Uninterrupted memory parking map construction methods, electronic devices and products

By conducting quality assessments and monitoring interruption status of memory parking map data, the problem of easy interruption in traditional memory parking map construction has been solved. It enables automatic continuation of construction after interruption, improving the success rate and reliability of map construction and reducing user operation costs.

CN123115290APending Publication Date: 2026-11-17ZHEJIANG GEELY HLDG GRP CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611370710.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-04
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Traditional memory-based parking map construction methods are easily interrupted by complex factors, resulting in a low success rate of map construction, increasing user operation and time costs, and affecting the user experience.

Method used

By evaluating the quality of the real-time recorded parking map data, storing high-quality data, monitoring interruption status, and merging high-quality new recorded data with old data after an interruption, a complete map is generated.

Benefits of technology

It improves the success rate and reliability of memory parking map construction, reduces user operation costs, enables automatic continuation of construction after interruption, and enhances user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123115290A_ABST
    Figure CN123115290A_ABST
Patent Text Reader

Abstract

This application provides an uninterrupted memory parking map construction method, electronic device, and product, relating to the field of vehicle-assisted driving technology. In response to the vehicle initiating memory parking map construction, the method performs a quality assessment on first memory parking map data recorded in real-time by the vehicle, obtaining a first quality assessment result; stores the first memory parking map data based on the first quality assessment result; and monitors the interruption status of the vehicle's memory parking map construction based on the first quality assessment result. Based on the interruption status, it merges second memory parking map data with the first memory parking map data to obtain a complete memory parking map. This application solves the problem that traditional memory parking map construction schemes cannot meet users' needs for efficient and convenient parking map construction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of vehicle driver assistance technology, and in particular to an uninterrupted memory parking map construction method, electronic device and product. Background Technology

[0002] Currently, memory parking has gradually become one of the core intelligent configurations of high-end smart cars, and is an important branch of advanced driver assistance technology. Memory parking allows the vehicle to collect environmental perception data along the target parking path under user guidance, autonomously learn and record spatial information of commonly used parking scenarios, and generate a high-precision parking map for the corresponding scenario. In subsequent use, the vehicle can automatically complete path planning and parking operations based on the constructed parking map, without requiring full user control, effectively reducing the difficulty of parking operations and significantly improving parking efficiency and driving convenience. It is an important manifestation of the advanced parking capabilities of smart cars. However, in the actual construction process of memory parking maps, the map recording process is easily interrupted by multiple complex factors. Once recording is interrupted, the user must restart the entire recording process, which not only significantly increases the user's operational and time costs, but also directly leads to a low success rate of memory parking map construction, thus seriously affecting the actual user experience. Summary of the Invention

[0003] The main objective of this application is to propose an uninterrupted memory parking map construction method, electronic device, and product, aiming to solve the problem that traditional memory parking map construction schemes cannot meet users' needs for efficient and convenient parking map construction.

[0004] To achieve the above objectives, a first aspect of this application proposes an uninterrupted memory parking map construction method, the method comprising: In response to the vehicle initiating the memory parking map construction, the quality of the first memory parking map data recorded in real time by the vehicle is evaluated to obtain the first quality evaluation result; The first memory parking map data is stored based on the first quality assessment result, and the interruption status of the vehicle's memory parking map construction is monitored based on the first quality assessment result. Based on the interruption state, the second memory parking map data is fused with the first memory parking map data to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0005] In some embodiments, the quality assessment of the first memory parking map data recorded in real time by the vehicle to obtain a first quality assessment result includes: Acquire multimodal sensor data when the vehicle records the first memory parking map data in real time; The reliability of the first memory parking map data is quantified based on the multimodal sensor data to obtain a first confidence score for the first memory parking map data; the first quality assessment result includes the first confidence score.

[0006] In some embodiments, when the first quality assessment result includes a first confidence score, storing the first memory parking map data based on the first quality assessment result includes: In response to the first confidence score being greater than the first score threshold, the first memory parking map data is marked as high-confidence map data and stored. In response to the first confidence score being less than or equal to a first score threshold and greater than or equal to a second score threshold, the first memory parking map data is marked as medium confidence map data and stored.

[0007] In some embodiments, monitoring the interruption status of the vehicle's memory parking map construction based on the first quality assessment result includes: In response to the first confidence score being less than the second score threshold, the storage of the first memory parking map data is paused and the real-time status information of the vehicle is obtained; the real-time status information includes the real-time pose information and sensor status information of the vehicle; The real-time pose information and the sensor status information are determined as the interruption state of the vehicle's construction of a memory parking map.

[0008] In some embodiments, when the interruption state is detected, the method further includes: In response to the vehicle restarting the memory parking map construction, dead reckoning is performed on the vehicle based on the vehicle's real-time sensor data to obtain reset guidance information; In response to the vehicle driving according to the reset guidance information, the quality of the second memory parking map data recorded in real time by the vehicle is evaluated to obtain a second quality evaluation result.

[0009] In some embodiments, when the interruption state is detected, the method further includes: Obtain the replanning path between the vehicle's current location and a preset trusted region; the trusted region is a geographical area where the vehicle's environmental perception performance meets a preset quality threshold. In response to the vehicle restarting the memory parking map construction and driving to the trusted area according to the replanning path, the quality of the second memory parking map data recorded by the vehicle in real time is evaluated to obtain a second quality evaluation result.

[0010] In some embodiments, fusing the second memory parking map data with the first memory parking map data based on the interruption state to obtain a complete memory parking map includes: The target pose information of the vehicle when recording the second memory parking map data in real time is compared with the real-time pose state information of the vehicle in the interrupted state to obtain the comparison result. In response to the comparison result indicating that the target pose information is the same as the real-time pose information, the second memory parking map data is stitched together with the stored first memory parking map data to obtain a complete memory parking map.

[0011] In some embodiments, the method further includes: In response to the vehicle restarting the memory parking map construction, the vehicle's pose information relative to the stored first memory parking map data is calculated; Based on the pose information, the second memory parking map data is fused with the first memory parking map data to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0012] To achieve the above objectives, a second aspect of this application proposes an uninterrupted memory parking map construction system, the system comprising: The quality assessment module is used to assess the quality of the first memory parking map data recorded in real time by the vehicle in response to the vehicle starting memory parking map construction, and to obtain the first quality assessment result. The storage module is used to store the first memory parking map data based on the first quality assessment result; An interruption monitoring module is used to monitor the interruption status of the vehicle's memory parking map construction based on the first quality assessment result; The intelligent continuation recording module is used to merge the second memory parking map data with the first memory parking map data based on the interruption state to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the interrupt-resistant memory parking map construction method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the interrupt-resistant memory parking map construction method described in the first aspect.

[0015] To achieve the above objectives, a fifth aspect of the present application provides a computer program product, which includes a computer program that, when executed by a processor, implements the interrupt-resistant memory parking map construction method provided in the first aspect above.

[0016] The uninterrupted memory parking map construction method, system, electronic device, computer-readable storage medium, and computer program product proposed in this application, in response to a vehicle initiating memory parking map construction, performs a quality assessment on first memory parking map data recorded in real time by the vehicle to obtain a first quality assessment result; stores the first memory parking map data based on the first quality assessment result; and monitors the interruption status of the vehicle's memory parking map construction based on the first quality assessment result; and fuses second memory parking map data with the first memory parking map data based on the interruption status to obtain a complete memory parking map; the second memory parking map data is memory parking map data recorded in real time after the vehicle restarts memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0017] Compared to traditional methods for constructing memory parking maps, this embodiment performs a quality assessment on the first memory parking map data recorded in real time when the vehicle starts constructing the memory parking map, obtaining a first quality assessment result. Then, based on this first quality assessment result, the first memory parking map data is stored, and the interruption status of the vehicle's memory parking map construction is monitored based on the first quality assessment result. Subsequently, when the vehicle restarts constructing the memory parking map, if the second quality assessment result of the second memory parking map data recorded in real time after the vehicle restarts construction is higher than or equal to the first quality assessment result, the second memory parking map data is fused with the first memory parking map data based on the previously monitored interruption status, thereby obtaining a complete memory parking map.

