Advanced reservation garage parking guiding method and system based on smart shopping mall

Abstract

The invention relates to the technical field of parking management, in particular to an advanced reservation garage parking guiding method and system based on a smart mall, and the method comprises the steps: obtaining information; screening residual parking spaces to obtain a distribution map; calculating a parking difficulty value; selecting a target parking space and planning a parking path; operating habits, parking difficulty values and parking time of a driver in the parking process are collected, proficiency degrees are divided according to the operating habits of the driver, and the proficiency degrees are stored in a novice data set and a proficiency data set respectively; training a parking time model; acquiring predicted parking time corresponding to each parking space in the parking lot; and according to the predicted parking time, correcting the parking path and guiding. According to the invention, the waiting time in the parking lot is reduced and the congestion probability is further reduced by reducing the passage retention caused by mismatching of parking spaces or too long time for waiting for the front vehicle of the vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of parking management, and in particular to a pre-booking garage parking guidance method and system based on a smart mall. BACKGROUND

[0002] With the acceleration of urbanization and the continuous growth of the number of cars, as a comprehensive commercial space integrating consumption, leisure and service, the service quality of the supporting parking lot of the smart mall has become the core link to improve user experience and the competitiveness of the mall. The traditional mall parking lot has been difficult to meet the needs of intelligent and refined services, and the upgrading of parking lot guidance technology has become an important issue in the construction of smart malls.

[0003] However, the existing parking lot guidance technology still has many drawbacks: on the one hand, the parking space selection and matching lack pertinence, and the existing system mostly recommends only according to the idle state of the parking space, without combining the specific characteristics of the vehicle to carry out adaptation. On the other hand, the parking difficulty evaluation and scheduling do not consider individual differences, and the existing technology mostly relies on subjective experience (such as distance from the entrance) to evaluate the "parking difficulty", without quantifying the influence of physical characteristics of the parking space (passage width, turning angle for entering the garage, etc.) on the parking operation, resulting in long parking time and causing passage congestion.

[0004] Therefore, the existing technology has defects and needs to be improved. SUMMARY

[0005] The purpose of the present application is to provide a pre-booking garage parking guidance method and system based on a smart mall, to solve the problem of lack of pertinence in parking space selection and matching and lack of consideration of individual differences in parking difficulty evaluation and scheduling in the prior art, resulting in long parking time and causing passage congestion.

[0006] The present application provides a pre-booking garage parking guidance method based on a smart mall, comprising: In response to the existence of a parking reservation, obtaining current vehicle information; According to the current vehicle information, the remaining parking spaces are selected, the image information in the parking lot is obtained, and the distribution map of the candidate parking spaces is obtained by identifying the image information in the parking lot; According to the distribution map and the current vehicle information, the parking difficulty value of each candidate parking space is calculated; Selecting the candidate parking space with the lowest parking difficulty value as the target parking space and planning the parking path; Collecting the driving habits, parking difficulty value and parking time of the driver during parking, dividing the proficiency according to the driving habits of the driver, and storing them in the novice data set and the proficiency data set respectively; According to the novice data set and the proficiency data set, the parking time model is trained respectively; In response to the entrance of a vehicle, the predicted parking time corresponding to each parking space in the parking lot is obtained; According to the predicted parking time, the parking path is corrected and parking guidance is performed.

[0007] As a preferred technical solution of the advanced reservation garage parking guidance method based on the intelligent mall, the current vehicle information includes: vehicle type, vehicle length, vehicle width, maximum steering angle, and whether it needs to be charged.

[0008] As a preferred technical solution of the advanced reservation garage parking guidance method based on the intelligent mall, the calculation of the parking difficulty value of each candidate parking space according to the distribution map and the current vehicle information includes: Obtain the image of each candidate parking space and identify the obstacles in the parking lot; According to the distribution map and the obstacles in the parking lot, the actual channel width of the road on which the car travels between the parking spaces is obtained; According to the vehicle type, the vehicle length, the vehicle width, the actual channel width, and the kinematic model, the garage entry trajectory of the current vehicle is simulated; Select the trajectory with the maximum and minimum radius, respectively calculate the corresponding steering angle, and form the steering angle interval with the maximum and minimum values respectively; Calculate the proportion of the interval in the steering angle interval where the steering angle is greater than the preset steering angle, and record it as the parking difficulty value.

