007] When the user drives the vehicle, collect the position information of the vehicle;

[0008] When collecting the position information, collect the actual vehicle driving scenario of the vehicle through the visual sensor of the vehicle;

[0009] Construct a first driving scenario according to the position information and the pre-set map data;

[0010] Compare the first driving scenario with the actual vehicle driving scenario to obtain a first verification result of the map data.

[0011] According to the above technical means, since the map data verification method provided by the present application can compare the real vehicle driving scenario collected by the sensors installed on the vehicle with the first driving scenario rendered based on the positioning information and map data during vehicle driving, thereby verifying the accuracy of the map data, it avoids the problems of the existing map data verification being divorced from the actual scenario, high cost, and low accuracy of map data verification.

[0012] Moreover, the map data verification method provided by the present application performs map data verification actions when the user is driving the vehicle. The test and verification behaviors are synchronized with the vehicle driving, and the map data at each moment is verified once. Only one complete vehicle driving process is required to verify the map data of the entire driving path, avoiding the repeated action of collecting map data through multiple real vehicle tests and improving the efficiency of map data verification.

[0013] Furthermore, in the map data verification method provided by the present application, it further includes:

[0014] When the user is driving the vehicle, obtain the positioning signal of the vehicle through a positioning sensor;

[0015] Calculate the coordinate system transformation according to the positioning signal and the map data to obtain the position information.

[0016] According to the above technical means, since the present application can also perform coordinate system transformation according to the difference between the positioning signal and the map data, thereby quickly determining the position information corresponding to the vehicle in the map data currently, it improves the efficiency of map data verification.

[0017] Furthermore, in the map data verification method provided by the present application, it further includes:

[0018] When the user is driving the vehicle, perform guided positioning on the vehicle through the global satellite navigation system to obtain a first positioning result;

[0019] Perform semantic positioning on the vehicle according to the semantic map to obtain a second positioning result;

[0020] Perform positioning convergence processing according to the first positioning result and the second positioning result to obtain the positioning signal.

[0021] According to the above technical means, since the technical solution provided by the present application can perform positioning convergence by integrating the semantic map and the GNSS signal, it avoids the problem of poor positioning effect caused by weak inertial navigation signals in the vehicle driving scenario, such as the problem of poor accuracy of the position information determined based on the GNSS signal due to weak inertial navigation signals in the underground parking lot, ensuring the accuracy of the vehicle's current position information, and further ensuring the accuracy of map data verification.

[0022] Further, the map data verification method provided by the present application further includes:

[0023] Performing visual simultaneous localization on the vehicle according to the visual feature map to obtain a third localization result;

[0024] Performing localization assistance on the positioning signal according to the third localization result to obtain the corrected positioning signal.

[0025] According to the above technical means, since the present application can also perform localization assistance through the visual feature map, it avoids the problem of poor positioning effect caused by weak inertial navigation signals in the vehicle driving scenario. For example, in an underground parking lot where the inertial navigation signal is weak and the accuracy of the position information determined based on the GNSS signal is poor, it ensures the accuracy of the vehicle's current position information, and further ensures the accuracy of map data verification.

[0026] Further, the map data verification method provided by the present application further includes:

[0027] Performing a positioning judgment based on the position information and a pre-set map loading area to obtain a positioning judgment result;

[0028] When the positioning judgment result indicates that the position information falls within the map loading area, constructing a second driving scenario according to the position information and the pre-set map data;

[0029] Comparing the second driving scenario with the actual vehicle driving scenario to obtain a second verification result of the map data.

[0030] According to the above technical means, since the present application can also perform a positioning area judgment, it avoids the continuous loading of map data and rendering of the driving scenario by the vehicle, reduces the data volume of map data verification, and improves the efficiency of map data verification.

[0031] Further, the map data verification method provided by the present application further includes:

[0032] Performing a simulation test according to the position information and the map data to obtain simulation test data;

[0033] Constructing a third driving scenario according to the simulation test data and the map data;

[0034] Comparing the third driving scenario with the actual vehicle driving scenario to obtain a third verification result of the map data.

[0035] According to the above technical means, since the present application can also re-verify the map data after real vehicle testing through simulation testing, and can perform simple inspection and processing on the collected parking lot map data according to the verification results of simulation, the time cost is greatly reduced. Moreover, the real vehicle verification only needs to be performed once, and subsequent simulations can be verified based on the data collected in the real vehicle verification, eliminating a large number of unnecessary repeated collections and also saving human resources.

[0036] Furthermore, in the map data verification method provided by the present application, it further includes:

[0037] Construct a first driving path according to the position information and a preset first coordinate refresh frequency, where the first driving path includes a plurality of coordinates determined based on the first coordinate refresh frequency;

[0038] Construct a second driving path according to the position information and a preset second coordinate refresh frequency, where the second driving path includes a plurality of coordinates determined based on the second coordinate refresh frequency, and the first coordinate refresh frequency is different from the second coordinate refresh frequency;

[0039] Perform a simulation test according to the first driving path and the map data to obtain the first simulation test data;

[0040] Perform a simulation test according to the second driving path and the map data to obtain the second simulation test data.

[0041] According to the above technical means, since the technical solution provided by the present application can also simulate the driving process at different vehicle speeds through different refresh frequencies, it is not necessary to perform repeated collections at different vehicle speeds during vehicle driving, reducing the real vehicle test volume and improving the map data verification efficiency.