[0018] Thus, this embodiment of the application can monitor the interruption status of the vehicle's memory parking map construction and, based on this interruption status, fuse the memory parking map data recorded in real time after the vehicle restarts memory parking map construction with the previously recorded and stored memory parking map data. This ensures that the memory parking map is restored at the interruption point, improving the reliability of the finally constructed memory parking map. In other words, this embodiment of the application can achieve uninterrupted memory parking map construction, automatically resuming after an interruption, thus eliminating the need for the user to restart the entire recording process. This improves the map construction success rate and reduces user operating costs, meeting users' needs for efficient and convenient parking map construction. Attached Figure Description

[0019] Figure 1 A flowchart illustrating the steps of the anti-interruption memory parking map construction method provided in some embodiments of this application; Figure 2 for Figure 1 A detailed flowchart of step S101; Figure 3 for Figure 1 A detailed flowchart of step S102; Figure 4 for Figure 1 A schematic diagram of another detailed step in step S102; Figure 5 A flowchart illustrating the steps of the anti-interruption memory parking map construction method provided in this application in some other embodiments; Figure 6 A flowchart illustrating the steps of the anti-interruption memory parking map construction method provided in some embodiments of this application; Figure 7 for Figure 1A detailed flowchart of step S103; Figure 8 A flowchart illustrating the steps of the anti-interruption memory parking map construction method provided in some embodiments of this application; Figure 9 A schematic diagram of the complete steps involved in one embodiment of the anti-interruption memory parking map construction method provided in this application; Figure 10 A schematic diagram of the structure of the anti-interruption memory parking map construction system provided in the embodiments of this application; Figure 11 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] It should be noted that although functional modules are divided in the device / system schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device / system or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0023] First, a brief explanation of the overall concept of the anti-interruption memory parking map construction method provided in the embodiments of this application will be given.

[0024] With the rapid development of automotive intelligent technology, Advanced Driver Assistance Systems (ADAS) have become one of the core competitive advantages of modern vehicles. Home-Zone Parking (HZP), as an important function of ADAS, aims to solve the parking problem for users in the last leg of their daily commute (such as the "last mile" from the community entrance to the fixed parking space). Its core value lies in improving the convenience and comfort of users.

[0025] However, real-world application scenarios are extremely complex. During the recording and reproduction of parking paths, vehicles are highly susceptible to various unpredictable external interferences and interruptions. These factors can not only disrupt the current mapping or parking task, but may also cause traditional systems based on the assumption of a perfect path in a single instance to construct incomplete, inaccurate, or even erroneous memory maps, leading to functional failures or safety risks during subsequent automatic parking. Therefore, how to enable memory parking systems to have strong anti-interruption capabilities, ensuring that they can robustly and safely complete map construction and path reproduction in complex and non-ideal real-world environments, has become a key technical bottleneck that academia and industry urgently need to overcome.

[0026] In related technologies, the technical approach of memory parking solutions mainly relies on the combination of multimodal sensor fusion and high-precision positioning technology. For example, the vehicle is first driven by the driver or activates the autonomous learning function to drive the target parking path once. During this process, the system collects surrounding environmental data in real time through various sensors deployed on the vehicle body. At the same time, the inertial measurement unit provides high-frequency vehicle acceleration and angular velocity information to calculate relative displacement over a short period of time, while real-time dynamic differential positioning provides the vehicle with absolute centimeter-level global pose in areas with good signal, serving as the spatiotemporal reference for the entire map. All these heterogeneous data are fed into a centralized fusion perception algorithm for processing, ultimately generating a vectorized memory map composed of a series of high-precision pose points and environmental feature points.

[0027] However, while traditional memory parking solutions perform exceptionally well under ideal conditions of high structure, static environment, and stable sensor signals, their inherent limitations become glaringly apparent when faced with the need for "interruption resistance." First, they heavily rely on continuous, high-precision signals from the Global Positioning System (GPS) / Real-Time Kinematic (RTK). In scenarios like underground parking garages, the loss of the positioning source leads to rapid global positioning drift, rendering the entire map's baseline ineffective. Second, visual and lidar sensors are susceptible to adverse weather conditions and sudden changes in lighting, causing misidentification or loss of environmental features and disrupting the map building process. Most critically, traditional memory parking solutions largely rely on the assumption of "successful recording on the first attempt." Their mapping process is a fragile sequence; any interruption will cause the recording task to fail, and the system lacks mechanisms for sensing, recording, and automatically recovering from or continuing from interrupted events. Therefore, developing an interruption-resistant memory parking map building method capable of sensing interruptions, adaptively adjusting, and possessing strong recovery capabilities is crucial for promoting the large-scale application of this technology.

[0028] In summary, traditional memory parking solutions have significant drawbacks in practical applications. First, poor anti-interference capabilities lead to low map building reliability: in complex environments, vehicle memory parking systems are susceptible to GPS signal loss, sensor performance degradation, or interference from dynamic obstacles, causing map building interruptions or failures, requiring users to re-record the route and degrading the user experience. Second, insufficient map accuracy and robustness affect parking safety: the constructed maps are often incomplete or inaccurate, potentially causing collision risks during automatic parking. Finally, low cost and time efficiency: the system requires successful route recording on the first attempt; any interruption leads to repetitive operations, increasing user time costs and system energy consumption.

[0029] To address the aforementioned issues, this application proposes an uninterrupted memory parking map construction method, system, electronic device, computer-readable storage medium, and computer program product. The aim is to solve the problem that traditional memory parking map construction schemes cannot meet users' needs for efficient and convenient parking map construction, improve the uninterrupted capability and robustness of the memory parking system, and ensure reliable map construction and path reproduction in complex and dynamic environments.

[0030] This application embodiment responds to the vehicle initiating memory parking map construction by performing a quality assessment on the first memory parking map data recorded in real time by the vehicle to obtain a first quality assessment result; storing the first memory parking map data based on the first quality assessment result; and monitoring the interruption status of the vehicle's memory parking map construction based on the first quality assessment result; and fusing the second memory parking map data with the first memory parking map data based on the interruption status to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0031] Compared to traditional methods for constructing memory parking maps, this embodiment performs a quality assessment on the first memory parking map data recorded in real time when the vehicle starts constructing the memory parking map, obtaining a first quality assessment result. Then, based on this first quality assessment result, the first memory parking map data is stored, and the interruption status of the vehicle's memory parking map construction is monitored based on the first quality assessment result. Subsequently, when the vehicle restarts constructing the memory parking map, if the second quality assessment result of the second memory parking map data recorded in real time after the vehicle restarts construction is higher than or equal to the first quality assessment result, the second memory parking map data is fused with the first memory parking map data based on the previously monitored interruption status, thereby obtaining a complete memory parking map.

[0032] Thus, this embodiment of the application can monitor the interruption status of the vehicle's memory parking map construction and, based on this interruption status, fuse the memory parking map data recorded in real time after the vehicle restarts memory parking map construction with the previously recorded and stored memory parking map data. This ensures that the memory parking map is restored at the interruption point, improving the reliability of the finally constructed memory parking map. In other words, this embodiment of the application can achieve uninterrupted memory parking map construction, automatically resuming after an interruption, thus eliminating the need for the user to restart the entire recording process. This improves the map construction success rate and reduces user operating costs, meeting users' needs for efficient and convenient parking map construction.

