Assisted parking method, server, vehicle, and assisted parking system

Abstract

The application discloses an assisted parking method, a server, a vehicle and an assisted parking system, relates to the technical field of driving assistance, and is applied to a server in an assisted parking system. When it is detected that a target user has a parking demand, the server acquires parking guidance information corresponding to a current parking lot where the target user is located, and the parking guidance information is generated in advance according to parking process data corresponding to the current parking lot. The parking guidance information comprises a parking lot target score, and the parking lot target score represents the parking difficulty of the current parking lot. The parking guidance information is sent to a vehicle corresponding to the target user for display, so as to assist the target user in parking in the current parking lot. In the application, the server quantitatively evaluates the parking difficulty of the parking lot in advance according to the parking process data of the parking lot, and sends the parking guidance information when it is detected that the user has a parking demand, so as to reduce the real-time decision pressure of the user and improve the parking experience of the user.

Description

Technical Field

[0001] This application relates to the field of driver assistance technology, and in particular to a parking assistance method, server, vehicle, and parking assistance system. Background Technology

[0002] With the development of intelligent driving technology, more and more vehicles on the market are equipped with assisted parking systems to optimize the user's parking experience.

[0003] However, existing assisted parking systems largely rely on onboard sensors to perceive the parking environment in real time. Therefore, users, especially novice drivers, face significant real-time decision-making pressure when parking, and the parking experience needs further improvement. Summary of the Invention

[0004] The main purpose of this application is to provide an assisted parking method, server, vehicle, and assisted parking system, aiming to solve the technical problem of high pressure on users to make real-time parking decisions.

[0005] To achieve the above objectives, this application proposes an assisted parking method, applied to a server in an assisted parking system, the method comprising: When a parking need is detected in a target user, parking guidance information corresponding to the current parking lot where the target user is located is obtained. The parking guidance information is pre-generated based on the parking process data corresponding to the current parking lot. The parking guidance information includes a parking lot target score, which represents the difficulty of parking in the current parking lot. The parking guidance information is sent to the vehicle corresponding to the target user and displayed to assist the target user in parking the current parking lot.

[0006] Furthermore, to achieve the above objectives, this application also proposes an assisted parking method for use in a vehicle within an assisted parking system, the method comprising: When a parking need is detected in a target user, parking guidance information of the current parking lot where the target user is located is obtained from the server and displayed. The parking guidance information is used to assist the target user in parking the current parking lot. The parking guidance information is pre-generated by the server based on the parking process data corresponding to the current parking lot; The parking guidance information includes a parking lot target score, which represents the difficulty of parking in the current parking lot.

[0007] In addition, to achieve the above objectives, this application also proposes a server for implementing the assisted parking method mentioned above.

[0008] In addition, to achieve the above objectives, this application also proposes a vehicle for implementing the assisted parking method mentioned above.

[0009] In addition, to achieve the above objectives, this application also proposes an assisted parking system, including the server mentioned above and the vehicle mentioned above.

[0010] This application discloses an assisted parking method, server, vehicle, and assisted parking system. This application relates to the field of driver assistance technology. The method is applied to a server in an assisted parking system. When the server detects that a target user has a parking need, it obtains parking guidance information corresponding to the current parking lot where the target user is located. The parking guidance information is pre-generated based on parking process data corresponding to the current parking lot. The parking guidance information includes a parking lot target score, which characterizes the difficulty of parking in the current parking lot. The parking guidance information is then sent to the vehicle corresponding to the target user for display, to assist the target user in parking in the current parking lot.

[0011] In this application, the server pre-assesses the parking difficulty based on the historical parking process data stored in the parking lot. Then, when it detects that a user has a parking need, it issues parking guidance information to provide the target user with forward-looking parking decision support, reducing the real-time decision-making pressure on novice drivers and improving the user's parking experience. Attached Figure Description

[0012] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a first flowchart illustrating the first embodiment of the assisted parking method of this application; Figure 2 This is a second flowchart illustrating the first embodiment of the assisted parking method of this application; Figure 3 This is a flowchart illustrating the second embodiment of the assisted parking method of this application; Figure 4 This is a system architecture diagram of the assisted parking method of this application; Figure 5 This is a flowchart illustrating the data interaction process of the assisted parking method described in this application.

[0015] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application. Furthermore, all actions involving the acquisition of signals, information, or data in this application are performed in accordance with the relevant data protection regulations of the country where the application is located and with authorization from the owner of the corresponding device.

[0017] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0018] The main solution of this application is that in the server of the assisted parking system, when the server detects that a target user has a parking need, it obtains the parking guidance information corresponding to the current parking lot where the target user is located. The parking guidance information is pre-generated based on the parking process data of the current parking lot. The parking guidance information includes a parking lot target score, which represents the difficulty of parking in the current parking lot. The parking guidance information is then sent to the vehicle corresponding to the target user for display to assist the target user in parking in the current parking lot.