[0042] In a second aspect, the present invention also provides a map data verification device, including:

[0043] A position information acquisition module, configured to acquire the position information of the vehicle when the user drives the vehicle;

[0044] A real vehicle scenario acquisition module, configured to acquire the real vehicle driving scenario of the vehicle through the visual sensor of the vehicle when acquiring the position information;

[0045] A first scenario construction module, configured to construct a first driving scenario according to the position information and a preset map data;

[0046] A first verification module, configured to compare the first driving scenario with the real vehicle driving scenario to obtain a first verification result of the map data.

[0047] In a third aspect, the present invention further provides a vehicle, comprising:

[0048] at least one processor; and,

[0049] a memory communicatively connected to the at least one processor; wherein,

[0050] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement the above-described map data verification method.

[0051] In a fourth aspect, the present invention further provides a computer-readable storage medium storing a computer program, which when executed by a processor can implement the above-described map data verification method.

[0052] Advantages of the present invention:

[0053] Since the map data verification method provided by the present application can compare the actual vehicle driving scenario collected by the sensors installed on the vehicle with the first driving scenario rendered based on the positioning information and map data during vehicle driving, thereby verifying the accuracy of the map data, avoiding the problems of the existing map data verification being divorced from the actual scenario, high cost, and low accuracy of map data verification.

[0054] Moreover, the map data verification method provided by the present application performs map data verification actions when the user drives the vehicle, and the testing and verification behaviors are synchronized with the vehicle driving. The map data at each moment is verified once, and only one complete vehicle driving process is required to verify the map data of the entire driving path, avoiding the repeated collection of map data in multiple actual vehicle tests and improving the efficiency of map data verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] One or more embodiments are exemplarily illustrated by the pictures in the corresponding drawings. These exemplary illustrations do not limit the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements, unless otherwise stated, and the drawings in the figures do not constitute a proportional limitation.

[0056] Figure 1 is one of the schematic diagrams of the map data verification method provided by the present application;

[0057] Figure 2 is an example of the communication architecture of the map data verification system provided by the present application;

[0058] Figure 3 is the second schematic diagram of the map data verification method provided by the present application;

[0059] Figure 4 is the third schematic diagram of the map data verification method provided by the present application;

[0060] Figure 5 It is the fourth schematic diagram of the map data verification method provided by this application;

[0061] Figure 6 It is the fifth schematic diagram of the map data verification method provided by this application;

[0062] Figure 7 It is the sixth schematic diagram of the map data verification method provided by this application;

[0063] Figure 8 It is the seventh schematic diagram of the map data verification method provided by this application;

[0064] Figure 9 It is an example of the map data verification process provided by this application;

[0065] Figure 10 It is the schematic diagram of the map data verification device provided by this application;

[0066] Figure 11 It is the schematic diagram of the structure of a vehicle provided by an embodiment of this application. Detailed implementation manners

[0067] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for explaining the present invention and not for limiting the protection scope of the present invention.

[0068] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and proportion of each component in its actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0069] With the continuous development and improvement of autonomous driving technology, many product functions around intelligent driving have been continuously installed in the vehicle environment and mass-produced. As one of the many functions, intelligent parking has high practicality and expandability, with good economic value and a lot of development space. With the continuous improvement of autonomous driving technology, the intelligent parking function has also achieved version iteration, from intelligent parking to valet parking scenario applications. Compared with traditional parking functions, the updated intelligent parking function can automatically select the optimal parking space and park automatically based on the existing parking lot map data. For example, an autonomous parking navigation method proposed in CN113506456A extracts driving information in real time through sensors such as surround-view cameras and millimeter-wave radars, and compares it with the high-precision map to achieve the positioning of the vehicle in the high-precision map of the parking lot, solving the problem of vehicle autonomous driving from the parking lot entrance to the parking lot. However, the feasibility of these intelligent parking methods and functions depends on the high-precision map. For example, the accuracy of the existing high-precision map of the parking lot. When the accuracy of the map data is poor, it is difficult to ensure the actual feasibility of the intelligent parking function.

[0070] To address the above problems, since the development process of the intelligent parking function not only requires a large number of tests to verify the feasibility and completeness of the function, but also the accuracy of the parking lot map data collected manually is crucial. Therefore, it is often necessary to verify the map data to ensure that the map data does not interfere with autonomous driving. For example, by building a simulation cloud platform for large-scale simulation tests. However, the development cost of this kind of simulation test is high, and a large number of scenario designs and constructions are required, and specific data needs to be constructed, with a large workload. For example, the high-precision map data verification device and method proposed in CN109189872A lack application in specific scenarios, such as the parking lot scenario. There is a gap between the verification data obtained by simulation and the actual scenario, such as the actual parking lot. And the on-road vehicle test is affected by factors such as a large amount of parking lot map data, and the uneven structure and quality of the parking lot. A large number of tests need to be carried out repeatedly, which is time-consuming, laborious, costly, and inefficient.

[0071] The technical solution provided in this application constructs a scenario during the on-road vehicle driving process according to the vehicle's position information and the pre-set map data, and compares the scenario construction result with the information collected by the vision sensor during the on-road vehicle driving process to achieve the verification effect of the map data, ensuring the correctness and reasonableness of the map data for autonomous driving, such as the parking lot map data for intelligent parking. In addition, this application can also perform subsequent simulation verification based on the data collected during the on-road vehicle driving process, eliminating a large number of unnecessary repeated collections and saving human resources at the same time.