[0009] As a preferred technical solution of the advanced reservation garage parking guidance method based on the intelligent mall, the operation habit of the driver during parking, the parking difficulty value, and the parking time are collected, the proficiency is divided according to the operation habit of the driver, and the novice data set and the proficiency data set are stored respectively, including: For a single parking space, the parking time data, parking habit data, and parking difficulty value of each vehicle from starting to reverse to parking in the garage are collected and stored in the form of data groups; Clean the data groups, filter out data groups with parking time less than a first threshold or greater than a second threshold, and obtain standard data groups; According to the operation habit, each of the standard data groups is divided into a proficiency standard data group and a novice standard data group; The proficiency standard data group is stored in the proficiency data set, and the novice standard data group is stored in the novice data set.

[0010] As a preferred technical solution of the advanced reservation garage parking guidance method based on the intelligent mall, the operation habit is one of the total steering wheel rotation required to complete parking, the maximum steering angle deviation during parking, or the interval time from starting to reverse to first adjusting the steering wheel.

[0011] As a preferred technical scheme of the advanced reservation garage parking guidance method based on the smart mall, the standard data set is divided into a skilled standard data set and a novice standard data set according to the operation habit, including: In response to the total steering wheel rotation number exceeding a preset number threshold, or the maximum steering angle deviation exceeding a preset angle threshold, or the interval time being less than a preset time threshold, the standard data set is divided into a novice standard data set, otherwise, it is divided into a skilled standard data set; The parking time model is trained according to the novice standard data set and the skilled standard data set, respectively.

[0012] As a preferred technical scheme of the advanced reservation garage parking guidance method based on the smart mall, the parking time model is built according to the corresponding relationship between the parking difficulty value, the current vehicle information and the parking time as input data, and if the parking difficulty value and the current vehicle information are input, the predicted parking time is output.

[0013] As a preferred technical scheme of the advanced reservation garage parking guidance method based on the smart mall, the parking time of each parking space in the parking lot is predicted in response to the vehicle entering the entrance, including: An image in the parking lot is acquired, each parking space in the image that is performing parking operation is recognized, and a parking difficulty value thereof is acquired; The driving person's current parking operation characteristics are collected, and the proficiency is evaluated; The corresponding parking time model is selected according to the proficiency evaluation result; The parking difficulty value and the current vehicle information are input into the corresponding parking time model, and the predicted parking time is output.

[0014] As a preferred technical scheme of the advanced reservation garage parking guidance method based on the smart mall, the parking path is corrected according to the predicted parking time, and the vehicle is guided according to the corrected parking path, including: The total time consumption for reaching the target parking space is calculated, the path with the lowest total time consumption is selected as the parking path, and the vehicle is guided; The total time consumption for reaching the target parking space is the sum of the driving time consumption in the channel and the waiting time consumption for driving to each parking space that is being parked.

[0015] The application also provides an advanced reservation garage parking guidance system based on the smart mall, including: A reservation module is configured to acquire current vehicle information in response to the existence of parking reservation; A parking space distribution module is configured to filter the remaining parking spaces according to the current vehicle information, and acquire a distribution map of the selected parking spaces. a difficulty value calculation module configured to calculate a parking difficulty value of each of the candidate parking spaces according to the distribution map and the current vehicle information; a path planning module configured to select a candidate parking space with the lowest parking difficulty value as a target parking space and plan a parking path; a data collection module configured to collect an operation habit of a driver, a parking difficulty value and a parking time during parking, divide proficiency according to the operation habit of the driver, and store into a novice data set and a proficiency data set respectively; a model construction module configured to train a parking time model according to the novice data set and the proficiency data set respectively; a result prediction module configured to acquire a predicted parking time corresponding to each parking space being parked in a parking lot in response to a vehicle entering the parking lot; a path correction module configured to correct the parking path according to the predicted parking time and perform parking guidance.