[0033] Next, the uninterrupted memory parking map construction method, system, electronic device, computer-readable storage medium, and computer program product provided in this application will be specifically described through the following embodiments, and the various detailed embodiments of the uninterrupted memory parking map construction method provided in this application will be described in detail first.

[0034] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0035] It should be noted that the uninterrupted memory parking map construction method provided in this application relates to the field of vehicle assisted driving technology. The uninterrupted memory parking map construction method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be an in-vehicle terminal, a smartphone, tablet, laptop, desktop computer, or other electronic device associated with a vehicle and capable of communicating and interacting with the vehicle via a network. The server can be a backend server terminal device of the terminal, which can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The software can be an application implementing the uninterrupted memory parking map construction method, a computer program, and a storage medium carrying the computer program. It should be understood that, based on different design needs of practical applications, the terminal, server, and software of the anti-interruption memory parking map construction method provided in this application embodiment may also be other forms not listed here, depending on the different feasible embodiments. The anti-interruption memory parking map construction method provided in this application embodiment does not specifically limit these.

[0036] Furthermore, this application can also be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: vehicle terminals, personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, personal computers (PCs), minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via communication networks. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0037] For ease of understanding and explanation, the following text will use the anti-interruption memory parking map construction method provided in the embodiments of this application as an example to describe the various specific embodiments of this application in detail. In some descriptions, the terminal device may be simply referred to as the terminal. The implementation of the anti-interruption memory parking map construction method provided in the embodiments of this application for any of the above-described subject matter can refer to the implementation process of the anti-interruption memory parking map construction method described below.

[0038] Please refer to Figure 1 , Figure 1 The flowchart illustrates the steps of the anti-interruption memory parking map construction method provided in some embodiments of this application. It should be understood that, although... Figure 1 The flowcharts illustrating subsequent steps show the execution order of some method steps. However, based on different design needs in practical applications, the uninterrupted memory parking map construction method provided in this application embodiment can, of course, employ a different execution order of method steps than shown in the figures. That is, Figure 1 The order of the method steps shown does not constitute a limitation on the execution logic order of the uninterrupted memory parking map construction method provided in the embodiments of this application. Any other method based on this method is not limited to this method. Figure 1 Reasonable changes to the sequence of steps shown should be included within the protection scope of the anti-interruption memory parking map construction method provided in the embodiments of this application.

[0039] like Figure 1 As shown, in some embodiments, the uninterrupted memory parking map construction method provided in this application may include steps S101 to S103 as shown below.

[0040] Step S101: In response to the vehicle starting memory parking map construction, perform quality assessment on the first memory parking map data recorded in real time by the vehicle to obtain the first quality assessment result.

[0041] It should be noted that the vehicle's initiation of memory parking map construction can be interpreted as the vehicle starting memory parking map recording. "Start recording" refers to the time when the vehicle enters a state where the memory parking function can be activated, and the relevant "start recording" button is clicked. Memory parking map recording refers to the process during which the vehicle, using its own cameras, radar, and other sensors, scans, records, and digitally constructs the environmental features surrounding the driving map in real time, including lane lines, speed bumps, and ramps. In some embodiments, in the environmental perception module used by the vehicle to record the first memory parking map data, the camera can use a multi-view stereo vision or multispectral solution instead of a monocular visible light camera to improve ranging capabilities or perception reliability in adverse weather conditions; the lidar can be replaced with a solid-state LiDAR, 4D millimeter-wave radar, or a ToF camera to optimize cost, size, or weather interference resistance; the IMU module can perform dead reckoning through multi-IMU array fusion or by combining wheel speed pulse signals to improve the redundancy of short-term pose prediction; the positioning module can use UWB indoor base stations, geomagnetic / Bluetooth beacon-assisted positioning, or a pure visual SLAM solution to address scenarios where GPS signals are missing. In addition, in some embodiments, the terminal device can perform quality assessment of the memorized parking map data through a GPU, FPGA, or cloud-based collaborative computing architecture, as well as perform subsequent parking map storage and interruption status monitoring and map continuation operations.

[0042] After the vehicle starts building the memory parking map, the terminal device immediately performs a quality assessment on the first memory parking map data recorded in real time by the vehicle, and obtains the first quality assessment result.

[0043] In some embodiments, after the vehicle completes real-time path recording of the target parking scenario (e.g., generating first memory parking map data through simultaneous localization and mapping (SLAM) fusion of surround-view vision and inertial measurement unit (IMU)), the terminal device can immediately perform an automated quality assessment on the first memory parking map data to obtain a first quality assessment result.

[0044] In some embodiments, when the terminal device performs quality assessment on the first memory parking map data recorded in real time by the vehicle, it can perform position accuracy verification, semantic integrity detection, and topology consistency verification on the first memory parking map data, and finally generate a first quality assessment result containing the overall quality level, abnormal area location, and defect category. Specifically, position accuracy verification involves extracting the coordinates of static key features such as parking space corners, curbs, and pillars within the map, comparing them with the true values ​​calculated from the mapping reference pose, calculating the root mean square error of the plane, and marking abnormal areas with deviations exceeding a 5cm threshold; semantic integrity detection involves statistically analyzing the detection completeness of parking semantic elements such as parking lines, ground markings, and speed bumps, and identifying feature breaks and missed segments; topology consistency verification involves verifying whether the connectivity relationships of parking path nodes and parking space entrances / exits match the actual traffic constraints.

[0045] Step S102: Store the first memory parking map data based on the first quality assessment result, and monitor the interruption status of the vehicle's memory parking map construction based on the first quality assessment result.

[0046] After obtaining the first quality assessment result, the terminal device further stores the first memory parking map data based on the first quality assessment result. Furthermore, while storing the first memory parking map data based on the first quality assessment result, the terminal device can also simultaneously monitor whether the vehicle's memory parking map construction is interrupted, and if an interruption is detected, record the interruption status of the vehicle's memory parking map construction.

[0047] In some embodiments, when storing first memory parking map data based on a first quality assessment result, the terminal device may store the first memory parking map data locally only if the first quality assessment result indicates that the first memory parking map data meets quality requirements. These quality requirements may include the absence of abnormal areas and feature breaks / missed detection segments in the first memory parking map data, and the matching of the connectivity relationships between parking path nodes and parking space entrances / exits with actual traffic constraints.

[0048] In some embodiments, when the terminal device stores the first memory parking map data based on the first quality assessment result, the memory used to store the first memory parking map data may be an eMMC / UFS embedded storage or a cloud storage solution.

[0049] In some embodiments, when the terminal device monitors the interruption status of the vehicle's memory parking map construction based on the first quality assessment result, it may confirm that the vehicle's current memory parking map construction is interrupted if the first quality assessment result indicates that the first memory parking map data does not meet the quality requirements. Furthermore, it may record the vehicle's pose information and sensor state snapshots as the interruption status of the vehicle's current memory parking map construction.

[0050] Step S103: Based on the interruption state, the second memory parking map data and the first memory parking map data are fused to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0051] After the terminal device detects an interruption in the construction of the memory parking map, when the vehicle restarts the construction of the memory parking map, if the second quality assessment result of the second memory parking map data recorded in real time after the vehicle restarts the construction of the memory parking map is higher than or the same as the first quality assessment result, the second memory parking map data is fused with the first memory parking map data based on the previously monitored interruption state to obtain a complete memory parking map.

[0052] In some embodiments, the terminal device may respond to the vehicle restarting the memory parking map construction by performing a quality assessment on the second memory parking map data recorded in real time by the vehicle to obtain a second quality assessment result. Then, if the second quality assessment result is higher than the first quality assessment result of the already stored first memory parking map data or is the same as the first quality assessment result, the second memory parking map data is stitched and fused with the first memory parking map data based on the previously monitored and recorded interruption status to obtain a complete memory parking map.