[0072] Taking the intelligent parking scenario in a parking lot as an example, the technical solution provided by this application can perform data verification on the parking lot map data relied on for intelligent parking. Specifically, during the in-vehicle test of intelligent parking, when the vehicle is parking in the parking lot, this application can determine the position information of the vehicle in the parking lot according to the vehicle positioning system in the in-vehicle system, and construct the driving scenario of the vehicle in the parking lot at this time based on the position information of the vehicle in the parking lot and the parking lot map data pre-stored in the PC host computer. While the positioning system determines the current position of the vehicle, multiple sensors installed on the vehicle in the in-vehicle system, such as visual sensors like cameras, collect data on the actual scenario of the vehicle in the parking lot, and compare the scenario constructed based on the parking lot map data and the position information of the vehicle with the scenario collected by the sensors, so as to complete the verification action of the parking lot map data at the current position.

[0073] The first embodiment of this application relates to a method for verifying map data, as Figure 1 shown, including:

[0074] Step 101: When the user drives the vehicle, collect the position information of the vehicle;

[0075] Step 102: When collecting the position information, collect the in-vehicle driving scenario of the vehicle through the visual sensor of the vehicle;

[0076] Step 103: Construct a first driving scenario according to the position information and the pre-set map data;

[0077] Step 104: Compare the first driving scenario with the in-vehicle driving scenario to obtain a first verification result of the map data.

[0078] Specifically, in the map data verification method provided by this application, it is necessary to first build a map data verification system. As Figure 2As shown in the figure, the map data verification system includes a vehicle-mounted system, a lower computer, a map data cloud platform, and a personal computer (PC) upper computer. Among them, the PC upper computer is the core of the entire map data verification system, which is used to generate driving routes and load map data, calculate relevant data during vehicle driving according to a pre-set algorithm, and broadcast the calculation results to the lower computer. The map data exists in the cloud of the map data cloud platform in the form of a semantic map and a visual feature map. When the map data cloud platform receives a loading request sent by the PC upper computer, it transmits the map data to the PC upper computer. The vehicle-mounted system includes a positioning system and multiple sensors installed on the vehicle, such as visual sensors like cameras, which are used to provide positioning information through the positioning system when the vehicle is actually driving in the scene corresponding to the map data, so that the PC upper computer can calculate the position information of the vehicle, and collect the actual scene information of the current position of the vehicle through the sensors. The lower computer, such as a display screen, can visualize the driving scene determined based on the vehicle's current position information and map data and output parameter logs, so as to compare the simulated scene obtained based on the map data with the actual scene collected by the sensors set on the vehicle, thereby verifying the accuracy of the map data. Among them, the sensors and the positioning system in the vehicle-mounted system can synchronize data and send the actual scene during vehicle driving to the PC upper computer through the positioning system.

[0079] Taking the intelligent parking scenario in autonomous driving as an example, after pre - building communication connections among the in - vehicle system, the lower computer, the map data cloud platform, and the PC upper computer in the map data verification system. First, when the vehicle is driving in the parking lot, the PC upper computer sends a map loading request to the parking lot map data cloud platform and receives the parking lot semantic map feedback from the map data cloud platform. The positioning system in the in - vehicle system collects the positioning information of the vehicle in the parking lot at the current moment. For example, after collecting the Global Navigation Satellite System (GNSS) positioning signal, it sends the GNSS signal to the PC upper computer for the PC upper computer to determine the current position information of the vehicle in the parking lot based on the parking lot semantic map and the GNSS signal. While the positioning system in the in - vehicle system is collecting the GNSS signal, sensors such as cameras and radars installed on the vehicle in the in - vehicle system collect the parking lot scene around the vehicle at the current moment as the real - vehicle driving scene and send the real - vehicle driving scene to the lower computer. When the PC upper computer completes the determination of the vehicle's position information, it performs data rendering based on the position information and the parking lot semantic data to obtain the first driving scene for parking simulation based on the parking lot map data and sends the rendering result to the lower computer. The lower computer compares the first driving scene for parking simulation based on the parking lot map data with the real - vehicle driving scene collected by the sensors during the vehicle parking driving process to complete the verification action of the parking lot semantic map stored in the parking lot map data cloud platform.

[0080] Since the parking path of intelligent parking often consists of a sequence of multiple path coordinate points, a time threshold for map verification can be preset. Starting from when the user drives the vehicle, sensors such as cameras and millimeter - wave radars continuously collect the real - vehicle driving scene during the vehicle driving process. And whenever the driving time of the vehicle reaches the preset time threshold, the positioning information at the current moment is obtained and the position of the vehicle in the parking lot at the current moment is determined. Then, based on the position information and the pre - set map data, the driving scene at the current moment is constructed. The first driving scene corresponding to the current moment obtained by the construction is compared with the real - vehicle driving scene collected by the camera at the current moment to verify the accuracy of the map data at the current moment. This action is repeated until the verification of the map data corresponding to each moment on the entire parking path is completed, thus realizing the verification function of the parking lot map data. Among them, the setting of the time threshold can be determined by the user according to their own needs. For example, different time thresholds can be used to represent the verification of the parking lot map data at different parking speeds, or verification can also be carried out in ways such as driving distance. This application does not make restrictions.