[0019] Currently, existing assisted parking systems can only rely on passive responses during the parking process. Due to the lack of systematic analysis of historical parking data in parking lots, they cannot provide targeted suggestions in advance. As a result, users, especially novice drivers, cannot predict the parking difficulty, common fault scenarios, and potential risks of the target parking lot. Therefore, there is a problem that novice drivers face great pressure in real-time decision-making, and the user experience needs to be further improved.

[0020] To address this issue, this application proposes a vehicle-cloud collaborative architecture. A cloud server aggregates massive amounts of vehicle parking process data, quantifies the difficulty of parking lots, and proactively pushes parking guidance information when a user has a parking need before or upon entering the parking lot, thus achieving a proactive parking assistance method. In this application, the server can pre-assess the parking lot's parking difficulty based on historically stored parking process data. Then, upon detecting a user's parking need, it sends parking guidance information to provide proactive parking decision support, reducing the real-time decision-making pressure on novice drivers and improving the user's parking experience.

[0021] It should be noted that the executing entity in this embodiment can be either the assisted parking system or a server within the assisted parking system; this embodiment does not specifically limit it in this way. The following description uses a server as the executing entity to illustrate this embodiment and the subsequent embodiments.

[0022] Based on this, the embodiments of this application provide an assisted parking method, referring to... Figure 1 , Figure 1 This is a first flowchart illustrating the first embodiment of the assisted parking method of this application.

[0023] In this embodiment, the assisted parking method is applied to the server in the assisted parking system, and the method includes steps S10~S20: Step S10: When a target user is detected to have a parking need, parking guidance information corresponding to the current parking lot where the target user is located is obtained. The parking guidance information is pre-generated based on the parking process data corresponding to the current parking lot. The parking guidance information includes a parking lot target score, which represents the difficulty of parking in the current parking lot. It is easy to understand that the aforementioned server can be a cloud-based big data service platform connected to the target user's terminal device or vehicle, and the target user can be a driver or passenger driving a vehicle equipped with an assisted parking system. In this case, the vehicle can have assisted parking functions, vehicle-to-everything (V2X) communication, GNSS (Global Navigation Satellite System) positioning and map services.

[0024] Understandably, the aforementioned parking demand can be triggered by conditions automatically determined by the server or a parking guidance request automatically generated and sent to the server by the vehicle. For example, conditions determining the existence of parking demand may include: the vehicle's GNSS positioning entering the electronic fence of the current parking lot; the vehicle's navigation destination being the current parking lot and less than 500 meters away; or the target user manually clicking the "Assisted Parking" button on the vehicle's infotainment system. Therefore, when the target user's vehicle enters the vicinity of the target parking lot, the vehicle can upload its location to the server in real time via GNSS+map; and the server can quickly determine that the target user has a parking demand based on the vehicle's location, navigation information, and / or user actions.

[0025] It should be noted that the aforementioned parking guidance information can be a set of information pre-calculated offline by the server based on historical parking data corresponding to the current parking lot. Therefore, in this embodiment, the server can complete the full calculation and update of parking guidance information based on historical multi-vehicle parking data before the target user arrives at the current parking lot, rather than calculating it temporarily after the user arrives, in order to improve the user's parking efficiency. The aforementioned parking process data can be the full data uploaded by the vehicle after each assisted parking. For example, parking condition data may include parking execution result, average driving speed, number of times the vehicle maneuvers into the parking space, parking direction, and parking position deviation.

[0026] At this point, the server can pre-establish a unique index for all parking lots where parking behavior has occurred after the vehicle parking process data is uploaded. The index can be based on: GNSS latitude and longitude + parking lot name + floor number, to ensure that different floors of the same parking lot can be scored independently, guaranteeing the accuracy of parking guidance. That is, the server can receive parking process data uploaded by all connected vehicles on the network in a long-term, real-time manner. Each data entry carries a unique parking lot index, and the server categorizes and stores it according to parking lot dimensions. Therefore, when the server detects parking demand in the current parking lot, it can immediately send the pre-generated parking guidance information (at least including the parking lot target score) to the target vehicle.

[0027] It's important to understand that this parking guidance information includes at least a parking lot target score, and can also be expanded to include parking scenario characteristics, failure scenario information, recommended parking direction, and risk warnings. The parking lot target score can be a quantitative assessment of the difficulty of parking in the parking lot, ranging from 0 to 100 points. A higher score indicates easier assisted parking, a higher success rate, and lower risk; a lower score indicates higher parking difficulty and higher risk.

[0028] In one feasible implementation, refer to Figure 2 , Figure 2 This is a second flowchart illustrating the first embodiment of the assisted parking method of this application. In this embodiment, steps A1 to A4 may be included before step S10: Step A1: Receive parking process data collected when at least one vehicle is parked in the current parking lot; It should be understood that the aforementioned parking process data may include parking lot location information and parking status data uploaded by at least one vehicle each time it executes the parking assistance function in the current parking lot (regardless of whether the parking is successful or not).