[0053] In this embodiment, after the vehicle initiates the construction of the memory parking map, the terminal device immediately performs a quality assessment on the first memory parking map data recorded in real time by the vehicle, obtaining a first quality assessment result. Then, the first memory parking map data is further stored based on this first quality assessment result. Furthermore, while storing the first memory parking map data based on the first quality assessment result, the terminal device can also simultaneously monitor whether the vehicle's memory parking map construction is interrupted, and record the interruption status if an interruption is detected. Finally, when the vehicle restarts memory parking map construction, if the second quality assessment result of the second memory parking map data recorded in real time after the vehicle restarts memory parking map construction is higher than or equal to the first quality assessment result, the second memory parking map data is fused with the first memory parking map data based on the previously monitored interruption status, thereby obtaining a complete memory parking map.

[0054] Therefore, compared to traditional methods of building memory parking maps, this embodiment monitors the interruption status of vehicle memory parking map construction and merges the memory parking map data recorded in real time after the vehicle restarts memory parking map construction with the previously recorded and stored memory parking map data. This ensures that the memory parking map is restored at the interruption point, improving the reliability of the final completed memory parking map. In other words, this embodiment can achieve uninterrupted memory parking map construction, automatically resuming after an interruption, thus eliminating the need for the user to restart the entire recording process. This improves the map construction success rate and reduces user operating costs, meeting users' needs for efficient and convenient parking map construction.

[0055] Please refer to Figure 2 , Figure 2 for Figure 1 A detailed flowchart of step S101.

[0056] like Figure 2 As shown, in some embodiments, the step of “performing a quality assessment on the first memory parking map data recorded in real time by the vehicle and obtaining a first quality assessment result” in step S101 above may include steps S201 and S202 as shown below.

[0057] Step S201: Obtain multimodal sensor data when the vehicle records the first memory parking map data in real time.

[0058] It should be noted that multimodal sensor data can come from various sensors on the vehicle, such as cameras, LiDAR, millimeter-wave radar, ultrasonic sensors, and inertial measurement units. This data can include visual information, point cloud data, motion status, and so on. The terminal device can fuse multimodal sensor data (i.e., integrate data from multiple sensors) to comprehensively evaluate the reliability and accuracy of the map segment currently recorded by the vehicle, thereby achieving a quality assessment of the first memory parking map data and obtaining a first quality assessment result.

[0059] When the terminal device performs a quality assessment on the first memory parking map data recorded in real time by the vehicle, it can first obtain the multimodal sensor data collected by the vehicle when the first memory parking map data was recorded in real time, and then perform a quality assessment on the first memory parking map data.

[0060] In some embodiments, the terminal device can simultaneously collect multimodal sensor data such as visual information, point cloud data, and motion status from various sensors, including cameras, lidar, millimeter-wave radar, ultrasonic sensors, and inertial measurement units, while the vehicle is recording the first memory parking map data in real time.

[0061] Step S202: Based on the multimodal sensor data, perform reliability quantification processing on the first memory parking map data to obtain a first confidence score of the first memory parking map data; the first quality assessment result includes the first confidence score.

[0062] After acquiring multimodal sensor data during the real-time recording of the first memory parking map data by the vehicle, the terminal device can further perform reliability quantification processing on the first memory parking map data based on this multimodal sensor data, thereby obtaining a first confidence score for the first memory parking map data. In this way, the terminal device can use this first confidence score as the first quality assessment result obtained from the quality assessment of the first memory parking map data.

[0063] In some embodiments, when the terminal device performs reliability quantification processing on the first memory parking map data based on multimodal sensor data, it can use a sensor fusion algorithm to perform spatiotemporal alignment and feature extraction on the multi-source heterogeneous multimodal sensor data to obtain the environmental features when the vehicle records the first memory parking map data in real time. Based on a deep learning model or probabilistic fusion method, it calculates the stability of the environmental features, the consistency between sensors, and the positioning accuracy, thereby outputting a confidence score in the range of 0 to 1. This confidence score is the first confidence score, which serves as a key basis for determining whether the first memory parking map data meets the relevant quality requirements for map recording in subsequent steps.

[0064] In some embodiments, the terminal device may also perform reliability quantification processing on the first memory parking map data based on a machine learning model, a Bayesian filtering framework, or a rule base expert system, thereby obtaining a first confidence score for the first memory parking map data.

[0065] Please refer to Figure 3 , Figure 3 for Figure 1 A detailed flowchart of step S102.

[0066] like Figure 3 As shown, in some embodiments, when the first quality assessment result includes a first confidence score, the step of "storing the first memory parking map data based on the first quality assessment result" in step S102 above may include steps S301 and S302 as shown below.

[0067] Step S301: In response to the first confidence score being greater than the first score threshold, the first memory parking map data is marked as high confidence map data and stored.

[0068] It should be noted that the first score threshold can be a reliability threshold preset by the terminal device (the specific value can be flexibly set according to actual usage needs), used to determine whether the map segments recorded by the vehicle have sufficient accuracy and robustness for subsequent positioning and navigation.

[0069] After calculating the first confidence score and using it as the first quality assessment result for the first memory parking map data, the terminal device can further compare the first confidence score with the first score threshold in real time. If the first confidence score is higher than the first score threshold, it can be determined that the first memory parking map data has sufficient accuracy and robustness to be used for subsequent positioning and navigation. Thus, the first memory parking map data is marked as high-confidence map data and stored in the local storage space.

[0070] In some embodiments, the terminal device compares a first confidence score with a first score threshold in real time. When the first confidence score exceeds the first score threshold, the terminal device triggers a map saving process via the domain controller. This process compresses and persistently stores the first memory parking map data (sensor data within the current time window, including keyframe images, point clouds, feature points, pose information, etc.) in the vehicle's storage unit as a valid fragment of a high-precision map for subsequent map planning and vehicle control. Simultaneously, the terminal device can also generate corresponding map fragment indexes and metadata to facilitate subsequent retrieval and stitching.

[0071] In some embodiments, when the terminal device persistently stores the first memory parking map data, it may mark the first memory parking map data as high confidence map data after the first confidence score is higher than the first score threshold, and then persistently store it in the first area of ​​the local storage space.

[0072] Step S302: In response to the first confidence score being less than or equal to the first score threshold and greater than or equal to the second score threshold, the first memory parking map data is marked as medium confidence map data and stored.

[0073] It should be noted that the second score threshold can be a minimum acceptable threshold value preset by the terminal device (the specific value can be flexibly set according to actual usage needs). The second score threshold is less than the first score threshold mentioned above, and the second score threshold is used to determine whether the map segment recorded by the vehicle is usable.

[0074] After calculating the first confidence score and using it as the first quality assessment result for the first memory parking map data, the terminal device can further compare the first confidence score with the second score threshold in real time. If the first confidence score is less than or equal to the first score threshold and greater than or equal to the second score threshold, it can determine that the environmental characteristic stability, sensor consistency, or positioning accuracy of the first memory parking map data has been noticeably degraded. Although it has not reached the high-quality standard, its data still has certain reference value. Therefore, the first memory parking map data is marked as medium confidence map data and stored in the local storage space.

[0075] In some embodiments, the terminal device compares a first confidence score with a first score threshold and a second score threshold in real time. If the first confidence score is less than or equal to the first score threshold and greater than or equal to the second score threshold, and the terminal device determines that the environment in which the vehicle recorded the first memory parking map data may have problems such as sensor obstruction, feature loss, or severe disturbance, then the terminal device can enter a warning state and save the first memory parking map data. Furthermore, in this warning state, the terminal device can issue visual or auditory prompts to the user through a human-computer interaction interface, but the process of the vehicle recording the memory parking map will not be interrupted.