[0081] Among them, the present application does not limit the specific method of map data verification. It can perform data similarity or similarity comparison between the driving data corresponding to the first driving scenario and the driving data corresponding to the actual vehicle driving scenario collected by the sensor, or perform similarity comparison according to the image data of the first driving scenario and the actual vehicle driving scenario. When the similarity is within the preset threshold range, it indicates that the map data has high accuracy; when the similarity is outside the preset threshold range, it indicates that the map data has low accuracy, so as to obtain the corresponding map data verification result.

[0082] In addition, the technical solution provided by the present application is not limited to the manual form of the user driving the vehicle. When the vehicle is controlled by the system to perform autonomous driving, the first driving scenario generated according to the positioning information and the map data can also be compared with the actual vehicle driving scenario while the vehicle is controlled by the in-vehicle system to perform autonomous driving, so as to obtain the verification result of the map data.

[0083] Since the map data verification method provided by the present application can compare the actual vehicle driving scenario collected by the sensors installed on the vehicle with the first driving scenario rendered based on the positioning information and the map data during vehicle driving, and then verify the accuracy of the map data, it avoids the problems of the existing map data verification being divorced from the actual scenario, high cost, and low accuracy of map data verification.

[0084] Moreover, the map data verification method provided by the present application performs the map data verification action when the user drives the vehicle. The test and verification behaviors are synchronized with the vehicle driving. The map data at each moment is verified once, and only one complete vehicle driving process is required to verify the map data of the entire driving path, avoiding the action of repeatedly collecting map data through multiple actual vehicle tests, and improving the efficiency of map data verification.

[0085] Based on the above embodiments, as Figure 3 shown, in the map data verification method provided by the present application, step 101 includes:

[0086] Step 111: When the user drives the vehicle, obtain the positioning signal of the vehicle through the positioning sensor;

[0087] Step 112: Calculate the coordinate system transformation according to the positioning signal and the map data to obtain the position information.

[0088] Specifically, in the map data verification method provided by this application, since the PC host computer has established network communication with the vehicle system in advance, the positioning system in the vehicle system can generate the driving path of the vehicle during driving in real time based on the positioning signal of the vehicle and the map positioning conversion matrix. The visual sensors such as the on-vehicle camera in the vehicle system can record the surrounding scenes on the driving path of the vehicle during driving in real time and determine the actual driving scene of the vehicle during driving based on these scenes. The actual parking path of the vehicle during driving and the synchronized driving scene data packet can both be transmitted through the network communication between the PC host computer and the vehicle system.

[0089] Since the coordinate system information in the positioning signal of the vehicle determined by the vehicle system is not consistent with that in the map data, it is necessary to synchronize the positioning information in the data format first and then calculate the relevant parameters. After the PC host computer obtains the positioning signal sent by the vehicle system, it performs coordinate system conversion based on the map data transmitted by the map data cloud platform. For example, in a parking lot scenario, the positioning signal transmitted by the vehicle is based on the global coordinate system, while the parking lot map data transmitted by the parking lot map data cloud platform is based on the relative coordinate system of the parking lot. Therefore, it is first necessary to perform coordinate system conversion on these two types of data to achieve data synchronization in format, so as to obtain the position information of the vehicle at the current moment.

[0090] Taking the intelligent parking scenario in autonomous driving as an example, when the vehicle conducts on-vehicle testing of the parking operation in the parking lot, first, the positioning system in the vehicle system, such as a positioning sensor, collects the positioning signal of the vehicle in the parking lot. However, this positioning signal is determined based on the global coordinate system, while the map data loaded by the parking lot map data cloud platform to the PC host computer is often only the parking lot map data, such as the parking lot semantic map data, and the position coordinates are often determined based on the relative coordinate system of the parking lot. Therefore, it is necessary to convert the positioning signal of the vehicle to calculate the position relationship between the vehicle and each map element in the parking lot at the current moment during the parking process, so as to determine the position information for constructing the scene.

[0091] Based on the above implementation manner, since this application can also perform coordinate system conversion according to the differences between the positioning signal and the map data, it can quickly determine the position information corresponding to the vehicle in the map data currently, improving the efficiency of map data verification.

[0092] Based on the above implementation manner, as Figure 4 shown, the map data includes a semantic map. In the map data verification method provided by this application, step 111 includes:

[0093] Step 113, when the user drives the vehicle, use the global satellite navigation system to guide and position the vehicle to obtain a first positioning result;

[0094] Step 114: Perform semantic positioning on the vehicle according to the semantic map to obtain a second positioning result;

[0095] Step 115: Perform positioning convergence processing based on the first positioning result and the second positioning result to obtain the positioning signal.

[0096] Specifically, in the map data verification method provided in this application, the map data includes a semantic map. First, it is necessary to perform initial positioning based on the GNSS signal provided by the in-vehicle system to obtain a first positioning result, and rely on the semantic map to perform semantic positioning on the current position of the vehicle to obtain a second positioning result. When the first positioning result and the second positioning result converge, for example, when the difference between the two positioning results is less than a preset value, it is determined that the positioning converges, and the positioning of the vehicle at the current moment is successful. Thus, the position information of the vehicle is determined according to the positioning signal fed back by the in-vehicle system at this time.

[0097] Taking the intelligent parking scenario in autonomous driving as an example, when the vehicle drives into the parking lot, first perform positioning tracking based on the initial positioning guided by the GNSS of the in-vehicle system, and perform semantic positioning according to the semantic map in the parking lot map data. When the two positioning results converge successfully, calculate and determine the position information corresponding to the vehicle in the parking lot map data at the current moment according to the GNSS signal fed back by the in-vehicle system at this time.