[0016] Compared with the prior art, the application has the beneficial effects that the application acquires vehicle information according to parking reservation and accurately selects candidate parking spaces, combines a kinematic model to calculate a parking difficulty value to realize reasonable allocation of parking spaces, divides a novice / proficiency data set according to collection of an operation habit of a driver and trains a corresponding parking time model, and finally dynamically corrects a path based on real-time predicted parking time, so as to ensure more accurate parking space matching and more reliable parking time prediction, effectively reduce channel retention caused by mismatched parking spaces or too long waiting time for a preceding vehicle, thereby reducing waiting time in a parking lot and further reducing congestion probability. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 a step flow chart of the intelligent market-based advanced reservation garage parking guidance method of the embodiment of the application; Figure 2 a structure block diagram of the intelligent market-based advanced reservation garage parking guidance system of the embodiment of the application. DETAILED DESCRIPTION

[0018] The features and exemplary embodiments of various aspects of the present application will be described below in detail with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended to explain the application, but not to limit the application. The application can be implemented without some of the specific details described below. The following description of the embodiments is merely intended to provide a better understanding of the application by showing examples of the application.

[0019] It is to be noted that the relative terms such as first and second and the like are used herein only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between such entities or operations. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by an "includes" statement does not exclude the existence of additional elements in the process, method, article, or apparatus that includes the element.

[0020] Referring to Figure 1 The figure is a flowchart of the intelligent market-based advanced reservation garage parking guidance method according to an embodiment of the present application, which includes the following steps: Step S1, in response to the existence of parking reservation, acquiring current vehicle information; Step S2, screening the remaining parking spaces according to the current vehicle information, acquiring image information in the parking lot, and identifying the image information in the parking lot to obtain a distribution map of the candidate parking spaces; Step S3, calculating the parking difficulty value of each candidate parking space according to the distribution map and the current vehicle information; Step S4, selecting the candidate parking space with the lowest parking difficulty value as the target parking space and planning a parking path; Step S5, collecting the driving habit, parking difficulty value, and parking time during parking, dividing the proficiency according to the driving habit of the driver, and storing them into the novice data set and the proficiency data set, respectively; Step S6, training the parking time model according to the novice data set and the proficiency data set, respectively; Step S7, in response to the entry of a vehicle, acquiring the predicted parking time corresponding to each parking space in the parking lot; Step S8, correcting the parking path according to the predicted parking time and guiding parking.

[0021] In detail, the present application accurately screens the parking spaces according to the vehicle information obtained by parking reservation, calculates the parking difficulty value in combination with the kinematic model, thereby reasonably allocating the parking spaces and motor vehicles according to the parking difficulty value, divides the novice / proficiency data set by collecting the driving habit of the driver, trains the corresponding parking time model, and finally dynamically corrects the path based on the real-time predicted parking time, thereby reducing the channel retention caused by the mismatch of parking spaces and the long waiting time for the previous vehicle, reducing the waiting time in the parking lot, and further reducing the congestion probability.

[0022] Further, the current vehicle information includes: vehicle type, vehicle length, vehicle width, maximum steering angle, and whether charging is needed.

[0023] In implementation, the user can initiate parking reservation through the mall APP / mini program, trigger vehicle information collection, and support two acquisition methods: Active input: the user manually fills in the vehicle type, vehicle length, vehicle width, maximum steering angle, and charging demand, etc. Automatic identification: the license plate is scanned through the entrance camera of the parking lot, the vehicle parameter database is associated (such as identifying the vehicle type through the license plate, calling the record data of the vehicle management department), or the real-time parameters are read through the OBD interface or Bluetooth, and the vehicle type, vehicle length, vehicle width, maximum steering angle, and charging demand, etc. are automatically identified.

[0024] Further, the present application acquires the full-dimensional information of the vehicle (vehicle type, size, steering angle, special demand, etc.) through active input and automatic identification, ensures accurate matching of vehicle characteristics (such as new energy matching charging pile, large vehicle avoiding narrow parking space) when parking space is selected, avoids parking failure or secondary adjustment due to information missing, reduces the invalid residence time of the vehicle in the parking space, thereby reducing the waiting time in the parking lot and further reducing the congestion probability.