[0076] In some embodiments, the terminal device can also continuously record first-memory parking map data in the warning state and save this first-memory parking map data as medium-confidence map data. The data content of these marked medium-confidence map segments may include dead reckoning maps based on inertial navigation, sparse environmental feature point clouds, and degraded visual features. This data can be used as auxiliary information in subsequent map building or relocation processes, and fused with high-confidence segments through a global optimization algorithm to correct accumulated errors and complete map information, thereby maximizing the use of all available sensor data while ensuring robustness.

[0077] In some embodiments, when the terminal device persistently stores the first memory parking map data, it may mark the first memory parking map data as medium confidence map data after the first confidence score is between the first score threshold and the second score threshold, and then persistently store it in the second area of ​​the local storage space.

[0078] Please refer to Figure 4 , Figure 4 for Figure 1 A schematic diagram of another detailed step in step S102.

[0079] like Figure 4 As shown, in some embodiments, when the first quality assessment result includes a first confidence score, the step of "monitoring the interruption status of the vehicle's memory parking map construction based on the first quality assessment result" in step S102 above may include steps S401 and S402 as shown below.

[0080] Step S401: In response to the first confidence score being less than the second score threshold, pause the storage of the first memory parking map data and obtain the real-time status information of the vehicle; the real-time status information includes the real-time pose information and sensor status information of the vehicle.

[0081] After calculating the first confidence score and using it as the first quality assessment result for the first memory parking map data, the terminal device can compare the first confidence score with the second score threshold in real time when monitoring the interruption status of the vehicle's memory parking map construction based on the first quality assessment result. If the first confidence score is less than the second score threshold, the storage of the first memory parking map data recorded by the vehicle in real time will be immediately paused, and the vehicle's real-time pose information and sensor status information, as well as other real-time status information of the vehicle, will be obtained.

[0082] Step S402: Determine the real-time pose state information and the sensor state information as the interruption state of the vehicle's memory parking map construction.

[0083] After obtaining the vehicle's real-time pose information and sensor status information as the vehicle's real-time status information, the terminal device further determines the real-time pose information and sensor status information as the interrupted state of the vehicle's current construction of the memory parking map.

[0084] In some embodiments, when the terminal device detects a first confidence score lower than a second score threshold (which represents the minimum tolerable reliability limit), indicating extremely poor sensor data quality during the real-time recording of the first memory parking map data, with severely missing or unreliable environmental features, directly saving the original map fragment would unacceptably contaminate the accuracy and reliability of the entire map. Therefore, the terminal device will no longer save the current low-quality original data but will immediately initiate an intelligent recording continuation process. The core of this process is to start a background monitoring thread to continuously monitor changes in the quality and confidence score of the sensor data until it recovers to an acceptable level. Furthermore, after initiating the intelligent recording continuation process, the terminal device, while pausing the saving of map fragments within the current low-confidence time period, records the real-time pose information and sensor status information of the interruption point.

[0085] In this embodiment, the terminal device integrates multimodal sensor data from the vehicle's real-time recording of the first memory parking map data to calculate a recording confidence score (i.e., the first confidence score). This first confidence score is then compared with a first score threshold and a second score threshold to determine whether the recording of the current memory parking map segment is in a normal state. If the first confidence score is lower than the second score threshold, it indicates that the current sensor data quality is extremely poor, with severely missing or unreliable environmental features. Directly saving the original map segment would cause unacceptable contamination to the accuracy and reliability of the entire map. Therefore, the terminal device no longer saves the current low-quality original data but immediately initiates an intelligent continuation recording process, recording the pose information and sensor state snapshot of the interruption point for subsequent construction to form a complete, usable, and high-precision memory parking map. Thus, this embodiment allows users to construct the desired memory parking map through this memory parking map construction method, and enables timely continuation recording when map recording is interrupted, thereby avoiding multiple repetitive recording tasks and improving user experience and resource utilization. Furthermore, this embodiment employs modular segmented storage and incremental algorithms in the terminal device to store and perform quality assessment processing on the first memory parking map data. It can also automatically resume construction of the memory parking map after an interruption in the vehicle's memory parking map construction process. Figure 1 Consistency checks are performed to improve map accuracy and robustness.

[0086] Please refer to Figure 5 , Figure 5The flowchart of the anti-interruption memory parking map construction method provided in this application is shown in some other embodiments.

[0087] like Figure 5 As shown, in some embodiments, when the interruption state is detected, the anti-interruption memory parking map construction method provided in this application embodiment may further include the following steps S501 and S502.

[0088] Step S501: In response to the vehicle restarting the memory parking map construction, dead reckoning is performed on the vehicle based on the vehicle's real-time sensor data to obtain reset guidance information.

[0089] It should be noted that restarting the memory parking map construction allows the terminal device to initiate the intelligent continuation recording process, after which the vehicle continues to collect multimodal sensor data to record new memory parking map data in real time.

[0090] After the vehicle restarts and the memory parking map is built, the terminal device can perform dead reckoning based on the vehicle's real-time sensor data to obtain reset guidance information.

[0091] In some embodiments, when the terminal device performs dead reckoning on the vehicle based on the vehicle's real-time sensor data after starting the intelligent recording process, it actively uses the remaining available sensor information (i.e., the vehicle's current real-time sensor data) to perform dead reckoning on the vehicle and attempts to identify an environmental area with rich features and high sensor data quality, thereby generating corresponding reset guidance information, which can facilitate the subsequent re-establishment of accurate vehicle positioning.

[0092] Step S502: In response to the vehicle driving according to the reset guidance information, perform a quality assessment on the second memory parking map data recorded in real time by the vehicle to obtain a second quality assessment result.

[0093] After obtaining the reset guidance information by performing dead reckoning processing, the terminal device further controls the vehicle to drive according to the reset guidance information, so that the vehicle can enter an environmental area with rich features and high sensor data quality. Moreover, while the vehicle is driving according to the reset guidance information, the terminal device can perform a quality assessment on the second memory parking map data recorded in real time by the vehicle, using the same process as the quality assessment of the first memory parking map data, and obtain a second quality assessment result.

[0094] In some embodiments, the vehicle may simultaneously acquire multimodal sensor data to record second-memory parking map data in real time when it begins to drive according to the reset guidance information. Alternatively, the vehicle may simultaneously acquire multimodal sensor data to record second-memory parking map data in real time after it has driven to an environmental area with rich features and high-quality sensor data according to the reset guidance information.

[0095] In some embodiments, after the terminal device initiates the intelligent recording process, it can continuously use dead reckoning data during the low confidence period to maintain a rough estimated map, while continuously analyzing the current limited sensor data to calculate a confidence score. When the confidence score recovers and continues to exceed a first score threshold, it indicates that the vehicle has entered a high-quality environmental area.

[0096] Please refer to Figure 6 , Figure 6 The flowchart of the anti-interruption memory parking map construction method provided in this application is shown in some other embodiments.

[0097] like Figure 6 As shown, in some embodiments, when the interruption state is detected, the anti-interruption memory parking map construction method provided in this application embodiment may further include steps S601 and S602 as shown below.

[0098] Step S601: Obtain the replanning path between the current location of the vehicle and a preset trusted area; the trusted area is a geographical area where the environmental perception performance of the vehicle meets a preset quality threshold.

[0099] After initiating the intelligent recording continuation process, the terminal device can also guide the vehicle to replan its route, thereby obtaining the replanned route between the vehicle's current location and a preset trusted area. Here, the trusted area is a geographical region where the vehicle's environmental perception performance meets a preset quality threshold; for example, this trusted area could be an environmental region rich in features and with high-quality sensor data.