[0098] When the map data includes a semantic map, this application also provides an example of a specific first driving scenario:

[0099] After the PC host computer determines the position of the vehicle in the map data at the current moment through the semantic map and the positioning information, the PC host computer can calculate relevant data in the parking lot such as the number of lanes where the vehicle is located, the number of surrounding parking spaces, and whether there is a special area ahead based on the map element information included in the semantic map, such as lanes, parking spaces, road marking facilities, etc., and send the above data to the slave computer. After the slave computer receives the semantic map, the real vehicle positioning data, and the calculated parking data broadcast by the PC host computer, it simulates the parking scenario in the parking lot through a rendering tool and displays it in real time, and verifies the displayed first driving scenario with the data collected by sensors such as cameras installed on the vehicle, so as to judge the accuracy of the semantic map.

[0100] Among them, the rendering tool can be set on the slave computer, and the slave computer renders according to the calculation result of the PC host computer to generate the first driving scenario, or it can be set on the PC host computer. After the PC host computer renders, it sends the constructed first driving scenario to the slave computer for the slave computer to output and display. This application does not make any restrictions.

[0101] Based on the above embodiments, since the technical solution provided by this application can perform positioning convergence by integrating semantic maps and GNSS signals, it avoids the problem of poor positioning effects caused by weak inertial navigation signals in vehicle driving scenarios. For example, in an underground parking lot where the inertial navigation signal is weak and the accuracy of the position information determined based on GNSS signals is poor, it ensures the accuracy of the vehicle's current position information, and further ensures the accuracy of map data verification.

[0102] Based on the above embodiments, as Figure 5 shown, the map data further includes a visual feature map. In the map data verification method provided by this application, after step 114, it further includes:

[0103] Step 116: Perform visual simultaneous localization on the vehicle according to the visual feature map to obtain a third positioning result;

[0104] Step 117: Perform positioning assistance on the positioning signal according to the third positioning result to obtain the corrected positioning signal.

[0105] Specifically, in the map data verification method provided by this application, the map data further includes a visual feature map. After convergence according to GNSS signals and semantic maps, it can also perform visual simultaneous localization (Visual Simultaneous Localization And Mapping, VSLAM) on the vehicle at the current moment according to the visual feature map to obtain a third positioning result, and perform positioning assistance and correction on the positioning signal determined after convergence according to the third positioning result.

[0106] Taking the intelligent parking scenario in autonomous driving as an example, when the vehicle enters the parking lot, it first performs positioning tracking based on the initial positioning guided by the GNSS of the in-vehicle system and relies on the semantic map in the parking lot map data for semantic positioning. When the positioning convergence is successful, the VSLAM positioning in the visual feature map in the parking lot map data is selectively enabled to assist the positioning signal. Among them, the selective enabling of VSLAM positioning depends on the actual needs of the user, and this application does not make any restrictions.

[0107] Based on the above embodiments, since this application can also perform positioning assistance through the visual feature map, it avoids the problem of poor positioning effects caused by weak inertial navigation signals in vehicle driving scenarios. For example, in an underground parking lot where the inertial navigation signal is weak and the accuracy of the position information determined based on GNSS signals is poor, it ensures the accuracy of the vehicle's current position information, and further ensures the accuracy of map data verification.

[0108] Based on the above embodiments, as Figure 6As shown in the figure, in the map data verification method provided by this application, after step 102, it further includes:

[0109] Step 105: Perform a positioning judgment based on the position information and a pre-set map loading area to obtain a positioning judgment result;

[0110] Step 106: When the positioning judgment result is that the position information falls within the map loading area, construct a second driving scenario based on the position information and pre-set map data;

[0111] Step 107: Compare the second driving scenario with the actual vehicle driving scenario to obtain a second verification result of the map data.

[0112] Specifically, in the map data verification method provided by this application, before constructing the scenario framework, this application can also perform a positioning judgment based on the position of the vehicle. When the position information of the vehicle in the map data falls within the pre-set map loading area in the map data, perform the action of constructing a driving scenario based on the map data; when the position information of the vehicle in the map data falls outside the pre-set map loading area in the map data, do not perform the scenario construction action.

[0113] Taking the intelligent parking scenario in autonomous driving as an example, since there is a valet parking area in intelligent parking, that is, a judgment area for determining whether to activate the vehicle's automatic parking function. When the vehicle drives into this area, the user can leave the vehicle and send instructions through methods such as the car key, mobile terminal, and in-vehicle input screen. After receiving the corresponding instructions, the in-vehicle system can control the vehicle to drive autonomously to the end of the parking path to complete the parking action. Since valet parking is usually that the user drives the vehicle to a designated valet parking area, such as the elevator entrance or exit of a parking lot, etc., then gets out of the vehicle and leaves, and the corresponding system controls the vehicle to drive autonomously to the end of the parking path to achieve the valet parking action. Therefore, the verification of the parking lot map data can start from when the vehicle enters the valet parking area and end when the vehicle reaches the end of the parking path, that is, the pre-determined parking space. Therefore, before rendering the driving scenario of the parking lot map data and comparing the driving scenario with the actual scenario inside the parking lot collected by the camera, a positioning judgment can be performed first. When it is judged that the current position information of the vehicle falls within the map loading area, such as the valet parking area, then perform scenario rendering and verification of the parking lot map data; when it is judged that the current position information of the vehicle does not fall within the map loading area, such as the valet parking area, do not perform scenario rendering and data verification actions.

