A method to improve memory parking efficiency

Abstract

The application provides a method for improving memory parking efficiency, which comprises the following steps: starting background path learning when a vehicle meets background path learning conditions, obtaining a current record route; obtaining a pre-recorded route, and fusing the current record route and the pre-recorded route to obtain a current optimized route; and performing memory parking based on the current optimized route. The method for improving memory parking efficiency is proposed, the background path learning is automatically started when the preset background path learning conditions are met, the delay record caused by other operations is avoided, the starting point positioning efficiency and accuracy of the background path learning are improved, the current record route data obtained is more accurate, the current optimized route is obtained by fusing the pre-recorded route and the current record route, a smoother route is provided for memory parking, and the memory parking process is smoother.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and in particular to a method for improving the efficiency of memory parking. Background Technology

[0002] Existing memory parking systems primarily use onboard sensors to record the vehicle's driving trajectory and guide the vehicle along that trajectory during subsequent parking. They rely on a series of sensors and control systems to achieve their function, including ultrasonic sensors, surround-view cameras, and controllers, all working together to complete the parking task. Furthermore, memory parking technology design elements also include memory mapping, intelligent parking entry, along-path parking space recognition, intelligent parking exit, fault handling during operation, dynamic and static obstacle handling, and video monitoring—all designed to ensure a smooth and safe parking process.

[0003] However, existing memory parking systems often lack precise positioning near the starting point, leading to a less than smooth parking process. Furthermore, current technologies haven't fully considered the accuracy of starting point positioning, the independence of system layers, and the intelligent processing of route splicing. This results in issues such as interference between background learning and local hardware data, and inconsistent splicing between pre-recorded and currently recorded routes. In conclusion, while memory parking improves parking convenience to some extent, its shortcomings cannot be ignored, requiring continuous improvement and refinement in both technology and service. Summary of the Invention

[0004] The present invention aims to provide a method to improve the efficiency of memory parking, thereby solving the above-mentioned technical problems, improving the accuracy and efficiency of memory parking learning and route positioning, and making the memory parking process smoother.

[0005] To address the aforementioned technical problems, this invention provides a method for improving memory parking efficiency, comprising the following steps:

[0006] When a vehicle meets the conditions for background path learning, background path learning is started to obtain the currently recorded route.

[0007] Obtain the pre-recorded route and fuse it with the currently recorded route to obtain the current optimized route;

[0008] Parking is performed based on the current optimized route.

[0009] In the above scheme, by pre-setting background path learning conditions, background path learning is automatically started when the conditions are met, avoiding delays caused by other operations, improving the efficiency and accuracy of background path learning starting point positioning, and making the current recorded route data more accurate; and by merging the pre-recorded route with the current recorded route to obtain the current optimized route, providing a smoother route for memory parking, making the memory parking process smoother.

[0010] Furthermore, when the vehicle meets the background path learning conditions, background path learning is initiated to obtain the currently recorded route, including: collecting current vehicle speed information and current gear information, and analyzing and processing them based on a preset logic algorithm to obtain the current detection result; if the current detection result meets the background path learning conditions, background path learning is initiated to obtain the currently recorded route.

[0011] In the above scheme, the current vehicle speed and current gear information are collected and analyzed as conditions for starting background path learning. The scheme is automatically activated when the detection results meet the preset conditions, without the need for manual operation. This makes the starting point positioning of background path learning more timely and accurate, and improves the accuracy and efficiency of background path learning.

[0012] Furthermore, the process involves acquiring a pre-recorded route and fusing the currently recorded route with the pre-recorded route to obtain the current optimized route. This includes: acquiring the pre-recorded route; matching marker points based on the currently recorded route and the pre-recorded route to determine the splicing position; performing fusion and splicing processing using a preset fusion algorithm based on the currently recorded route, the pre-recorded route, and the splicing position to obtain the current spliced ​​route; and performing optimization processing based on the current spliced ​​route and a route intelligent algorithm to obtain the current optimized route.

[0013] In the above scheme, by performing marker point matching analysis on the current recorded route and the pre-recorded route, and using intelligent algorithms to fuse, splice, and optimize them to obtain the current optimized route, a current optimized parking route is provided for memory parking, making the memory parking process smoother.