[0025] Specifically, the parking difficulty value of each candidate parking space is calculated according to the distribution map and the current vehicle information, including: Obtaining the image of each candidate parking space, identifying the obstacles in the parking lot; According to the distribution map and the obstacles in the parking lot, the actual channel width of the road on which the vehicle travels between the parking spaces is obtained; According to the vehicle type, vehicle length, vehicle width, actual channel width, and kinematic model, the warehouse entry trajectory of the current vehicle is simulated; Selecting the trajectory with the maximum and minimum radius, respectively calculating the corresponding steering angle, and respectively taking the maximum and minimum values to form a steering angle interval; The proportion of the interval in the steering angle interval where the steering angle is greater than the preset steering angle is calculated and recorded as the parking difficulty value.

[0026] In implementation, the parking space selection and the distribution map are connected through the real-time data of the parking lot geomagnetic sensor and camera, the remaining parking space list is obtained, the vehicle information (such as filtering the micro parking space with a width less than 2.2m when the SUV width is greater than 1.9m, and reserving the charging pile parking space for new energy vehicles) is used to select the adaptive parking space, and the 3D modeling is used to generate the candidate parking space distribution map (labeling the parking space position, channel width, and adjacent parking space state).

[0027] Parking difficulty value calculation: It can be understood that the trajectory formed during the vehicle reversing process will correspond to a unique steering angle, which can be obtained according to the included angle between the trajectory tangent and the vehicle direction, or through a kinematics model, simulate the parking trajectory of each candidate parking space, calculate the proportion of the steering angle > 30° (i.e. the preset angle in the present application, the selection of the preset angle value is determined according to the actual situation, for example: when the simulation trajectory finds that the distance between the current vehicle and the surrounding environment is less than 10 cm, there may be a risk of scratching in the actual parking process, so the steering angle corresponding to this simulation trajectory is recorded as the preset angle value, or other determination methods that meet the actual situation) in the trajectory, as the parking difficulty value (for example, the steering angle interval of a parking space is 10° to 35°, and the preset steering angle is 30°, so the parking difficulty value is 5 / 30=16.7%).

[0028] It can be understood that in the process of vehicle reversing into the garage, the larger the angle required for vehicle steering, the smaller the space available during reversing, the more difficult the parking, for example, in extreme cases: the vehicle can only be parked in the garage at the farthest distance from the garage with the maximum steering angle, there are only two parking methods, the fault tolerance during parking is extremely small, parking is difficult, calculating the parking difficulty value can dataize the difficulty of parking in the garage, on the one hand, it provides a basis for the establishment of the subsequent parking time model, on the other hand, it provides the possibility for subsequent dynamic adjustment of the path based on the predicted parking time, reduces the waiting time in the parking lot, thereby further reducing the congestion probability.

[0029] In detail, the operation habits of the driver during parking, the parking difficulty value and the parking time are collected, the proficiency is divided according to the operation habits of the driver, and the new driver data set and the skilled data set are stored respectively, including: For a single parking space, the parking time data, parking habit data and parking difficulty value of each vehicle from starting reversing to parking in the garage are collected and stored in the form of data groups; The data groups are cleaned, and the data groups with parking time less than the first threshold value or greater than the second threshold value are screened out to obtain standard data groups; Each standard data group is divided into skilled standard data groups and novice standard data groups according to the operation habits; The skilled standard data groups are stored in the skilled data set, and the novice standard data groups are stored in the novice data set.