[0100] In some embodiments, the terminal device can determine a feature-rich environment area with high sensor data quality from the stored first memory parking map data as a trusted area, and then use the trusted area as the target location area to plan a replanning path from the vehicle's current location to the target location area.

[0101] Step S602: In response to the vehicle restarting the memory parking map construction and driving to the trusted area according to the replanning path, the quality of the second memory parking map data recorded by the vehicle in real time is evaluated to obtain a second quality evaluation result.

[0102] After obtaining the replanning path, the terminal device immediately performs a quality assessment on the second memory parking map data recorded in real time by the vehicle when the vehicle restarts the memory parking map construction and drives to a trusted area according to the replanning path, thereby obtaining a second quality assessment result.

[0103] In some embodiments, the vehicle may simultaneously acquire multimodal sensor data to record second-memory parking map data in real time as it begins to travel along the replanned path. Alternatively, the vehicle may begin simultaneously acquiring multimodal sensor data to record second-memory parking map data in real time only after it has traveled to a trusted area along the replanned path.

[0104] In some embodiments, after the terminal device starts the intelligent recording process, it can also adopt the "pause recording - manual intervention" mode, in which the vehicle user drives the vehicle to a trusted area according to the prompts, and performs a quality assessment on the second memory parking map data recorded in real time by the vehicle, thereby obtaining a second quality assessment result.

[0105] Please refer to Figure 7 , Figure 7 for Figure 1 A detailed flowchart of step S103.

[0106] like Figure 7 As shown, in some embodiments, the step S103 above, "fusing the second memory parking map data with the first memory parking map data based on the interruption state to obtain a complete memory parking map", may include steps S701 and S702 as shown below.

[0107] Step S701: Compare the target pose information of the vehicle when it records the second memory parking map data in real time with the real-time pose state information of the vehicle in the interrupted state to obtain the comparison result.

[0108] After the vehicle restarts the memory parking map construction and performs a quality assessment on the second memory parking map data recorded in real time to obtain a second quality assessment result, the terminal device then merges the second memory parking map data with the stored first memory parking map data based on the previously monitored and recorded interruption state. If the second quality assessment result is higher than or the same as the first quality assessment result, the terminal device first compares the target pose information when the vehicle recorded the second memory parking map data in real time with the real-time pose information recorded during the vehicle terminal memory parking map construction in the interruption state, thereby obtaining the comparison result between the two pose information.

[0109] Step S702: In response to the comparison result indicating that the target pose information is the same as the real-time pose information, the second memory parking map data is stitched together with the stored first memory parking map data to obtain a complete memory parking map.

[0110] After obtaining the comparison result, if the comparison result indicates that the target pose information is the same as the real-time pose information, the terminal device will stitch the second memory parking map data with the currently stored first memory parking map data to construct a complete memory parking map.

[0111] In some embodiments, after triggering the intelligent recording process, the terminal device may attempt to maintain a coarse-grained internal state estimate by performing sensor self-testing, data cleaning, or switching to a backup sensor fusion strategy in order to maintain continuous tracking of the vehicle's position.

[0112] In some embodiments, when the second quality assessment result is also a confidence score, the terminal device can continuously monitor and detect when the confidence score rises again and exceeds the second score threshold. Instead of immediately saving the second memory parking map data, it waits for the confidence score to further stabilize and exceed the first score threshold before restarting the saving of high-quality map segments. Then, it uses an optimization algorithm to intelligently stitch and globally optimize the newly recorded high-quality second memory parking map data with the high-quality first memory parking map data stored before the interruption, thereby supplementing the map information missing during the interruption and finally constructing a complete and high-precision memory parking map.

[0113] Please refer to Figure 8 , Figure 8 The flowchart of the anti-interruption memory parking map construction method provided in this application is shown in some other embodiments.

[0114] like Figure 8 As shown, in some embodiments, the anti-interruption memory parking map construction method provided in this application may further include steps S801 and S802 as shown below.

[0115] Step S801: In response to the vehicle restarting the memory parking map construction, calculate the pose information of the vehicle relative to the stored first memory parking map data.

[0116] After the vehicle restarts and the memory parking map data is restarted, the terminal device can calculate the vehicle's pose information relative to the stored first memory parking map data based on repositioning technology.

[0117] In some embodiments, after starting the intelligent recording process, the terminal device can continuously analyze the limited sensor data at present while maintaining a rough map estimate using dead reckoning data during the low confidence period. If the confidence score recovers and continuously exceeds the first score threshold, it can determine that the vehicle has entered a high-quality perception area (such as a reliable area) and determine that the current position of the vehicle is a feasible reset point. Then, it can start the relocation module to quickly match the environmental features currently perceived by the vehicle with the stored high-precision map fragments (first memory parking map data) to calculate the precise pose information of the vehicle relative to the recorded map.

[0118] Step S802: Based on the pose information, the second memory parking map data and the first memory parking map data are fused to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0119] After the terminal device calculates the precise pose information of the vehicle relative to the recorded map, if the pose information is the same as the real-time pose information in the previously monitored interrupted state, and the second quality assessment result is also a confidence score, the terminal device can continuously monitor and detect if the confidence score rises again and exceeds the first score threshold (i.e., the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result). Then, it can restart the saving of high-quality map segments and use optimization algorithms to intelligently stitch and globally optimize the newly recorded high-quality second memory parking map data with the high-quality first memory parking map data stored before the interruption, thereby completing the map information missing during the interruption and finally constructing a complete and high-precision memory parking map.

[0120] In this embodiment, after the vehicle restarts the memory parking map data, the terminal device calculates the vehicle's pose information relative to the stored first memory parking map data based on repositioning technology. Then, if the pose information is the same as the real-time pose information in the previously monitored interrupted state, and the second quality assessment result is also a confidence score, the terminal device can continuously monitor and detect that the confidence score has risen again and exceeded the first score threshold. Then, it can restart the saving of high-quality map segments and use optimization algorithms to intelligently stitch and globally optimize the newly recorded high-quality second memory parking map data and the high-quality first memory parking map data stored before the interruption, thereby completing the map information missing during the interruption and finally constructing a complete and high-precision memory parking map. Thus, this embodiment achieves the synergistic effect of the adaptive mapping process, the interruption-resumption function, and the real-time interruption detection and recording mechanism when building a memory parking map for a vehicle. It can continue recording with an intelligent resume process even if the memory parking map building is interrupted, resulting in a complete and high-precision memory parking map. This solves the problem of the function being unavailable due to the interruption of memory parking recording, allowing users to build the desired memory parking map with anti-interruption capabilities, improving the user experience and resource utilization.

[0121] Please refer to Figure 9 , Figure 9 This is a schematic diagram illustrating the complete steps involved in one embodiment of the anti-interruption memory parking map construction method provided in this application.

[0122] In some embodiments, when a terminal device applies the uninterrupted memory parking map construction method provided in this application, it can calculate a recording confidence score by fusing multimodal sensor data, and compare it with a first score threshold and a second score threshold to determine whether the current memory parking map segment recording is in a normal state. If it is lower than the second score threshold, an intelligent continuation recording process is promptly activated to obtain a complete, usable, and high-precision memory parking map. Specifically, the terminal device can employ methods such as... Figure 9 Steps S901 to S906, as shown, allow the user to construct the desired memory parking map and enable timely resumption of recording when recording is interrupted, avoiding repeated recording work and thus improving the user experience and resource utilization. Specifically: Step S901 is: Start recording the parking map.