[0014] Furthermore, the process involves matching marker points based on the current recorded line and the pre-recorded line to determine the splicing position. This includes: preprocessing the data based on the current recorded line and the pre-recorded line, and setting several common marker points according to a preset selection method. The marker points include location information. The process also involves matching and analyzing the marker points to identify location information that meets the preset marker conditions, thereby determining the splicing position.

[0015] In the above scheme, by preprocessing the data of the currently recorded line and the pre-recorded line, the accuracy and consistency of the line data are ensured. This allows for the selection of several common identification points between the pre-recorded line and the currently recorded line according to the preset selection method, and the matching analysis is performed to determine the splicing position, making the splicing position more accurate and providing a foundation for subsequent line fusion processing.

[0016] Furthermore, the process of matching marker points based on the current recorded line and the pre-recorded line to determine the splicing position also includes: when the accuracy verification of the splicing position does not meet the preset accuracy requirements, re-preprocessing the data based on the current recorded line and the pre-recorded line, and setting several common marker points according to the preset selection method.

[0017] The above solution improves the accuracy and reliability of the splicing position by verifying the accuracy of the splicing position and adjusting and reselecting the marker points according to the actual situation.

[0018] Furthermore, based on the current recorded line, the pre-recorded line, and the splicing position, a preset fusion algorithm is used for fusion and splicing processing to obtain the current splicing line, which includes several current data points in the current recorded line and several pre-recorded data points in the pre-recorded line. The preset fusion algorithm satisfies the following calculations:

[0019]

[0020] In the formula: (x fused y fused The numbers () represent the position coordinates after fusion and splicing, A represents the currently recorded line, B represents the pre-recorded line, nA represents the current number of data points, and nB represents the pre-recorded number of data points. and This represents the position coordinates of the i-th data point in A. and This represents the position coordinates of the j-th data point in B. This represents the weight assigned to the i-th data point in A. This represents the weight assigned to the j-th data point in B.

[0021] In the above scheme, by assigning weights to the pre-data points and the current data points respectively, it is convenient to perform weighted fusion calculations based on different actual factors and their magnitude of influence. This allows for more accurate determination of the splicing position under different line influence factors, thereby improving the reliability of splicing position positioning.

[0022] Furthermore, based on the current splicing route and the intelligent route algorithm, optimization processing is performed to obtain the current optimized route, including: using the intelligent route algorithm to optimize the current splicing route and obtaining the route optimization processing result; constructing a simulation environment to verify the current splicing route and obtaining the verification result; and obtaining the current optimized route based on the route optimization processing result and the verification result.

[0023] In the above solution, an intelligent algorithm is introduced to verify and optimize the current splicing line, and a simulation environment is constructed to verify its rationality, thereby obtaining an optimized and feasible line, which improves the accuracy and smoothness of line splicing.

[0024] Furthermore, methods to improve memory parking efficiency include: setting up a non-hardware recording database and associating it with other databases through software logic; after the background automatically learns and confirms that there are no errors, storing the current optimized route in the non-hardware recording database as a pre-recorded route for subsequent memory parking.

[0025] In the above scheme, the hardware recording database is used to store and process data directly collected by the hardware. A non-hardware recording database is set up to store the current optimized route and automatically expands the route through software algorithms. This is different from other database storage and realizes an independent storage and processing mechanism, avoiding interference from other data on the background path learning data.

[0026] Furthermore, methods to improve memory parking efficiency include: when the vehicle meets the front-end path learning conditions, enabling front-end path learning and performing memory parking based on the front-end path learning.

[0027] In the above solution, by setting up front-end path learning, users can directly control the activation of the memory parking function, realizing user interaction and improving the user experience.

[0028] Furthermore, methods to improve memory parking efficiency include: setting priorities so that when front-end path learning is enabled, background path learning is automatically paused, and only one of front-end path learning and background path learning is active at any given time.

[0029] In the above scheme, by setting priorities and ensuring mutual exclusion of states, only one of the front-end path learning and the back-end path learning is performed at a time, avoiding confusion caused by repeated learning records when they are performed simultaneously, and ensuring the accuracy and independence of the data.