[0030] In implementation, the path planning aims at "the lowest parking difficulty value + the nearest distance to the entrance", plans the optimal path from the garage entrance to the target parking space, and synchronously generates path navigation instructions (such as "drive straight for 50 meters, turn left, and the target parking space is the third one on the right side"). Data collection is performed through a camera beside the parking space, a steering wheel sensor, and a vehicle-mounted timer. The first threshold value and the second threshold value are used to filter out data sets with problems, such as short parking time due to parking process violations or excessively long parking time due to novice practice of reversing into the garage. In this case, the data sets are not representative and should be filtered out. The values of the first threshold value and the second threshold value should be selected according to the actual situation. Preferably, the first threshold value is selected as the average parking time of drivers with more than fifteen years of driving experience, and the second threshold value is selected as the average parking time of drivers with less than one year of driving experience.

[0031] Further, the present application cleans the standard data sets by collecting data in the form of data sets, and divides the standard data sets according to operating habits, thereby providing a basis for subsequent training of parking time models based on skilled standard data sets and novice standard data sets, and predicting parking time during the reversing process of the driver, correcting the parking path according to the prediction result, reducing the waiting time in the parking lot, and further reducing the congestion probability.

[0032] Further, the operating habit is any one of the following habits: The total number of steering wheel rotations required to complete parking is recorded by a steering wheel rotation sensor. The maximum steering angle deviation during parking, i.e., the difference between the actual steering angle and the optimal trajectory steering angle. The interval time from starting reversing to first adjusting the steering wheel, i.e., the time from starting reversing to first turning the steering wheel.

[0033] In detail, dividing the standard data sets into skilled standard data sets and novice standard data sets according to the operating habit includes: In response to the total number of steering wheel rotations exceeding a preset number of rotations threshold, or the maximum steering angle deviation exceeding a preset angle threshold, or the interval time being less than a preset time threshold, the standard data set is divided into a novice standard data set, otherwise it is divided into a skilled standard data set. The parking time model is trained based on the novice standard data set and the skilled standard data set, respectively.

[0034] In the implementation, the preset number of laps threshold, the preset angle threshold and the preset time threshold are valued to divide the proficiency of the driver, and the values meet the actual situation and are reasonable. Preferably, the values of the preset number of laps threshold, the preset angle threshold and the preset time threshold in the embodiment of the application are determined by collecting the average values of drivers with a driving age of more than fifteen years, and the second threshold selects the average values of drivers with a driving age of less than one year.

[0035] Specifically, by dividing the proficiency of the driver, the corresponding relationship model is trained based on the proficiency, which can reduce the influence of the proficiency of the driver on the parking time in the parking process, make the two types of models be able to be fitted to the parking rules of the corresponding population (new driver high difficulty long time consumption, skilled low difficulty short time consumption), improve the parking time prediction accuracy, reduce the improper scheduling and vehicle waiting caused by the fuzzy classification, thereby reducing the waiting time in the parking lot and further reducing the congestion probability.

[0036] Further, based on the parking time model, the corresponding relationship between the parking difficulty value, the current vehicle information and the parking time is taken as input data, and if the parking difficulty value and the current vehicle information are input, the predicted parking time is output.

[0037] In the implementation, the random forest regression algorithm is used for training the parking time model, the novice / skilled data set is taken as a sample, the input features are the parking difficulty value and the vehicle information (vehicle length, steering angle), and the output is the parking time. The model is optimized through cross-validation (for example, the novice model focuses on fitting the “high difficulty parking space-long time consumption” correlation, and the skilled model focuses on the “low difficulty parking space-short time consumption” rule).

[0038] Further, in response to the entrance having a vehicle entering, the parking time of each parking space in the parking lot is predicted, including: An image in the parking lot is acquired, each parking space in the image in which a parking operation is being performed is recognized, and the parking difficulty value thereof is acquired; The driving person currently performing the parking operation feature is collected to evaluate the proficiency; The corresponding parking time model is selected according to the proficiency evaluation result; The parking difficulty value and the current vehicle information are input into the corresponding parking time model, and the predicted parking time is output.

[0039] Further, the proficiency is evaluated by collecting the driving person operation feature in real time, the corresponding parking time model is called to predict the time consumption of the parking space, the prediction result is dynamically updated to match the actual parking progress, so that the system can perceive the potential congestion point (such as a parking space with a predicted time consumption that is too long) in the channel in advance, provide timely reference for subsequent vehicle path planning, reduce the channel congestion caused by prediction lag, thereby reducing the waiting time in the parking lot and further reducing the congestion probability.