[0029] The parking condition data can include parking lot information: GNSS coordinates, floor, parking lot number; and parking condition data: parking execution result (success / failure), average driving speed, number of times the vehicle traversed the parking space, parking direction, and parking position deviation relative to the center of the parking space. The server can then receive one or more parking process data entries uploaded by one or more vehicles in the current parking lot via the MQTT / HTTP protocol. Each data entry can include a timestamp, vehicle identifier, parking lot information, and parking condition data. Based on a fixed update cycle (e.g., 15 minutes / time) or a data volume trigger (every 50 new parking data entries), the server analyzes the uploaded parking process data for each parking lot to generate a dynamic parking lot target score.

[0030] Step A2: Standardize the parking process data to obtain optimized parking data; Understandably, to ensure the quality of input data, the server needs to perform data cleaning and outlier filtering during the above standardization process. For example, the server can remove outliers (such as speed > 1.5 m / s, number of parking maneuvers > 10 times, or position deviation > 1 m) from the collected parking process data; perform Z-Score standardization to eliminate dimensional differences; and use Kalman filtering to smooth continuous data such as speed and deviation, outputting optimized parking data.

[0031] Step A3: Perform data statistics on the optimized parking data based on preset parking dimensions to obtain key parking indicators; the key parking indicators include parking success rate, speed coefficient, number of times the car dribble is used, and position deviation coefficient. It's easy to understand that the preset parking dimensions can be four core dimensions: success rate, speed efficiency, number of parking maneuvers, and parking accuracy. At this point, the server can perform the following statistical analysis on all valid optimized parking data in the current parking lot to obtain key parking metrics: It can calculate the total number of parking attempts (i.e., the number of times all vehicles in the current parking lot have used parking and entered the parking process) and the number of successful attempts (i.e., the number of times all vehicles in the current parking lot have successfully completed parking) to generate a parking success rate; it can calculate the average speed to generate a speed coefficient; it can calculate the average number of parking maneuvers to generate a 1 / (parking maneuver count + 1) calculation; and it can calculate the average position deviation to generate a position deviation coefficient.

[0032] For example, the key parking metrics that the server can obtain from the optimized parking data based on preset parking dimensions may be: parking success rate = number of successful parking attempts / total number of parking attempts; speed coefficient = actual average speed / standard speed (standard speed is set according to parking lot type, such as 0.6m / s for commercial areas and 0.8m / s for residential areas); number of times the vehicle enters and exits the parking space, i.e., the number of times the vehicle enters and exits the parking space; and position deviation coefficient = 1. |Actual deviation|÷0.5 (0.5m is the maximum permissible deviation).

[0033] Step A4: The key parking indicators are weighted and calculated using a preset parking scoring formula to obtain the parking lot target score.

[0034] At this point, the aforementioned preset parking score formula can be: Basic Score = α·Success Rate + β·Speed ​​Coefficient + γ·(1 / (Number of Parking Attempts + 1)) + δ·Position Deviation Coefficient. Where α, β, γ, and δ are the score weight coefficients corresponding to different indicators. The calculated result can then be further mapped to a target score ranging from 0 to 100 points. Finally, the server can bind the latest target score to the parking lot's current parking lot identifier, overwriting the old score and completing the pre-generation process.

[0035] Therefore, in response to the problem that existing assisted parking methods cannot provide a multi-dimensional, objective, and quantifiable assessment of parking difficulty, and can only rely on subjective human judgment or real-time attempts, this embodiment can establish a unified, reproducible, and iterative parking lot scoring model, and comprehensively reflect the real parking difficulty through multi-index weighting.

[0036] It is easy to understand that fixed weights cannot adapt to different parking lot types and environments, easily leading to inaccurate scoring and distorted guidance information. Therefore, in a feasible implementation, step A4 may include steps A41 to A43: Step A41: Obtain the area attribute information and current environmental status information of the current parking lot; Step A42: Dynamically adjust the pre-set initial analysis weights based on the regional attribute information and the current environmental state information to obtain the scene analysis weights corresponding to the current parking lot; Step A43: Based on the scenario analysis weights, the key parking indicators are weighted and calculated using a preset parking scoring formula to obtain the parking target score for the current parking lot.

[0037] It's important to understand that the aforementioned area attribute information can refer to the parking lot type, which may include: commercial area, residential area, office building, underground parking garage, outdoor parking lot, and multi-level parking garage. The current environmental status information can include real-time weather (sunny / rainy / snowy / foggy), light intensity, day / night time, and parking space occupancy rate. In this case, the server can determine the current parking lot type via the map API (Application Programming Interface) or vehicle-uploaded information; and determine the current environmental status information via the weather API, vehicle-uploaded light values, and camera status.