[0030] The above solution provides a method to improve the efficiency of memory parking. It controls the activation of background path learning by pre-setting background path learning conditions, making the currently recorded route data more accurate. It employs an independent data storage and processing mechanism to distinguish route data from hardware-recorded data to avoid data interference and ensure data accuracy. It introduces an intelligent route splicing and fusion method to improve the accuracy and efficiency of route splicing and fusion. Furthermore, it sets up front-end and back-end path learning, enabling manual and automatic modes, thus enhancing the user experience of memory parking and improving the overall performance of the memory parking system. Attached Figure Description

[0031] Figure 1 This is a schematic flowchart of a method for improving memory parking efficiency according to an embodiment of the present invention;

[0032] Figure 2 This is a schematic diagram of a system for improving memory parking efficiency according to an embodiment of the present invention;

[0033] Figure 3 This is a schematic diagram illustrating the specific process of a system for improving memory parking efficiency according to an embodiment of the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Example 1:

[0036] Please see Figure 1 This embodiment provides a method for improving memory parking efficiency, including the following steps:

[0037] When a vehicle meets the conditions for background path learning, background path learning is started to obtain the currently recorded route.

[0038] Obtain the pre-recorded route and fuse it with the currently recorded route to obtain the current optimized route;

[0039] Parking is performed based on the current optimized route.

[0040] In the specific implementation process, the above solution automatically starts background path learning when the preset background path learning conditions are met, avoiding delays caused by other operations, improving the efficiency and accuracy of background path learning starting point positioning, and making the current recorded route data more accurate; and by merging the pre-recorded route with the current recorded route to obtain the current optimized route, providing a smoother route for memory parking, making the memory parking process smoother.

[0041] Furthermore, when the vehicle meets the background path learning conditions, background path learning is initiated to obtain the currently recorded route, including: collecting current vehicle speed information and current gear information, and analyzing and processing them based on a preset logic algorithm to obtain the current detection result; if the current detection result meets the background path learning conditions, background path learning is initiated to obtain the currently recorded route.

[0042] Optionally, information sharing and transmission can be achieved through vehicle bus communication technology. This can be accomplished using CAN (Controller Area Network) bus or similar technologies to realize the transmission and sharing of data signals and to provide feedback mechanisms for real-time display. For example, the current vehicle speed and gear information can be displayed in real time on the vehicle's dashboard or the driver's display screen for the driver's reference.

[0043] In practice, this feedback mechanism helps drivers understand the current status of the vehicle and take appropriate actions when necessary.

[0044] Optionally, vehicle speed sensors can be used to collect current vehicle speed data. Vehicle speed sensors include, but are not limited to: radar sensors, such as millimeter-wave radar, which measure the relative speed between the vehicle and its surrounding environment by emitting and receiving electromagnetic waves. They are mainly used to measure the relative speed with other objects, rather than directly measuring the current vehicle speed; wheel speed sensors, which are directly mounted on the wheels and calculate the current vehicle speed by measuring the rotational speed of the wheels; and vision cameras, which, although mainly used for image processing and recognition, can also assist in determining the current vehicle speed in certain situations, such as by recognizing changes in the relative position of road markings or vehicles ahead.

[0045] Optionally, a gear position sensor is used to collect the current gear position data. The gear position sensor is directly mounted on the transmission and is used to detect the current gear position status. The current gear position data is an electrical signal. The gear position sensor converts the current gear position status (such as P, R, N, D, etc.) into an electrical signal and transmits it through vehicle bus communication (such as CAN bus) to determine whether the current gear position information meets the background path learning conditions based on a preset logic algorithm.

[0046] Optionally, the preset logic algorithm includes: signal parsing, specifically parsing the received current gear position data and converting it from a physical signal into a recognizable logical state or digital coded signal; mapping relationship, specifically matching the parsed signal with the specific gear position according to the mapping relationship defined by the vehicle manufacturer. This mapping relationship is usually stored in the controller's memory and configured during the vehicle manufacturing process; and logical judgment, specifically determining the corresponding gear position of the signal and then making further judgments based on preset logic. For example, if the received signal indicates "D" (drive gear), it is confirmed that the vehicle is currently in a forward state, and the corresponding operation or function is triggered accordingly.

[0047] In the specific implementation process, the background path conditions can be set as follows: the current vehicle speed information meets the condition of being less than or equal to 30km / h and the current gear information meets the condition of being in "D" (drive gear) signal. At this time, background path learning will be automatically started.