[0040] Further, according to the predicted parking time, the parking path is corrected, and the vehicle is guided according to the corrected parking path, comprising: calculating the total time consumption of reaching the target parking space, selecting the path with the lowest total time consumption as the parking path, and guiding the vehicle; The total time consumption of reaching the target parking space is the sum of the driving time consumption in the channel and the waiting time consumption of driving to each parking space being parked.

[0041] Further, the total time consumption is calculated by combining the channel driving time consumption and the predicted parking time of the parking space being parked, the path with the lowest total time consumption is preferentially selected and dynamically corrected, which can avoid channel congestion caused by long parking time in a certain area, reduce invalid waiting and secondary detour of vehicles in congested road sections, improve the turnover efficiency of vehicle flow in the parking lot, thereby reducing the waiting time in the parking lot and further reducing the congestion probability.

[0042] Please refer to Figure 2 The structure block diagram of the super-advance reservation garage parking guidance system based on intelligent market of the embodiment of the application is shown, which comprises: The reservation module acquires the current vehicle information in response to the existence of parking reservation; The parking space distribution module is used for screening the remaining parking spaces according to the current vehicle information, and acquiring the distribution diagram of the selected parking spaces; The difficulty value calculation module is used for calculating the parking difficulty value of each selected parking space according to the distribution diagram and the current vehicle information; The path planning module selects the selected parking space with the lowest parking difficulty value as the target parking space and plans the parking path; The data acquisition module is used for acquiring the driving habit, parking difficulty value and parking time during parking, dividing the proficiency according to the driving habit of the driver, and storing them into the novice data set and the proficiency data set respectively; The model construction module is used for training the parking time model according to the novice data set and the proficiency data set respectively; The result prediction module acquires the predicted parking time corresponding to each parking space being parked in the parking lot in response to the vehicle entering the entrance; The path correction module is used for correcting the parking path according to the predicted parking time and guiding the parking.

[0043] Obviously, the above embodiments of the present application are merely exemplary but not intended to limit the embodiments of the present application. Based on the above description, any modification, equivalent replacement and improvement etc. within the spirit and principle of the present application should be included in the protection scope of the claims of the present application.

Claims

1. The advance reservation garage parking guidance method based on smart shopping mall is characterized by: include: In response to the existence of a parking reservation, obtaining current vehicle information; Filter the remaining parking spaces according to the current vehicle information, obtain image information in the parking lot, and identify the image information in the parking lot to obtain a distribution map of the parking spaces to be selected; Calculating the parking difficulty value of each of the to-be-selected parking spaces according to the distribution map and the current vehicle information; Select the parking space with the lowest parking difficulty value as the target parking space and plan the parking path; The driver's operating habits, parking difficulty, and parking time during parking are collected, and the proficiency level is divided according to the driver's operating habits. The data are stored in a novice dataset and a skilled dataset respectively. Training parking time models according to the novice dataset and the experienced dataset respectively; In response to a car entering the entrance, obtaining a predicted parking time corresponding to each parking space in the parking lot; The parking route is corrected based on the predicted parking time and parking guidance is performed.

2. The advance reservation garage parking guidance method based on smart shopping mall according to claim 1 is characterized in that: The current vehicle information includes: vehicle model, vehicle length, vehicle width, maximum steering angle, and whether charging is required.

3. The advance reservation garage parking guidance method based on smart shopping mall according to claim 1 is characterized in that: The calculating the parking difficulty value of each of the to-be-selected parking spaces according to the distribution map and the current vehicle information includes: Obtain images of each parking space to be selected and identify obstacles in the parking lot; Obtaining the actual passage width of the road between the parking spaces according to the distribution map and obstacles in the parking lot; Simulating the entry trajectory of the current vehicle based on the vehicle model, length, width, actual channel width, and kinematic model; Select the trajectory with the largest and smallest radius, calculate the corresponding steering angles, and use them as the maximum and minimum values ​​to form the steering angle range; The proportion of the steering angle interval in which the steering angle is greater than the preset steering angle in the steering angle interval is calculated and recorded as the parking difficulty value.