[0114] For example, in a parking lot scenario, an area within a radius of 500 meters around an electronic fence in the parking lot map data is set as the map loading area, such as the valet parking area. When the position information calculated by the PC host computer based on the map data and the positioning signal provided by the in-vehicle system is within this area, the PC host computer starts the scene construction action. Among them, the valet parking area can be pre-stored in the PC host computer. When it is detected that the vehicle is currently in the valet parking area, for example, the real-time obtained GSNN signal falls within the valet parking area, the PC host computer will send a map data loading request to the map data cloud platform and receive the map data. In addition, after the map data is loaded, the present application can also prompt the user to enter the valet parking area for map data verification.

[0115] Based on the above embodiments, since the present application can also perform positioning area judgment, it avoids continuous map data loading and driving scene rendering of the vehicle, reduces the data volume of map data verification, and improves the efficiency of map data verification.

[0116] Based on the above embodiments, as Figure 7 shown, in the map data verification method provided by the present application, after step 104, it further includes:

[0117] Step 108: Perform a simulation test according to the position information and the map data to obtain simulation test data;

[0118] Step 109: Construct a third driving scene according to the simulation test data and the map data;

[0119] Step 110: Compare the third driving scene with the actual vehicle driving scene to obtain a third verification result of the map data.

[0120] Specifically, in the map data verification method provided by the present application, after verifying the map data during vehicle driving, it can also compare the actual vehicle driving scene collected during actual vehicle driving with the simulation test scene, so as to complete the simulation verification of the map data.

[0121] Taking the intelligent parking scenario in autonomous driving as an example, after the in-vehicle test of intelligent parking and the verification of the parking lot map data, the technical solution provided by this application can also perform simulation tests based on the information collected during the in-vehicle test of the parking lot map data. For example, during the in-vehicle test, when the vehicle travels from the valet parking area to the end of the parking path in the parking lot, such as the surrounding environment of the parking space, it is collected through sensors such as cameras. During the simulation test, only the data collected by the sensors is needed to verify the map data. For subsequent verification of the parking lot map data, such as the verification of the parking path at different vehicle speeds during the vehicle parking process, it can be verified based on the data collected by the sensors, reducing the repeated collection actions of the in-vehicle test.

[0122] In addition, this application can also compare the scenario based on simulation with the scenario rendered based on the parking lot map data and vehicle positioning information, so as to verify the parking lot map data through simulation, reducing the time cost of verifying the parking lot map data.

[0123] It should be emphasized that the main difference between verifying the parking lot data through simulation tests and through in-vehicle tests lies in the difference in the process of obtaining the parking path in the parking lot. For simulation verification, based on the existing parking lot map data, the position of the vehicle itself and the target position are determined through a PC host computer. The target position is any parking space in the parking lot, and a parking path consisting of a sequence of the shortest path coordinate points relative to the entrance and exit positions is generated according to the set starting position and target position; while the parking path for in-vehicle measurement verification is to transmit the real-time parking path of the vehicle through the positioning system of the vehicle head unit and the parking lot map positioning transformation matrix, and the surrounding situations during parking are recorded in real time through sensors such as on-vehicle cameras. After the parking path is obtained through simulation tests, based on the parking path consisting of the coordinate point sequence relative to the parking lot entrance and exit, the driving of the vehicle in the corresponding parking lot is simulated through the change process of the sequential broadcast of the coordinate points. The PC host computer can simulate the driving process at different vehicle speeds by changing the refresh frequency of the coordinate points. In fact, each relative coordinate point is a simulation of the vehicle's positioning situation in the parking lot. Based on the simulated positioning information and combined with the map element information such as the parking lot entrance and exit, the passable area of the road, the road edge line, the parking space, and the planar marking line in the existing parking lot map data, various parameters required during the parking process are calculated, and the scenario simulation and the verification of the parking lot map data are carried out.

[0124] Based on the above embodiments, since the present application can also re-verify the map data after real vehicle testing through simulation testing, and can perform simple inspection and processing on the collected parking lot map data according to the simulation verification results, the time cost is greatly reduced. Moreover, the real vehicle verification only needs to be performed once, and subsequent simulations can be verified based on the data collected in the real vehicle verification, eliminating a large number of unnecessary repeated collections and saving human resources at the same time.

[0125] Based on the above embodiments, as Figure 8 shown, the simulation test data includes first simulation test data and second simulation test data. In the map data verification method provided by the present application, step 108 includes:

[0126] Step 181, constructing a first driving path according to the position information and a preset first coordinate refresh frequency, wherein the first driving path includes a plurality of coordinates determined based on the first coordinate refresh frequency;

[0127] Step 182, constructing a second driving path according to the position information and a preset second coordinate refresh frequency, wherein the second driving path includes a plurality of coordinates determined based on the second coordinate refresh frequency, and the first coordinate refresh frequency is different from the second coordinate refresh frequency;

[0128] Step 183, performing a simulation test according to the first driving path and the map data to obtain the first simulation test data;

[0129] Step 184, performing a simulation test according to the second driving path and the map data to obtain the second simulation test data.