[0048] Furthermore, the process involves acquiring a pre-recorded route and fusing the currently recorded route with the pre-recorded route to obtain the current optimized route. This includes: acquiring the pre-recorded route; matching marker points based on the currently recorded route and the pre-recorded route to determine the splicing position; performing fusion and splicing processing using a preset fusion algorithm based on the currently recorded route, the pre-recorded route, and the splicing position to obtain the current spliced ​​route; and performing optimization processing based on the current spliced ​​route and a route intelligent algorithm to obtain the current optimized route.

[0049] Optionally, based on the currently recorded route, the pre-recorded route, and the splicing position, a preset fusion algorithm is used for fusion and splicing processing. When obtaining the current spliced ​​route, the following steps are included but not limited to: similarity calculation, specifically using the Dynamic Time Warping (DTW) algorithm to calculate the similarity between the two routes, handling minor differences and offsets in the routes, thereby identifying the overlapping parts of the two routes; segmented matching, specifically dividing the long route into multiple short segments and matching each segment, which can more effectively handle interruptions and differences in the route while reducing computational complexity; feature extraction and comparison, specifically extracting key feature points in the route (such as turning points, intersections, etc., which can be used as subsequent marker points) and comparing them. These feature points can help identify the overlapping and different parts of the two routes.

[0050] Optionally, optimization processing is performed based on the current splicing route and intelligent route algorithm. When obtaining the current optimized route, the following steps are included but not limited to: fusion based on preset fusion rules, specifically, based on the route matching results, preset fusion rules are formulated to fuse two routes. For example, if a route segment exists in both data and has a high similarity, then the route segment is retained first; if a route segment is missing in one data but exists in another data, then an attempt is made to add the route segment to the missing route, etc.; path planning algorithm, specifically, using path planning algorithms (such as A* algorithm, Dijkstra algorithm, etc.) to optimize the fused route to ensure the rationality and efficiency of the route; simulation verification, specifically, verifying the fused route in a simulation environment to ensure that it can work normally and meet the requirements in actual applications.

[0051] In the specific implementation process, the route fusion process needs to comprehensively consider factors such as data accuracy, algorithm efficiency, system security, and robustness. Attention should be paid to: ensuring data security, i.e., ensuring data security and privacy when processing route data to avoid leaking user information and sensitive data; considering the real-time performance of the algorithm, as memory parking is an application scenario with high real-time requirements, therefore it is necessary to ensure that the algorithm can complete the route matching and fusion task in a short time; and enhancing system robustness, i.e., in practical applications, various complex and changing environmental factors may be encountered, therefore it is necessary to enhance the system's robustness to cope with various unexpected situations.

[0052] Furthermore, the process involves matching marker points based on the current recorded line and the pre-recorded line to determine the splicing position. This includes: preprocessing the data based on the current recorded line and the pre-recorded line, and setting several common marker points according to a preset selection method. The marker points include location information. The process also involves matching and analyzing the marker points to identify location information that meets the preset marker conditions, thereby determining the splicing position.

[0053] Optionally, data preprocessing can be performed based on the currently recorded line and the pre-recorded line, including but not limited to the following steps: data cleaning, specifically cleaning the data of the currently recorded line and the pre-recorded line to remove noise and outliers, and ensuring the consistency and accuracy of the data; data alignment, aligning the two sets of data in time and space for subsequent comparison and analysis, which can be achieved through timestamps, GPS coordinates, etc.

[0054] Optionally, when setting up markers, elements with significant features or easy identification along the route are generally selected as markers. These elements can be fixed geographical landmarks, road features, building outlines, traffic facilities, etc. Marker points should adhere to the following principles: Significance: Marker points should have significant visual or geographical features to facilitate easy identification and location on maps or in the field; Stability: Marker points should be relatively stable and not easily affected by environmental changes or human factors; Uniformity: Marker points should be placed at key nodes or turning points along the route to ensure a uniform distribution of marker points throughout the route, facilitating subsequent data processing and stitching.