[0123] "Start recording" refers to the time when the vehicle enters a state where the memory parking function can be activated, and the relevant "start recording" button is clicked. "Memory parking map recording" refers to the process during which the vehicle, in conjunction with its own cameras, radar, and other sensors, scans, records, and digitally constructs the environmental features around the driving map in real time, including lane lines, speed bumps, slopes, and other information.

[0124] Step S902 is: fusing multimodal sensor data and calculating the recording confidence score.

[0125] Multimodal sensor data fusion refers to the comprehensive evaluation of the reliability and accuracy of currently recorded map segments by integrating data from multiple sensors. The terminal device simultaneously collects data from various sensors, including cameras, LiDAR, millimeter-wave radar, ultrasonic sensors, and inertial measurement units; this data includes visual information, point cloud data, and motion status. Confidence score calculation involves real-time processing and fusion of this data; the system then generates a "confidence score" to quantify the reliability of the currently recorded segment.

[0126] The terminal device uses sensor fusion algorithms to perform spatiotemporal alignment and feature extraction on multi-source heterogeneous data. Based on deep learning models or probabilistic fusion methods, it calculates the stability of current environmental features, consistency between sensors, and positioning accuracy, thus outputting a confidence score within the range of 0 to 1. This score will serve as a key criterion for determining whether the map recording quality requirements are met in subsequent steps.

[0127] Based on the recording confidence score obtained during the memory parking map recording process, the following step S903 is performed by comparing it with a first threshold.

[0128] Step S903 is: If the score is higher than the first score threshold, save the current map segment.

[0129] The first score threshold is a preset reliability threshold used to determine whether the recorded map segment has sufficient accuracy and robustness for subsequent positioning and navigation. If the confidence score of the current segment is higher than this threshold, it indicates that the environmental features of the segment are clear, the sensor data is consistent, and the positioning results are stable, meeting the mapping quality requirements.

[0130] The terminal device compares the confidence score with a first score threshold in real time. When the score exceeds the threshold, the domain controller triggers the map saving process. Sensor data within the current time window (including keyframe images, point clouds, feature points, pose information, etc.) is compressed and persistently stored in the vehicle's storage unit as a valid fragment of the high-precision map for subsequent map planning and vehicle control. Simultaneously, the terminal device generates corresponding map fragment indexes and metadata to facilitate subsequent retrieval and stitching.

[0131] If the confidence score is lower than or equal to the first score threshold, it indicates that there may be problems such as sensor occlusion, feature loss or severe disturbance in the current environment, and the following steps S904 are performed.

[0132] Step S904 is as follows: If the value is between the first threshold and the second threshold, the system enters the warning state and saves the current map fragment.

[0133] The second score threshold is the minimum acceptable threshold for determining whether a map segment is usable (i.e., the first score threshold > the second score threshold). When the confidence score is between the first and second score thresholds, it indicates a noticeable degradation in the environmental stability, sensor consistency, or positioning accuracy of the current map segment. Although it does not meet the high-quality standard, its data still has some reference value. The terminal device will then enter an alert state and issue visual or auditory prompts to the user through the human-computer interaction interface, but the recording process will not be interrupted. Simultaneously, the terminal device will save the current map segment, but will add a special marker to it and store it separately from high-confidence segments.

[0134] Even in the warning state, the terminal device continues to record and save data. These labeled medium-confidence segments may contain data such as dead reckoning maps based on inertial navigation, sparse environmental feature point clouds, and degraded visual features. This data can be used as auxiliary information in subsequent map building or relocation processes, fused with high-confidence segments through a global optimization algorithm to correct accumulated errors and supplement map information. This ensures system robustness while maximizing the use of all available sensor data.

[0135] When the recorded confidence score continues to drop below the second threshold, perform the following step S905.

[0136] Step S905 is: if the score is below the second threshold, then start the intelligent recording process.

[0137] The second score threshold represents the minimum reliability that the terminal device can tolerate. When the confidence score falls below this threshold, it indicates that the current sensor data quality is extremely poor, environmental features are severely lacking or unreliable, and directly saving the original map fragment will cause unacceptable contamination to the accuracy and reliability of the entire map. The terminal device will no longer save the current low-quality original data but will immediately initiate an intelligent data continuation process. The core of this process is to start a background monitoring thread to continuously monitor changes in the quality and confidence score of the sensor data until it recovers to an acceptable level.

[0138] Once the intelligent recording process is triggered, the terminal device first pauses the saving of map segments within the current low-confidence period and records the pose information and sensor state snapshots of the interruption points. Simultaneously, the terminal device attempts to maintain a coarse-grained internal state estimate through sensor self-checks, data cleaning, or switching to a backup sensor fusion strategy to ensure continuous tracking of the vehicle's position. When the terminal device detects through continuous monitoring that the confidence score has risen again and exceeded the second threshold, it does not immediately save the data. Instead, it waits for the score to further stabilize and exceed the first threshold before resuming the saving of high-quality map segments. It then uses optimization algorithms to intelligently stitch and globally optimize the newly recorded high-quality segments with the high-quality segments from before the interruption, thereby completing the map information missing during the interruption and ultimately forming a complete and high-precision memory parking map.

[0139] After starting the intelligent recording process, perform the following steps S906.

[0140] Step S906 is: guide the vehicle to find the reset point, and continue recording through repositioning technology to obtain a complete and high-precision memory parking map.

[0141] Guiding the vehicle to find a reset point refers to the system proactively using remaining available sensor information to perform dead reckoning during the intelligent recording process, and attempting to identify an environmental area rich in features and with high-quality sensor data in order to re-establish accurate vehicle positioning. Re-relocation technology for continuous recording means that when the vehicle enters a suitable reset point area, the system will invoke a relocation algorithm to match and align the vehicle's current precise pose with the high-precision map segments recorded before the interruption, thereby seamlessly continuing subsequent map recording and ensuring that the final generated map is a complete map that is geometrically and logically continuous and accurate.

[0142] After initiating the intelligent recording process, the terminal device continuously utilizes dead reckoning data from the low-confidence period to maintain a rough estimated map. Simultaneously, the terminal device continuously analyzes the limited sensor data. Once the confidence score recovers and consistently exceeds a first threshold, indicating that the vehicle has entered a high-quality perception area, the terminal device determines this to be a feasible reset point. Subsequently, the terminal device activates the relocalization module, quickly matching the currently perceived environmental features with stored high-precision map fragments to calculate the vehicle's precise pose relative to the recorded map. After successful relocalization, the terminal device continues saving high-quality map fragments from this pose point, and globally optimizes and stitches the newly recorded fragments with the original map, ultimately resulting in a seamless, complete, and high-precision memory parking map.

[0143] In summary, this embodiment, by introducing a recording confidence score and applying judgment conditions of a first score threshold and a second score threshold, can determine the status of the currently recorded map segment. Furthermore, this embodiment incorporates an intelligent continuation recording process to continue recording, resulting in a complete, high-precision memory parking map. This solves the problem of functional unavailability caused by interrupted memory parking recording, allowing users to construct the desired memory parking map with uninterrupted recording capabilities, thus improving user experience and resource utilization.

[0144] Please refer to Figure 10 This application also provides an uninterrupted memory parking map construction system, which can implement the above-mentioned uninterrupted memory parking map construction method.

[0145] like Figure 10 As shown, the uninterrupted memory parking map construction system provided in this application embodiment may include: The quality assessment module is used to assess the quality of the first memory parking map data recorded in real time by the vehicle in response to the vehicle starting memory parking map construction, and to obtain the first quality assessment result. The storage module is used to store the first memory parking map data based on the first quality assessment result; An interruption monitoring module is used to monitor the interruption status of the vehicle's memory parking map construction based on the first quality assessment result; The intelligent continuation recording module is used to merge the second memory parking map data with the first memory parking map data based on the interruption state to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0146] In some embodiments, the quality assessment module is further configured to acquire multimodal sensor data when the vehicle records the first memory parking map data in real time; perform reliability quantification processing on the first memory parking map data based on the multimodal sensor data to obtain a first confidence score of the first memory parking map data; and the first quality assessment result includes the first confidence score.