[0038] The initial analysis weights can be the initial scoring weights corresponding to each key parking indicator pre-configured for the parking lot based on the actual situation, while the scenario analysis weights can be the optimized scoring weights corresponding to each key parking indicator after dynamic adjustment based on the area type and environment.

[0039] At this point, the above dynamic adjustment rules can be as follows: if it is a commercial area, increase the weight of the speed coefficient (β can be determined to be 0.4 for parking scene characteristics from 0.2); if it is a residential area, increase the weight of success rate and deviation; if it is a rainy or snowy day, increase the weight of the speed coefficient; if it is night / underground parking garage, increase the weight of the deviation coefficient.

[0040] For example, assuming the initial analysis weights are: α=0.6, β=0.3, γ=0.05, δ=0.05, the server adjusts the weights of the four indicators according to the scenario rules to obtain the scenario analysis weights; 1) Commercial area: α=0.55, β=0.4, γ=0.03, δ=0.02; 2) Rainy and snowy days: α=0.55, β=0.35, γ=0.05, δ=0.05; 3) Underground / Nighttime: α=0.6, β=0.25, γ=0.05, δ=0.1.

[0041] Finally, the server can combine the key parking metrics calculated in real time and the scene analysis weights updated in real time into the preset parking scoring formula to obtain a dynamic score that is more in line with the current scene, which will then be used as the parking target score for the vehicle.

[0042] In this embodiment, the weights can be adaptively adjusted according to different environments and scenarios. Specifically, commercial areas prioritize efficiency, residential areas emphasize safety and accuracy, while in adverse weather / low-light conditions, the accuracy weight is automatically increased to improve reliability. Therefore, in this embodiment, the scoring weights of each key parking indicator can be dynamically adapted to the scenario and environment, improving the accuracy of the scoring.

[0043] It is important to understand that while the target score is intuitive, it cannot determine the specific reasons for parking difficulties based solely on the score, and therefore cannot achieve automatic adaptation of parking strategies. Therefore, in a feasible implementation, the parking guidance information further includes: parking scene features, which characterize the parking scene type of the current parking lot; in this embodiment, steps B1~B2 may also be included before step S10: Step B1: Obtain the preset scene feature label corresponding to the current parking lot; Step B2: Perform cluster analysis based on the preset scene feature labels and the key parking indicators to obtain the parking scene features.

[0044] It should be understood that the aforementioned preset scenario feature labels may include high-efficiency, high-risk, and high-density types. For example, high-efficiency can be a parking lot characterized by a success rate >85%, fast speed, few parking maneuvers, or a time of <25 seconds; high-risk can be a parking lot characterized by a collision / failure rate >5% or an obstacle-to-the-point ratio >20%; high-density can be a parking lot characterized by narrow parking spaces, many parking maneuvers (>5 times), small deviations, but a success rate >80%.

[0045] Therefore, cluster analysis can refer to the server performing K-Means unsupervised clustering on the four standardized parking indicators read: success rate, speed coefficient, number of parking attempts, and deviation coefficient, to obtain the parking lot type for strategy adjustment, i.e., parking scenario characteristics.

[0046] For example, the server can set the number of cluster centers K=3, corresponding to three types of labels, and then perform K-Means clustering, normalize the four indicators, calculate the Euclidean distance, and assign them to the nearest cluster center; then iteratively update the centers until convergence.

[0047] Finally, the server can determine the parking scenario characteristics based on the clustering results. If the cluster center corresponds to a high success rate (success rate > 85%), high speed (average speed > 0.6 m / s), or low time consumption (parking time < 25 s), the parking scenario can be identified as high-efficiency. If the cluster center corresponds to high parking space utilization (number of utilizations > 5), small deviation (position deviation < 0.2 m), and a success rate meeting the standard (success rate > 80%), the parking scenario can be identified as high-density. If the cluster center corresponds to a high failure rate (success rate < 60%) and high collision rate (collision / system failure rate > 5%), the parking scenario can be identified as high-risk. Then, the server can add the scenario characteristics to the parking guidance information and send it along with the parking lot target score to the vehicle to guide the target user to park.

[0048] In this embodiment, the parking scene type can be automatically identified to provide precise auxiliary parking strategy input to the vehicle user. For example, if the parking scene is characterized as high-risk, the user can be guided to use a conservative parking strategy with low-speed driving; if the parking density is high, the user can be guided to use a high-precision conservative parking strategy; and if the efficiency is high, the user can be guided to park quickly. Therefore, this embodiment can make parking guidance information more targeted and accurate by leveraging parking scene characteristics.

[0049] In one feasible implementation, the parking guidance information further includes failure scenario information, which represents the scenario warning information of the current parking lot; step A4 may be followed by steps C1 to C2: Step C1: When the target score of the parking lot is detected to be less than the preset scenario score, the failure labels of the current parking lot are statistically analyzed to obtain the label statistics results. Step C2: Sort the tags according to the frequency of occurrence of each failure tag in the tag statistics results, and generate the failure scenario information based on the failure tag that ranks first in the sort.