[0055] In the specific implementation process, the following methods can be used to set up marker points: field investigation, through field investigation, select significant landmarks or feature points in the route as marker points, and record their detailed location information (such as latitude and longitude, coordinates, etc.); map comparison, using high-precision maps or satellite imagery, combined with the results of field investigation, to mark all selected marker points on the map; and technical assistance, using modern technical means such as GPS positioning technology, drone aerial photography, and LiDAR scanning to improve the accuracy and efficiency of marker point setting.

[0056] Furthermore, the process of matching marker points based on the current recorded line and the pre-recorded line to determine the splicing position also includes: when the accuracy verification of the splicing position does not meet the preset accuracy requirements, re-preprocessing the data based on the current recorded line and the pre-recorded line, and setting several common marker points according to the preset selection method.

[0057] In the specific implementation process, the steps for determining the splicing position of two lines through marker points (location, distribution trajectory) include: position matching, specifically comparing and matching the location information of marker points on the two lines to be spliced, and finding marker points with similar or overlapping locations as candidate splicing points; trajectory analysis, specifically analyzing the distribution trajectory of marker points on the two lines to determine whether they have similar directions or shape characteristics. If the marker point trajectories of the two lines show obvious intersection or overlap in a certain area, then that area is likely to be the splicing position of the two lines; accuracy verification, specifically verifying the accuracy of the splicing position. This can be done by measuring parameters such as the distance difference and angle difference between the two lines at the splicing position to evaluate whether the splicing accuracy meets the requirements, ensuring the accuracy of the splicing; and optimization and adjustment, specifically, if the accuracy of the splicing position does not meet the requirements, it is necessary to adjust the marker point positions according to the actual situation or select new marker points for splicing. At the same time, algorithms can also be used to optimize the splicing position to improve the accuracy and reliability of the splicing.

[0058] In the specific implementation process, when setting up marker points for line splicing, it is necessary to ensure that the data used is consistent and accurate to avoid splicing errors or mistakes caused by inconsistent data. Considering the differences and complexities of line characteristics under different environments, it is necessary to flexibly select marker points and adjust splicing strategies to adapt to different environmental conditions. With the continuous development of technology, attention should be paid to the application of new technologies in marker point setting and line splicing so as to update and optimize existing methods and tools in a timely manner.

[0059] Furthermore, the currently recorded line includes several current data points, and the pre-recorded line includes several pre-recorded data points. The preset fusion algorithm satisfies the following calculations:

[0060]

[0061] In the formula: (x fused y fused The numbers () represent the position coordinates after fusion and splicing, A represents the currently recorded line, B represents the pre-recorded line, nA represents the current number of data points, and nB represents the pre-recorded number of data points. and This represents the position coordinates of the i-th data point in A. and This represents the position coordinates of the j-th data point in B. This represents the weight assigned to the i-th data point in A. This represents the weight assigned to the j-th data point in B.

[0062] In the specific implementation process, assume that A and B each have 3 data points, and the location coordinates and assigned weights are as follows:

[0063] Data point 1 for A:

[0064] Data point 2 for A:

[0065] Data point 3 for A:

[0066] Data point 1 for B:

[0067] Data point 2 for B:

[0068] Data point 3 for B:

[0069] The splicing position coordinates obtained based on the preset fusion algorithm are as follows:

[0070]

[0071] In the specific implementation process, the calculated splicing positions are combined to form a trajectory to obtain the current splicing line.

[0072] Optionally, several current data points and pre-recorded data points may include, but are not limited to, location coordinates (latitude and longitude), direction, speed, acceleration, timestamps, route markers, etc. The weights are typically selected based on factors such as the accuracy, reliability, and importance of the data. A preset fusion algorithm is used for fusion processing to ensure a smooth transition at route joints and avoid conflicts.

[0073] Furthermore, a line intelligent algorithm is used to optimize the current spliced ​​line and obtain the line optimization results; a simulation environment is constructed to verify the current spliced ​​line and obtain the verification results; based on the line optimization results and the verification results, the current optimized route is obtained.

[0074] In the specific implementation process, during the fusion of pre-recorded routes and currently recorded routes, a certain degree of route optimization is required to improve driving efficiency and safety. During the splicing process, if a conflict is found between the two routes (such as path intersections or repetitions), conflict detection needs to be performed and corresponding solutions need to be taken. For example, subsequent routes can be replanned to avoid conflict areas, or a prompt can be issued to the user and a request can be made to relearn the currently recorded route. Pre-set intelligent algorithms such as machine learning can also be used to optimize the current spliced ​​route, removing redundant parts to improve the efficiency and accuracy of the path. At the same time, the memory parking route learning can be continuously learned and optimized based on the user's actual usage.