[0130] Specifically, in the map data verification method provided by the present application, taking the scenario of parking in a parking lot as an example, the simulation verification can be implemented in the following manner:

[0131] Based on the existing parking lot map data, the position of the vehicle itself and the target position are determined by the PC host computer. The target position is any parking space in the parking lot. A parking path consisting of a sequence of shortest path coordinate points relative to the entrance and exit positions is generated according to the set starting position and target position. Subsequently, based on the parking path consisting of the coordinate point sequence relative to the parking lot entrance and exit, the driving of the vehicle in the corresponding parking lot is simulated through the changing process of the coordinate point broadcast in order. The PC host computer can simulate the driving process at different vehicle speeds by changing the refresh frequency of the coordinate points. Based on the simulated positioning information and combined with the map element information such as the parking lot entrance and exit, the passable area of the road, the road edge line, the parking space, and the planar marking in the existing parking lot map data, various parameters required during parking are calculated. Subsequently, according to the above parameters, through the semantic map covering the relevant information of each element of each layer of the parking lot, the third driving scenario is generated through the rendering platform or rendering tool, the scenario in the parking lot is reproduced, and the reproduced third driving scenario is compared with the actual vehicle driving scenario collected during vehicle driving to obtain the verification result of the map data.

[0132] Based on the above implementation manner, since the technical solution provided by the present application can also simulate the driving process at different vehicle speeds through different refresh frequencies, it is not necessary to repeatedly collect at different vehicle speeds during vehicle driving, reducing the actual vehicle test volume and improving the map data verification efficiency.

[0133] Based on the above implementation manner, as Figure 9 shown, the present application also provides an example of a map data verification process:

[0134] After the network communication between the PC host computer, the parking lot map data, the lower computer, and the in-vehicle unit is pre-constructed, before the vehicle enters the parking lot, for example, at the entrance of the parking lot, the in-vehicle unit positioning system transmits GNSS signals to the PC host computer in real time through the ZMQ bus, and the initial positioning is guided by GNSS until the initial positioning is completed. The PC host computer judges according to the received actual vehicle position coordinates. When the actual vehicle position is within 500 meters of the parking lot electronic fence buffer zone, it is judged that the vehicle has entered the map loading range, and a map loading request is sent to the cloud. After the vehicle enters the parking lot, the positioning position converges based on the loaded semantic map, enters the positioning tracking state, and transmits the positioning information to the PC host computer. The PC host computer calculates the relationship data between the vehicle and the parking lot map elements during the parking process according to the received positioning information and the parking lot map data, and transmits the above intelligent driving domain data, such as relationship data, semantic map, and other information, to the lower computer. The lower computer generates the rendering effect of the parking lot according to the received semantic map, and simulates the parking scenario according to the actual vehicle position, so as to generate the rendering picture of the parking scenario in the parking lot. While the vehicle is moving, sensors such as the in-vehicle camera transmit the real-time data collected during the parking process to the lower computer, and verify the parking lot map data by comparing the simulated parking scenario presented by the lower computer with the actual driving scenario.

[0135] The second embodiment of the present application relates to a map data verification device, as Figure 10 shown, including:

[0136] A position information acquisition module 201, configured to acquire the position information of the vehicle when the user drives the vehicle;

[0137] An actual vehicle scenario acquisition module 202, configured to acquire the actual vehicle driving scenario of the vehicle through the visual sensor of the vehicle when acquiring the position information;

[0138] A first scenario construction module 203, configured to construct a first driving scenario according to the position information and the pre-set map data;

[0139] A first verification module 204, configured to compare the first driving scenario with the actual vehicle driving scenario to obtain a first verification result of the map data.

[0140] Based on the above embodiment, in the map data verification device provided by the present application, the position information acquisition module 201 includes:

[0141] A positioning signal acquisition unit, configured to acquire the positioning signal of the vehicle through a positioning sensor when the user drives the vehicle;

[0142] A position information calculation unit for calculating coordinate system conversion based on the positioning signal and the map data to obtain the position information.

[0143] Based on the above embodiments, the map data includes a semantic map. In the map data verification device provided in the present application, the positioning signal acquisition unit includes:

[0144] A first positioning result determination subunit for guiding and positioning the vehicle through a global satellite navigation system when the user drives the vehicle to obtain a first positioning result;

[0145] A second positioning result determination subunit for performing semantic positioning on the vehicle according to the semantic map to obtain a second positioning result;

[0146] A positioning convergence subunit for performing positioning convergence processing based on the first positioning result and the second positioning result to obtain the positioning signal.

[0147] Based on the above embodiments, the map data further includes a visual feature map. In the map data verification device provided in the present application, the positioning signal acquisition unit further includes:

[0148] A second positioning result determination subunit for performing visual simultaneous localization on the vehicle according to the visual feature map to obtain a third positioning result;

[0149] A positioning signal correction subunit for performing positioning assistance on the positioning signal according to the third positioning result to obtain the corrected positioning signal.

[0150] Based on the above embodiments, the map data verification device provided in the present application further includes:

[0151] A positioning judgment module for performing a positioning judgment based on the position information and a pre-set map loading area to obtain a positioning judgment result;

[0152] A second scenario construction module for constructing a second driving scenario according to the position information and pre-set map data when the positioning judgment result is that the position information falls within the map loading area;

[0153] A second verification module for comparing the second driving scenario with the actual vehicle driving scenario to obtain a second verification result of the map data.

[0154] Based on the above embodiments, the map data verification device provided in the present application further includes:

[0155] A simulation test module for performing a simulation test based on the position information and the map data to obtain simulation test data;

[0156] A third scenario construction module, configured to construct a third driving scenario according to the simulation test data and the map data;

[0157] A third verification module, configured to compare the third driving scenario with the actual vehicle driving scenario to obtain a third verification result of the map data.