[0050] Understandably, the parking process data mentioned above may also include warning labels corresponding to the most frequent parking failure reasons in the current parking lot, i.e., the aforementioned failure labels. In this embodiment, the failure labels may include excessive time consumption, obstacles too close, collision emergency stop, poor perception quality, or system malfunction.

[0051] At this point, the aforementioned preset scenario score can be a pre-set low threshold, which can be set to 50 points by default. A score below this value is considered a difficult / high-risk parking lot. Therefore, if the server can determine whether the target score of the parking lot is lower than the preset scenario score after determining the score, and if a score < 50 points is detected, the failure label statistics process can be further initiated.

[0052] At this point, the server can extract all historical failure tags for the current parking lot from the parking process data. The aforementioned tag statistics can be a frequency count of all historical failure tags for that parking lot, and each failure tag can be sorted from highest to lowest frequency. Correspondingly, the failure scenario information can be scenario warning information generated based on the most frequent failure tag corresponding to the current parking lot.

[0053] For example, if the first item in the ranking is "obstacle too close", the generated failure scenario information is "easily scratched, beware of obstacles"; if the first item in the ranking is "poor perception quality", the generated failure scenario information is "complex environment, high perception difficulty". Then, the server can add the failure scenario information to the parking guidance information, and the vehicle can display the failure scenario information simultaneously when displaying the score to provide risk warnings.

[0054] In this embodiment, to address the problem that existing low-scoring parking lots only know that the problem is "difficult" but do not know where the difficulty lies or what the risks are, this embodiment can provide accurate risk warnings based on the statistical results of failure tags, guiding users to avoid the risks in advance.

[0055] Furthermore, the server can build a cloud-based parking scenario library based on the aforementioned parking lot target scores, parking scenario characteristics, and failure scenario information. This parking scenario library can display basic information, scores, characteristics, and supplementary characteristics (failure scenario characteristics) of each parking lot in the form of a parking scenario map. The assisted parking development team can use the scenario library to specifically increase data collection and performance scenario stress testing for a particular parking lot, and promote targeted optimization. For example, for parking lots where the failure scenario characteristic is a long time consumption, the focus can be on stress testing to improve performance in handling multi-vehicle interference and planning rationality.

[0056] Step S20: The parking guidance information is sent to the vehicle corresponding to the target user for display, so as to assist the target user in parking in the current parking lot.

[0057] Understandably, the specific display method of the parking guidance information after it is sent to the vehicle can be through text / icons / pop-ups on the vehicle's central control screen, prompts on the instrument panel, HUD (Head-Up Display) or TTS (Text-to-Speech) voice broadcast to present the parking guidance information to the user.

[0058] For example, after receiving parking guidance information from the server, the vehicle can display the following information via a pop-up window on the central control screen: parking lot name; parking difficulty rating (0-100); difficulty level (easy / moderate / hard / extremely hard); and simultaneously provide voice prompts, such as: "Current parking lot parking difficulty: 85 points, automatic parking recommended"; or "Current parking lot parking difficulty: 40 points, parking is difficult, please be aware of surrounding obstacles." At this time, the user can choose whether to use assisted parking, change parking spaces, or drive cautiously based on the rating and prompts.

[0059] In summary, existing assisted parking systems rely solely on real-time environmental perception, which fails to assess parking difficulty in advance. This leads to novice drivers experiencing significant real-time decision-making pressure, panic, and parking failures when entering complex parking lots. This embodiment addresses these issues by providing advance awareness, visualization, and early warning of parking difficulty based on parking guidance information. Furthermore, the parking guidance information is pre-calculated in the cloud, eliminating the need for on-vehicle computing power and ensuring a seamless response. The quantitative scoring is intuitive and easy to understand, significantly reducing user psychological stress and operational burden.

[0060] This embodiment provides an assisted parking method. This method is applied to a server in an assisted parking system. When the server detects a parking need from a target user, it obtains parking guidance information corresponding to the current parking lot where the target user is located. The parking guidance information is pre-generated based on parking process data for the current parking lot. The parking guidance information includes a parking lot target score, which represents the difficulty level of parking in the current parking lot. The parking guidance information is then sent to the target user's vehicle for display, assisting the target user in parking the current parking lot.

[0061] In this embodiment, the server pre-assesses the parking difficulty based on the historical parking process data stored in the parking lot. Then, when it detects that a user has a parking need, it issues parking guidance information to provide the target user with forward-looking parking decision support, reducing the real-time decision-making pressure on novice drivers and improving the user's parking experience.

[0062] Based on the first embodiment of this application, in the second embodiment of this application, the same or similar content as the first embodiment described above can be referred to the above description, and will not be repeated hereafter.