[0075] In practice, conflict detection is usually based on the following situations: spatial intersection, specifically when two routes intersect or overlap in a certain area and there are safety hazards at the intersection; traffic flow conflict, specifically when the traffic flow of two routes is too large at the intersection or adjacent areas, leading to traffic congestion or increased risk of accidents; functional conflict, specifically when the functional positioning of two routes conflicts with each other, such as one being an expressway and the other being an internal road in a residential area, causing mutual interference of traffic flows.

[0076] In the specific implementation process, route planning to avoid conflicts or optimize existing spliced ​​routes is generally based on the following principles or factors: Safety principle, ensuring the route plan meets safety standards and avoids or reduces traffic accidents, such as by installing traffic lights, signs, and markings at intersections; Efficiency principle, optimizing route layout and traffic organization to improve traffic efficiency, such as by rationally setting the number of lanes and adjusting signal timing; Economic principle, considering the economic efficiency of route planning while meeting safety and efficiency requirements, such as optimizing route alignment to reduce construction and maintenance costs; Environmental principle, focusing on the environmental impact of route planning and reducing noise and air pollution, such as by using low-noise pavement materials and installing green belts.

[0077] Furthermore, a non-hardware recording database is set up and associated with other databases through software logic; after the background automatically learns and confirms that there are no errors, the current optimized route is stored in the non-hardware recording database as a pre-recorded route for subsequent memory parking.

[0078] In the specific implementation process, by setting up a non-hardware record database, which is distinguished from other data storage (such as mileage recorded by hardware), an independent storage and processing mechanism is adopted to avoid interference from other data to the background learning data. Furthermore, the data is associated through software logic rather than physical mixing, so even if the route is expanded or modified, the original mileage data recorded by hardware will not be affected.

[0079] In the actual implementation process, once the background path learning is confirmed to be correct, the new route data (i.e. the current optimized route) is synchronously stored for subsequent use. This process is controllable and will not affect the data recorded by the hardware.

[0080] Furthermore, when the vehicle meets the conditions for front-end path learning, front-end path learning is initiated and memory parking is performed based on the front-end path learning.

[0081] In the specific implementation process, the operation of enabling front-end path learning is set up, allowing users to directly control and enable the memory parking function, realizing user interaction and improving the user experience.

[0082] Furthermore, a priority setting is implemented so that when front-end path learning is enabled, back-end path learning is automatically paused, and only one of front-end path learning and back-end path learning can be active at any given time.

[0083] In the specific implementation process, by setting priorities and ensuring the principle of mutual exclusion of states, only one of the front-end path learning and the back-end path learning is carried out at the same time, avoiding the confusion caused by repeated learning records when they are carried out at the same time, and ensuring the accuracy and independence of the data.

[0084] The above solution provides a method to improve the efficiency of memory parking. It controls the activation of background path learning by pre-setting background path learning conditions, making the currently recorded route data more accurate. It employs an independent data storage and processing mechanism to distinguish route data from hardware-recorded data to avoid data interference and ensure data accuracy. It introduces an intelligent route splicing and fusion method to improve the accuracy and efficiency of route splicing and fusion. Furthermore, it sets up front-end and back-end path learning, enabling manual and automatic modes, thus enhancing the user experience of memory parking and improving the overall performance of the memory parking system.

[0085] Example 2:

[0086] This embodiment provides a system for improving memory parking efficiency, such as Figure 2 As shown, the method for improving memory parking efficiency as provided in Embodiment 1 includes:

[0087] The sensor detection module detects the current vehicle speed and current gear data;

[0088] The vehicle intelligent driving controller analyzes and processes current vehicle speed data and current gear data to obtain current vehicle speed information and current gear information, controls the activation of front-end path learning and back-end path learning, and is responsible for the data fusion, optimization and storage of parking routes;

[0089] The user interface displays current vehicle speed and gear information, and enables user interaction; the vehicle bus communication system is used to enable information transmission and sharing between various parts of the system.