[0158] Based on the above implementation, in the map data verification device provided by the present application, the simulation test module includes:

[0159] A first path construction unit, configured to construct a first driving path according to the position information and a preset first coordinate refresh frequency, where the first driving path includes a plurality of coordinates determined based on the first coordinate refresh frequency;

[0160] A second path construction unit, configured to construct a second driving path according to the position information and a preset second coordinate refresh frequency, where the second driving path includes a plurality of coordinates determined based on the second coordinate refresh frequency, and the first coordinate refresh frequency is different from the second coordinate refresh frequency;

[0161] A first simulation test unit, configured to perform a simulation test according to the first driving path and the map data to obtain the first simulation test data;

[0162] A second simulation test unit, configured to perform a simulation test according to the second driving path and the map data to obtain the second simulation test data.

[0163] The third implementation of the present application relates to an electronic device, as Figure 11 shown, including:

[0164] At least one processor 301; and,

[0165] A memory 302 communicatively connected to the at least one processor 301; where

[0166] The memory 302 stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor 301 so that the at least one processor 301 can implement the map data verification method according to the first implementation of the present application.

[0167] Among them, the memory and the processor are connected in a bus manner. The bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power deployment circuits, etc., which are well known in the art, so they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be an element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor is transmitted on the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor.

[0168] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory can be used to store the data used by the processor when executing operations.

[0169] The fourth embodiment of the present application relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the map data verification method described in the first embodiment of the present application.

[0170] That is, those skilled in the art can understand that all or part of the steps of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a program. This program is stored in a storage medium and includes several instructions to enable a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, etc., which can store program codes.

[0171] After considering the specification and practicing the application disclosed herein, those skilled in the art will readily think of other embodiments of the present application. The present application aims to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include the common general knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0172] It should be understood that the present application is not limited to the precise structures already described and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A method for validating map data, characterized in that, The method includes: When the user is driving the vehicle, collecting the position information of the vehicle; When collecting the position information, collecting the actual driving scene of the vehicle through the visual sensor of the vehicle; Constructing a first driving scene according to the position information and pre-set map data; Comparing the first driving scene with the actual driving scene to obtain a first verification result of the map data.

2. The method according to claim 1, wherein The step of "When the user is driving the vehicle, collecting the position information of the vehicle" includes: When the user is driving the vehicle, obtaining the positioning signal of the vehicle through a positioning sensor; Calculating the coordinate transformation according to the positioning signal and the map data to obtain the position information.

3. The method according to claim 2, wherein The map data includes a semantic map. The step of "When the user is driving the vehicle, obtaining the positioning signal of the vehicle through a positioning sensor" includes: When the user is driving the vehicle, guiding and positioning the vehicle through a global satellite navigation system to obtain a first positioning result; Performing semantic positioning on the vehicle according to the semantic map to obtain a second positioning result; Performing positioning convergence processing on the first positioning result and the second positioning result to obtain the positioning signal.

4. The method according to claim 3, wherein The map data further includes a visual feature map. After obtaining the second positioning result by performing semantic positioning on the vehicle according to the semantic map, it further includes: Performing visual simultaneous localization on the vehicle according to the visual feature map to obtain a third positioning result; Performing positioning assistance on the positioning signal according to the third positioning result to obtain the corrected positioning signal.

5. The method according to claim 1, wherein After collecting the actual driving scene of the vehicle through the visual sensor of the vehicle when collecting the position information, it further includes: Performing positioning judgment according to the position information and a pre-set map loading area to obtain a positioning judgment result; When the positioning judgment result is that the position information falls within the map loading area, constructing a second driving scene according to the position information and pre-set map data; Comparing the second driving scene with the actual driving scene to obtain a second verification result of the map data.

6. The method according to claim 1, characterized in that After comparing the first driving scene with the actual driving scene to obtain a first verification result of the map data, it further includes: Performing a simulation test according to the position information and the map data to obtain simulation test data; Constructing a third driving scene according to the simulation test data and the map data; Comparing the third driving scene with the actual driving scene to obtain a third verification result of the map data.

7. The method according to claim 6, characterized in that, The simulation test data includes first simulation test data and second simulation test data. The step of "Performing a simulation test according to the position information and the map data to obtain simulation test data" includes: Constructing a first driving path according to the position information and a pre-set first coordinate refresh frequency, where the first driving path includes multiple coordinates determined based on the first coordinate refresh frequency; Construct a second driving path based on the position information and a pre-set second coordinate refresh frequency, wherein the second driving path includes a plurality of coordinates determined based on the second coordinate refresh frequency, and the first coordinate refresh frequency is different from the second coordinate refresh frequency; Perform a simulation test based on the first driving path and the map data to obtain the first simulation test data; Perform a simulation test based on the second driving path and the map data to obtain the second simulation test data.

8. A map data verification device, characterized in that, Comprising: A position information acquisition module, configured to acquire the position information of the vehicle when the user is driving the vehicle; A real vehicle scenario acquisition module, configured to acquire the real vehicle driving scenario of the vehicle through the visual sensor of the vehicle when acquiring the position information; A first scenario construction module, configured to construct a first driving scenario based on the position information and pre-set map data; A first verification module, configured to compare the first driving scenario with the real vehicle driving scenario to obtain a first verification result of the map data.

9. A vehicle, characterized in that, Comprising: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement the map data verification method according to any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the map data verification method according to any one of claims 1-7.

Images