[0063] It is readily understood that the executing entity in this embodiment can also be a computing service device with data processing, network communication, and program execution functions in an assisted parking system, such as a tablet computer, personal computer, mobile phone, or vehicle terminal, or an assisted parking device capable of performing the above functions, such as a vehicle. This embodiment does not specifically limit this. The following uses a vehicle as the executing entity as an example to describe this embodiment and the following embodiments.

[0064] Based on this, the embodiments of this application provide an assisted parking method, referring to... Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the assisted parking method of this application.

[0065] In this embodiment, the assisted parking method is applied to the vehicle in the assisted parking system, and the method includes step S11: Step S11: When a target user is detected to have a parking need, the parking guidance information of the current parking lot where the target user is located is obtained from the server and displayed. The parking guidance information is used to assist the target user in parking the current parking lot. The parking guidance information is pre-generated by the server based on the parking process data corresponding to the current parking lot; wherein, the parking guidance information includes a parking lot target score, which represents the difficulty of parking in the current parking lot.

[0066] It is understood that, in this embodiment, the vehicle may be an intelligent vehicle equipped with an assisted parking system, GNSS, and a vehicle-mounted display screen for displaying parking guidance information.

[0067] At this point, the vehicle can use GNSS for real-time positioning and, combined with the navigation map, match the current parking lot where the target user is located. Upon detecting entry into a fence / navigation endpoint / user manually activating the assisted parking function—in other words, determining that the target user has a parking need—the vehicle can upload the parking lot signage, floor, and location to the server via the network to retrieve parking guidance information for the current parking lot. The parking guidance information can be displayed as a pop-up on the central control screen combined with voice TTS (Text-to-Speech) announcements. The specific display process can be referred to in Example 1, and will not be repeated here.

[0068] At this point, users can make parking decisions based on the parking guidance information that is issued and displayed. Therefore, this implementation method can achieve vehicle-side lightweighting, meaning that the vehicle does not need big data processing capabilities to obtain parking guidance information pre-generated based on parking experience across the entire network.

[0069] In one feasible implementation, this embodiment may further include steps D1~D2 after step S11: Step D1: If the target parking score is detected to be less than the preset parking score, obtain the data collection level corresponding to the target parking score. It's important to understand that the aforementioned preset parking score can be a pre-set threshold for triggering high-sensitivity data acquisition, with a default score of 60. The data acquisition level, on the other hand, can be a level of data representing the real-time acquisition sensitivity of the vehicle's data acquisition unit. This data acquisition level can include normal mode, enhanced mode, and ultra-high precision mode. For example, 50–80 points corresponds to normal acquisition; 30–50 points corresponds to enhanced acquisition; and <30 points corresponds to ultra-high precision acquisition.

[0070] For example, after obtaining the parking target score from the server, the vehicle can determine whether the target score is lower than the preset parking score. If the parking target score is less than 50 points from the preset parking score, the data collection level can be adjusted to enter a higher collection level (i.e., enhanced mode or ultra-high precision mode).

[0071] Step D2: Adjust the acquisition sensitivity of the data acquisition unit according to the data acquisition level.

[0072] It's easy to understand that the aforementioned data acquisition sensitivity can be the frame rate and exposure strategy corresponding to the vehicle's surround-view camera; the sampling frequency corresponding to the ultrasonic sensor; the parking space detection frequency; the image point cloud density; and the data upload frequency. Then, the vehicle can execute the assisted parking function according to the adjusted sensitivity configuration of the data acquisition unit, and send the enhanced acquired data back to the cloud for model optimization.

[0073] Therefore, this implementation can automatically enhance perception in difficult parking lots, significantly improving parking success rate and safety.

[0074] This embodiment provides an assisted parking method applied to a vehicle in an assisted parking system. The method includes: upon detecting a parking need from a target user, obtaining and displaying parking guidance information of the current parking lot where the target user is located from a server. The parking guidance information is used to assist the target user in parking at the current parking lot. The parking guidance information is pre-generated by the server based on parking process data corresponding to the current parking lot. The parking guidance information includes a parking lot target score, which characterizes the difficulty of parking at the current parking lot.

[0075] In this embodiment, the vehicle does not require big data support. When a user's parking needs are detected, parking guidance information can be directly obtained from the server to provide the target user with forward-looking parking decision support, reducing the real-time decision-making pressure on novice drivers and improving the user's parking experience.

[0076] For example, to help understand the technical concept or principle of the assisted parking method after combining this embodiment with the above-described Embodiments 1 and 2, please refer to Figure 4 and Figure 5 , Figure 4This is a system architecture diagram of the assisted parking method of this application. Figure 5 The data interaction flowchart for the assisted parking method of this application is as follows: like Figure 4 As shown, the assisted parking system may include two modules: cloud and vehicle.