4. The advance reservation garage parking guidance method based on smart shopping mall according to claim 3 is characterized in that: The driver's operating habits, parking difficulty value and parking time during the parking process are collected, and the proficiency level is divided according to the driver's operating habits, and stored in a novice data set and a skilled data set respectively, including: For each parking space, the parking time data, parking habit data, and parking difficulty value of each vehicle from the start of reversing to parking in the garage are collected and stored in the form of data groups; Cleaning the data group, filtering out data groups with parking time less than a first-level threshold or greater than a second-level threshold, and obtaining a standard data group; Dividing each of the standard data groups into an experienced standard data group and a novice standard data group according to the operating habits; The skilled standard data set is stored in the skilled data set, and the novice standard data set is stored in the novice data set.

5. The advance reservation garage parking guidance method based on smart shopping mall according to claim 4 is characterized in that: The operating habit is one of a total number of steering wheel rotations required to complete parking, a maximum steering angle deviation during parking, or an interval from starting to reverse to first adjusting the steering wheel.

6. The advance reservation garage parking guidance method based on smart shopping mall according to claim 5 is characterized in that: The dividing of the standard data groups into skilled standard data groups and novice standard data groups according to the operating habits comprises: In response to the total number of steering wheel rotations exceeding a preset number threshold, or the maximum steering angle deviation exceeding a preset angle threshold, or the interval time being less than a preset time threshold, classifying the standard data group into a novice standard data group, otherwise classifying the standard data group into an experienced standard data group; The parking time model is trained according to the novice standard data set and the skilled standard data set respectively.

7. The advance reservation garage parking guidance method based on smart shopping mall according to claim 6 is characterized in that: Based on the parking time model, the corresponding relationship between the parking difficulty value, current vehicle information and parking time is used as input data. If the parking difficulty value and current vehicle information are input, the predicted parking time is output.

8. The advance reservation garage parking guidance method based on smart shopping mall according to claim 6 is characterized in that: The method of predicting the parking time corresponding to each parking space in the parking lot in response to a car entering the entrance includes: Obtain an image of the parking lot, identify each parking space in the image where a parking operation is being performed, and obtain its parking difficulty value; Collect parking operation characteristics of the driver currently parking and conduct proficiency assessment; Select the corresponding parking time model based on the proficiency assessment results; The parking difficulty value and current vehicle information are input into a corresponding parking time model, and a predicted parking time is output.

9. The advance reservation garage parking guidance method based on smart shopping mall according to claim 1 is characterized in that: The method of correcting the parking path according to the predicted parking time and guiding the vehicle according to the corrected parking path includes: Calculate the total time taken to reach the target parking space, select the path with the lowest total time as the parking path, and guide the vehicle; The total time taken to reach the target parking space is the sum of the time taken to travel in the passage and the waiting time taken to travel to each parking space currently being parked.

10. A smart shopping mall-based advance reservation garage parking guidance system, used to implement the smart shopping mall-based advance reservation garage parking guidance method according to any one of claims 1 to 9, characterized in that: include: a reservation module, in response to the existence of a parking reservation, obtaining current vehicle information; A parking space distribution module is used to screen the remaining parking spaces according to the current vehicle information and obtain a distribution map of the parking spaces to be selected; A difficulty value calculation module is used to calculate the parking difficulty value of each of the to-be-selected parking spaces based on the distribution map and the current vehicle information; The path planning module selects the parking space with the lowest parking difficulty value as the target parking space and plans the parking path; The data collection module is used to collect the driver's operating habits, parking difficulty value and parking time during the parking process, classify the proficiency level according to the driver's operating habits, and store them in novice data sets and skilled data sets respectively; A model building module, configured to train parking time models based on the novice dataset and the experienced dataset respectively; A result prediction module, in response to a car entering the entrance, obtains the predicted parking time corresponding to each parking space in the parking lot; The path correction module is used to correct the parking path and provide parking guidance according to the predicted parking time.

Images