[0090] Data storage module: Set up a non-hardware recording database to store the memory parking line data, which is different from the hardware recording data;

[0091] The memory parking module completes memory parking based on the on-board intelligent driving controller.

[0092] In specific implementation, the sensing and detection module includes, but is not limited to: radar sensors, used to measure the relative speed between the vehicle and the surrounding environment; wheel speed sensors, used to be installed on the wheels to measure the wheel rotation speed to calculate the vehicle speed; vision cameras, used for image processing and recognition to assist in judging the vehicle speed and the surrounding environment; and gear position sensors, used to be installed on the transmission to detect the current gear position.

[0093] In practice, the mileage recorded by the hardware is data directly collected by the front-end sensor module and stored in the vehicle ECU or related hardware. The data storage module is responsible for processing additional information beyond this data, such as automatically expanding the route through software algorithms and recording the current optimized route.

[0094] In the specific implementation process, the system adopts a layered design, and the memory parking module includes a front-end path learning layer, a back-end path learning layer, and a user interaction layer.

[0095] This embodiment provides a system for improving the efficiency of memory parking. It implements a hierarchical system design and an independent storage and processing mechanism, optimizes background learning conditions, and introduces intelligent route splicing and fusion methods. This improves the efficiency of memory parking route splicing, making the parking process smoother. It avoids mutual interference between background learning and hardware recording of mileage, ensuring the accuracy and independence of data, and improving the accuracy and smoothness of the route. At the same time, it sets up a user interface to improve the overall performance and user experience of the memory parking system.

[0096] Example 3:

[0097] This embodiment provides a specific system flow for improving memory parking efficiency, executing the method and system for improving memory parking efficiency provided in Embodiments 1 and 2, as illustrated in the flowchart below. Figure 3 As shown, the specific steps include:

[0098] S1: Initially, the system is in standby mode;

[0099] S2: The sensor collects data to obtain the current vehicle speed and current gear data;

[0100] S3: Receive sensor data, process and analyze the data, and obtain current vehicle speed and current gear information;

[0101] S4: Determine whether the current vehicle speed information and current gear information meet the preset background path learning conditions; if not, return to step S1; if they meet, proceed to step S5.

[0102] S5: Enable background path learning and retrieve the currently recorded route;

[0103] S6: Obtain the pre-recorded route, perform route fusion processing based on the currently recorded route and the pre-recorded route, and obtain the current optimized route;

[0104] S7: Data synchronization and update;

[0105] S8: End.

[0106] Furthermore, in step S2, the sensor collects data to obtain current vehicle speed data and current gear data, and also collects data on the vehicle's surrounding environment and vehicle status; specifically, radar sensors, wheel speed sensors, vision cameras, gear sensors, etc. can be used.

[0107] Furthermore, in step S3, the data processing and analysis specifically involves: receiving sensor data and calculating the current vehicle speed information and current gear information through a preset logic algorithm.

[0108] Optionally, in step S4, determining whether the preset background path learning conditions are met specifically means that if the current vehicle speed is less than or equal to 30km / h and the current gear is D (forward gear), the preset background path learning conditions are met; otherwise, they are not met.

[0109] Further, in step S6, the line fusion process specifically involves: firstly, matching the identifier points based on the currently recorded line and the pre-recorded line to determine the splicing position; then, based on the currently recorded line, the pre-recorded line, and the splicing position, performing fusion splicing processing using a preset fusion algorithm to obtain the current spliced ​​line; finally, performing optimization processing based on the current spliced ​​line and the line intelligent algorithm, and performing conflict detection, resolution, and simulation verification to obtain the current optimized line.

[0110] Furthermore, in step S7, the data synchronization update specifically involves: setting up a non-hardware recording database, associating it with other databases through software logic, and storing the current optimized route in the non-hardware recording database after the background automatic learning is completed and confirmed to be correct, as a pre-recorded route for subsequent memory parking, stored separately from other data.

[0111] Optionally, after step S8 is completed, the system enters the next cycle or standby state.

[0112] The specific implementation of the system for improving memory parking efficiency provided in this embodiment realizes an independent storage and processing mechanism for the memory parking system, optimizes the background learning conditions, and introduces intelligent route splicing and fusion methods, thereby improving the efficiency of memory parking route splicing, making the parking process smoother, avoiding mutual interference between background learning and hardware recording of mileage, ensuring the accuracy and independence of data, improving the accuracy and smoothness of the route, and improving the overall performance of the memory parking system.