[0077] The upper-layer cloud is responsible for data processing, scoring mechanisms, and tag generation. During data processing, the cloud performs data cleaning, outlier removal, Kalman filtering, and Z-score normalization on the parking process data uploaded by the vehicle when using the assisted parking function, generating optimized parking data. In the scoring mechanism stage, based on key parking indicators (success rate, speed coefficient, number of parking maneuvers, and position deviation coefficient), and combined with regional attributes and environmental conditions, dynamically adjusts the weights to perform dynamic weighted calculations, generating a target parking lot score. In the tag generation stage, based on key parking indicators, it generates scenario features (high-efficiency / high-risk / high-density), and statistically analyzes high-frequency failure reasons for low-scoring parking lots, generating failure scenario tags.

[0078] The lower-level vehicle-side component is responsible for assisted parking, data uploading, and receiving parking lot information. During the assisted parking phase, the vehicle-side can retrieve pre-generated parking guidance information (including parking lot target score, scene features, and failure labels) from the cloud upon entering the parking lot to perform adaptive assisted parking. It also dynamically adjusts the data collection level and parking strategy based on the parking lot score, completing parking space detection, path planning, and vehicle control. During the data uploading phase, after each use of assisted parking, the vehicle uploads parking status data (success / failure, speed, number of parking maneuvers, deviation, and failure label) to the cloud.

[0079] Furthermore, such as Figure 5 As shown, the entire assisted parking process can include the following four steps: Step 1: When a vehicle equipped with an assisted parking system uses assisted parking, it uploads information to the cloud: Each time a vehicle completes parking / parking using the assisted parking function, the data acquisition unit collects complete parking process data (including parking lot GNSS positioning, floor, parking result, average speed, number of times the vehicle traverses the parking space, position deviation, failure scenario label, etc.), and uploads the data to the cloud server through the vehicle network communication module. This step forms the data source foundation for big data analysis in the cloud, ensuring that the cloud can aggregate parking behavior data from all vehicles on the network, providing support for subsequent scoring calculations.

[0080] Step 2: The cloud platform uses the information uploaded by the vehicles to generate a rating and characteristics for the parking lot based on the scoring mechanism. At this point, after receiving the massive amount of parking data uploaded from the vehicle, the cloud server executes the following complete processing flow: 1) Data standardization processing: outlier removal, filtering, and normalization to generate optimized parking data; 2) Key performance indicator statistics: Success rate, speed coefficient, number of parking maneuvers, and position deviation coefficient are statistically analyzed by parking lot dimension; 3) Dynamic weighted scoring: The weights are dynamically adjusted based on the parking area attributes and environmental conditions to calculate and generate the target score for the parking lot; 4) Scene feature clustering: Parking scene features (high efficiency / high risk / high density) are generated through K-Means clustering. 5) Failure Tag Generation: Analyze the high-frequency failure reasons for low-scoring parking lots and generate failure scenario tags; Ultimately, the cloud integrates scores, features, and tags into parking guidance information, pre-generated and stored, and completes offline calculations, eliminating the need for temporary calculations upon vehicle arrival.

[0081] Step 3: The vehicle enters a parking lot, and the parking lot's rating and feature information are obtained from the cloud based on GNSS positioning. The vehicle can obtain its location in real time through the GNSS positioning module. Combined with map electronic fences and navigation information, it can determine that the vehicle has entered the target parking lot and has a parking need. Then, the vehicle sends information such as parking lot signage and floor to the cloud through the communication module to request parking guidance information. The cloud immediately sends the pre-generated rating, feature, and tag information of the parking lot to the vehicle, and the vehicle completes the information reception and parsing.

[0082] Step 4: Based on the obtained parking information, the vehicle provides assisted parking recommendations, usage instructions, and user-friendly prompts. After receiving parking guidance information from the cloud, the vehicle can perform two core operations: 1) Information display and prompts: The HMI (Human-Machine Interface) module (central control screen, instrument panel, voice) displays the parking lot target score, difficulty level, scene characteristics, and failure scene labels to the user, such as "The current parking lot parking difficulty is 40 points, high risk, easy to scratch obstacles", to provide users with decision-making reference; 2) Adaptive Strategy Adjustment: The data collection level is dynamically adjusted based on the parking lot rating. For parking lots with low ratings, the camera frame rate and ultrasonic sampling rate are automatically increased, the parking speed is reduced, and the safety distance is increased to implement enhanced assisted parking, thereby improving the parking success rate and safety.

[0083] In summary, this application collects usage data from vehicles equipped with parking assistance each time, and pre-generates a rating and characteristics for each parking lot in the cloud based on a scoring mechanism. When a user drives a vehicle into a parking lot, the parking assistance system on the vehicle can obtain information such as the parking lot's rating and characteristics from the cloud through GNSS positioning and maps, and use this information to provide parking assistance usage recommendations, usage instructions, and user-friendly interactions.