[0147] In some embodiments, when the first quality assessment result includes a first confidence score, the storage module is further configured to, in response to the first confidence score being greater than a first score threshold, mark the first memory parking map data as high confidence map data for storage; and in response to the first confidence score being less than or equal to the first score threshold and greater than or equal to a second score threshold, mark the first memory parking map data as medium confidence map data for storage.

[0148] In some embodiments, the interruption monitoring module is further configured to, in response to the first confidence score being less than a second score threshold, suspend the storage of the first memory parking map data and obtain the real-time status information of the vehicle; the real-time status information includes the real-time pose information and sensor status information of the vehicle; and determine the real-time pose information and the sensor status information as an interruption state for the vehicle to build the memory parking map.

[0149] In some embodiments, when the interruption state is detected, the quality assessment module is further configured to, in response to the vehicle restarting the memory parking map construction, perform dead reckoning processing on the vehicle based on the vehicle's real-time sensor data to obtain reset guidance information; and in response to the vehicle driving according to the reset guidance information, perform quality assessment on the second memory parking map data recorded by the vehicle in real time to obtain a second quality assessment result.

[0150] In some embodiments, when the interruption state is detected, the quality assessment module is further configured to obtain the replanning path between the current position of the vehicle and a preset trusted area; the trusted area is a geographical area where the environmental perception performance of the vehicle meets a preset quality threshold; in response to the vehicle restarting the memory parking map construction and driving to the trusted area according to the replanning path, the quality assessment is performed on the second memory parking map data recorded by the vehicle in real time to obtain a second quality assessment result.

[0151] In some embodiments, the intelligent recording module is further configured to compare the target pose information of the vehicle when recording the second memory parking map data in real time with the real-time pose state information of the vehicle in the interrupted state to obtain a comparison result; in response to the comparison result indicating that the target pose information is the same as the real-time pose information, the second memory parking map data is spliced ​​with the stored first memory parking map data to obtain a complete memory parking map.

[0152] In some embodiments, the intelligent recording module is further configured to, in response to the vehicle restarting the memory parking map construction, calculate the pose information of the vehicle relative to the stored first memory parking map data; and fuse the second memory parking map data with the first memory parking map data based on the pose information to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

[0153] It should be noted that the specific implementation of the anti-interruption memory parking map construction system provided in this application is basically the same as the specific implementation of the anti-interruption memory parking map construction method described above, and will not be repeated here.

[0154] Please see Figure 11 This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described anti-interruption memory parking map construction method.

[0155] In some embodiments, the electronic device may be any smart terminal such as an in-vehicle terminal, an in-vehicle hardware platform (e.g., an in-vehicle computer), a tablet computer, a smartphone, or a wearable device; or, the electronic device may be a vehicle including a memory and a processor.

[0156] like Figure 11 As shown, the electronic device provided in this application embodiment may include: The processor 1101 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1102 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1102 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1102 and is called and executed by the processor 1101 to implement the interrupt-resistant memory parking map construction method of the embodiments of this application. Input / output interface 1103 is used to implement information input and output; The communication interface 1104 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1105 transmits information between various components of the device (e.g., processor 1101, memory 1102, input / output interface 1103, and communication interface 1104); The processor 1101, memory 1102, input / output interface 1103 and communication interface 1104 are connected to each other within the device via bus 1105.

[0157] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described interrupt-resistant memory parking map construction method.

[0158] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0159] This application also provides a computer program product, including a computer program. The steps implemented by the computer program when executed by a processor are basically the same as those in the specific embodiments of the above-described anti-interruption memory parking map construction method, and will not be repeated here.

[0160] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0161] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0162] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0163] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0164] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0165] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding factor is divided by the following factor, or that the related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0166] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0167] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0168] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0169] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0170] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for constructing an uninterrupted memory parking map, characterized in that, The method includes: In response to the vehicle initiating the memory parking map construction, the quality of the first memory parking map data recorded in real time by the vehicle is evaluated to obtain the first quality evaluation result; The first memory parking map data is stored based on the first quality assessment result, and the interruption status of the vehicle's memory parking map construction is monitored based on the first quality assessment result. Based on the interruption state, the second memory parking map data is fused with the first memory parking map data to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

2. The method according to claim 1, characterized in that, The quality assessment of the first memory parking map data recorded in real time by the vehicle to obtain a first quality assessment result includes: Acquire multimodal sensor data when the vehicle records the first memory parking map data in real time; The reliability of the first memory parking map data is quantified based on the multimodal sensor data to obtain a first confidence score for the first memory parking map data; the first quality assessment result includes the first confidence score.

3. The method according to claim 1, characterized in that, If the first quality assessment result includes a first confidence score, storing the first memory parking map data based on the first quality assessment result includes: In response to the first confidence score being greater than the first score threshold, the first memory parking map data is marked as high-confidence map data and stored. In response to the first confidence score being less than or equal to a first score threshold and greater than or equal to a second score threshold, the first memory parking map data is marked as medium confidence map data and stored.

4. The method according to claim 3, characterized in that, The monitoring of the interruption status of the vehicle's memory parking map construction based on the first quality assessment result includes: In response to the first confidence score being less than the second score threshold, the storage of the first memory parking map data is paused and the real-time status information of the vehicle is obtained; the real-time status information includes the real-time pose information and sensor status information of the vehicle; The real-time pose information and the sensor status information are determined as the interruption state of the vehicle's construction of a memory parking map.

5. The method according to claim 1, characterized in that, If the interruption state is detected, the method further includes: In response to the vehicle restarting the memory parking map construction, dead reckoning is performed on the vehicle based on the vehicle's real-time sensor data to obtain reset guidance information; In response to the vehicle driving according to the reset guidance information, the quality of the second memory parking map data recorded in real time by the vehicle is evaluated to obtain a second quality evaluation result.

6. The method according to claim 1, characterized in that, If the interruption state is detected, the method further includes: Obtain the replanning path between the vehicle's current location and a preset trusted region; the trusted region is a geographical area where the vehicle's environmental perception performance meets a preset quality threshold. In response to the vehicle restarting the memory parking map construction and driving to the trusted area according to the replanning path, the quality of the second memory parking map data recorded by the vehicle in real time is evaluated to obtain a second quality evaluation result.

7. The method according to claim 1, characterized in that, The process of fusing the second memory parking map data with the first memory parking map data based on the interruption state to obtain a complete memory parking map includes: The target pose information of the vehicle when recording the second memory parking map data in real time is compared with the real-time pose state information of the vehicle in the interrupted state to obtain the comparison result. In response to the comparison result indicating that the target pose information is the same as the real-time pose information, the second memory parking map data is stitched together with the stored first memory parking map data to obtain a complete memory parking map.

8. The method according to claim 1, characterized in that, The method further includes: In response to the vehicle restarting the memory parking map construction, the vehicle's pose information relative to the stored first memory parking map data is calculated; Based on the pose information, the second memory parking map data is fused with the first memory parking map data to obtain a complete memory parking map; the second memory parking map data is the memory parking map data recorded in real time after the vehicle restarts the memory parking map construction, and the second quality assessment result of the second memory parking map data is higher than the first quality assessment result, or the second quality assessment result is the same as the first quality assessment result.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the uninterrupted memory parking map construction method according to any one of claims 1 to 8.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the interruptible memory parking map construction method as described in any one of claims 1 to 8.