[0113] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for improving the efficiency of a memory parking, characterized in that, The method comprises the following steps: When the vehicle meets the background path learning condition, start the background path learning, and obtain a current recorded route; Obtain a pre-recorded route, and fuse the current recorded route and the pre-recorded route to obtain a current optimized route; Based on the current optimized route, perform memory parking, and update the pre-recorded route based on the current optimized route to obtain an updated pre-recorded route for the next memory parking; The current recorded route comprises a plurality of current data points, and the pre-recorded route comprises a plurality of pre-recorded data points; The plurality of current data points comprise a current position coordinate data point, a current direction data point, a current speed data point, a current acceleration data point, a current timestamp data point, and a current route marker data point; The plurality of pre-recorded data points comprise a pre-recorded position coordinate data point, a pre-recorded direction data point, a pre-recorded speed data point, a pre-recorded acceleration data point, a pre-recorded timestamp data point, and a pre-recorded route marker data point.

2. The method of claim 1, wherein, When the vehicle meets the background path learning condition, start the background path learning, and obtain a current recorded route, comprising: Collect current vehicle speed information and current gear information, and obtain a current detection result based on a preset logic algorithm analysis and processing; If the current detection result meets the background path learning condition, start the background path learning to obtain the current recorded route.

3. The method of claim 1, wherein, Obtain a pre-recorded route, and fuse the current recorded route and the pre-recorded route to obtain a current optimized route, comprising: Obtain a pre-recorded route; Based on the current recorded route and the pre-recorded route, perform marker point matching to determine a splicing position; Based on the current recorded route, the pre-recorded route, and the splicing position, perform fusion splicing processing by using a preset fusion algorithm to obtain a current spliced route; Based on the current spliced route and a route intelligent algorithm, perform optimization processing to obtain a current optimized route.

4. The method of claim 3, wherein, Based on the current recorded route and the pre-recorded route, perform marker point matching to determine a splicing position, comprising: Based on the current recorded route and the pre-recorded route, perform data preprocessing, and set a plurality of common marker points according to a preset selection method, wherein the marker points comprise position information; Perform matching analysis on the marker points to identify position information that meets a preset marker condition to determine the splicing position.

5. The method of claim 4, wherein, Based on the current recorded route and the pre-recorded route, perform data preprocessing, and set a plurality of common marker points according to a preset selection method when the splicing position accuracy verification does not meet the preset accuracy requirement.

6. The method of claim 3, wherein, Based on the current recorded route, the pre-recorded route, and the splicing position, perform fusion splicing processing by using a preset fusion algorithm to obtain a current spliced route, comprising: the current recorded route comprises a plurality of current data points, the pre-recorded route comprises a plurality of pre-recorded data points, and the preset fusion algorithm meets the following calculation: wherein (x fused , y fused ) denotes the position coordinates after fusion splicing processing, A denotes a current recording line, B denotes a pre-recording line, n A denotes a current data point number, n B denotes a pre-data point number, and denote position coordinates of an i-th data point of A, and denote position coordinates of a j-th data point of B, denotes an allocation weight of the i-th data point of A, denotes an allocation weight of the j-th data point of B.

7. The method of claim 3, wherein, Based on the current spliced route and a route intelligent algorithm, perform optimization processing to obtain a current optimized route, comprising: The current splicing line is optimized by using a line intelligent algorithm to obtain a line optimization processing result; An analog environment is constructed to verify the current splicing line to obtain a verification result; A current optimized route is obtained based on the line optimization processing result and the verification result.

8. The method of claim 1, wherein, Further comprising: A non-hardware record database is set up to be associated with other databases through software logic; When the background automatic learning is completed and confirmed to be correct, the current optimized route is stored in the non-hardware record database as a pre-recorded route for subsequent memory parking.

9. The method of claim 1, wherein, Further comprising: When the vehicle meets the front-end path learning condition, the front-end path learning is started and memory parking is performed based on the front-end path learning.

10. The method of claim 9, wherein, Further comprising: A priority is set, when the front-end path learning is started, the background path learning is automatically paused, and only one of the front-end path learning and the background path learning is in an active state at the same time.

Images