[0084] Therefore, this application can quantify the assisted parking results of parking lots in advance through a multi-dimensional dynamic scoring mechanism and feature analysis, and send this information to the vehicle to realize intelligent recommendations, risk warnings, and dynamic adjustment of parking strategies, thereby improving the user experience of assisted parking; and through the scoring and features of parking lots, data-driven continuous optimization of assisted parking can be achieved.

[0085] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the assisted parking method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0086] To achieve the above objectives, this application also proposes a server that can be used to implement the assisted parking method mentioned in Embodiment 1 above. The collaborative workflow of the assisted parking method can be referred to Embodiment 1 above. Since this server adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be elaborated here.

[0087] To achieve the above objectives, this application also proposes a vehicle that can be used to implement the assisted parking method mentioned in Embodiment 2 above. The collaborative workflow of the assisted parking method can be referred to Embodiment 2 above. Since this vehicle adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be described in detail here.

[0088] To achieve the above objectives, this application also proposes an assisted parking system, which includes a server and a vehicle as mentioned above. The specific execution process of the server and vehicle can be referred to the above embodiments. Since this assisted parking system adopts all the technical solutions of all the above embodiments, it possesses at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be elaborated upon here. The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0089] The above are only some embodiments of this application and do not limit the scope of the solution of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included within the protection scope of this application.

Claims

1. A parking assistance method, characterized in that, The method is applied to a server in an assisted parking system, and the method includes: When a parking need is detected in a target user, parking guidance information corresponding to the current parking lot where the target user is located is obtained. The parking guidance information is pre-generated based on the parking process data corresponding to the current parking lot. The parking guidance information includes a parking lot target score, which represents the difficulty of parking in the current parking lot. The parking guidance information is sent to the vehicle corresponding to the target user and displayed to assist the target user in parking the current parking lot.

2. The assisted parking method as described in claim 1, characterized in that, Before obtaining parking guidance information for the current parking lot where the target user is located when a parking need is detected, the method further includes: The parking process data is standardized to obtain optimized parking data; Based on preset parking dimensions, the optimized parking data is statistically analyzed to obtain key parking indicators; the key parking indicators include parking success rate, speed coefficient, number of times the car dribble is made, and position deviation coefficient. The key parking indicators are weighted and calculated using a preset parking scoring formula to obtain the parking lot target score.

3. The assisted parking method as described in claim 2, characterized in that, The step of obtaining the parking lot target score by weighting the key parking indicators through a preset parking score formula includes: Obtain the area attribute information and current environmental status information of the current parking lot; Based on the regional attribute information and the current environmental state information, the pre-set initial analysis weights are dynamically adjusted to obtain the scene analysis weights corresponding to the current parking lot. Based on the scenario analysis weights, the key parking indicators are weighted and calculated using a preset parking scoring formula to obtain the parking target score for the current parking lot.

4. The assisted parking method as described in claim 2, characterized in that, The parking guidance information also includes: parking scene characteristics. Before obtaining the parking guidance information corresponding to the current parking lot where the target user is located when a parking need is detected, the information further includes: Obtain the preset scene feature tags corresponding to the current parking lot; Cluster analysis is performed based on the preset scene feature labels and the key parking indicators to obtain the parking scene features.

5. The assisted parking method as described in claim 4, characterized in that, The parking guidance information also includes failure scenario information, which represents the scenario warning information of the current parking lot; After obtaining the parking scene features by performing cluster analysis based on the preset scene feature labels and the key parking indicators, the method further includes: When the target score of the parking lot is detected to be less than the preset scenario score, the failure labels of the current parking lot are statistically analyzed to obtain the label statistics results; The tags are sorted according to the frequency of occurrence of each failure tag in the tag statistics results, and the failure scenario information is generated based on the failure tag that ranks first in the sort.

6. A parking assistance method applied to a vehicle in a parking assistance system, the method comprising: When a parking need is detected in a target user, parking guidance information of the current parking lot where the target user is located is obtained from the server and displayed. The parking guidance information is used to assist the target user in parking the current parking lot. The parking guidance information is pre-generated by the server based on the parking process data corresponding to the current parking lot; The parking guidance information includes a parking lot target score, which represents the difficulty of parking in the current parking lot.

7. The assisted parking method as described in claim 6, characterized in that, After detecting that a target user has a parking need, the method of obtaining and displaying parking guidance information of the current parking lot where the target user is located from the server also includes: If the target parking score is detected to be less than the preset parking score, the data collection level corresponding to the target parking score is obtained; Adjust the acquisition sensitivity of the data acquisition unit according to the data acquisition level.

8. A server for implementing the assisted parking method according to any one of claims 1 to 5.

9. A vehicle for implementing the assisted parking method according to any one of claims 6 to 7.

10. A parking assistance system, comprising the server as described in claim 8 and the vehicle as described in claim 9.

Images